From: Mike Buckler Newsgroups: alt.folklore.computers Subject: Does the word "mainframe" still have a meaning? Date: Tue, 20 Jun 2000 21:22:27 +0800 Message-ID: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 16 NNTP-Posting-Host: 203.59.77.120 X-Trace: news.iinet.net.au 961509163 24357 v0yeus@203.59.77.120 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!harbinger.cc.monash.edu.au!news.mel.connect.com.au!news.per.connect.com.au!newsfeed.iinet.net.au!news.iinet.net.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58106 After searching whatis.com, webopedia.com, Google and Altavista, the only definition I could find was "a computer that supports hundreds of simultaneous users". Is that it? (Is that all there is, my friend? Then let's keep dancing. Let's break out the booze and have a ball... but I digress) If so does that mean that any modern box with a multiuser OS is a mainframe? Even IBM appears to be using the phrase "enterprise server" (whatever that means). So what's the problem, does a mainframe have to look like the cooling jacket of a nuclear reactor? Can nothing be sold unless "enterprise" is in the product name? Mike ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 21 Jun 00 12:50:50 -0800 Organization: http://extra.newsguy.com Lines: 49 Message-ID: <2510.207T2542T7706179@sky.bus.com> References: <394F3A96.534E7003@trailing-edge.com> NNTP-Posting-Host: p-309.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!news1 Xref: chonsp.franklin.ch alt.folklore.computers:58280 In article Brian.dot.Inglis@SystematicSw.ab.ca (Brian Inglis) writes: >The one distinguishing characteristic I have noticed on mainframe >systems is the presence of multiple I/O channels to which any I/O >interface could be attached. As compared to the minicomputer CPU >and I/O interfaces on the backplane, or the microcomputer CPU and >interfaces on the motherboard. Perhaps this might be a better >taxonomy. My MF experience was IBM but I have seen comments >implying similar architectures on DEC-10s, Univac 1100s, HIS/GE, >etc. Anyone care to compare/contrast across MF manufacturer >lines? Mainframes tend to have massive I/O capabilities, even if they're not so powerful CPU-wise. This is ideal in a corporate environment, where you're doing things like adding up columns of numbers or printing accounts receivable statements. These are not complicated operations, but boy do you do a lot of them. It usually makes more sense to spend your money on faster peripherals than on a CPU that's just going to be waiting for them anyway. Current desktop computers have CPU power, memory, and disk space that the manager of a data centre 20 years ago could only have dreamed of (although he'd be dreaming about the disk capacity much more than the CPU power). But his printers would run rings around anything you'll see even on today's desktops. As a programmer, though, I like to distinguish mainframes from minis or micros on a conceptual level. Mainframes tend to do record-level I/O - even on terminals - to reduce CPU overhead, which typically was extremely expensive. Mainframes excel at handling fixed-length records and batch processing, again because these tend to make more efficient use of both I/O and CPU resources. Such techniques are well suited to typical mainframe applications, which involve things like printing columns of numbers much more than dealing with, say, free-form text. Interactive processing on a mainframe (CICS, for example) might be a terror to program and cumbersome to use, but it enabled a CPU that could do only a fraction of a MIP to service dozens of terminals. This leads to a noticeably different mindset between mainframe and mini programmers, and it's not just a matter of size: a 360/20 programmer would have the "mainframe" mindset, while someone using a much larger PDP-10 would probably have the "mini" mindset. -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. ###### Message-ID: <394F3A96.534E7003@trailing-edge.com> Date: Tue, 20 Jun 2000 09:34:14 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 20 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader0.news.uu.net 961508055 15100 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!news.maxwell.syr.edu!newsfeed.enteract.com!hermes.visi.com!news-out.visi.com!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader0.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58155 Mike Buckler wrote: > > After searching whatis.com, webopedia.com, Google and Altavista, the only > definition I could find was "a computer that supports hundreds of > simultaneous users". Is that it? (Is that all there is, my friend? Then let's > keep dancing. Let's break out the booze and have a ball... but I digress) > If so does that mean that any modern box with a multiuser OS is a mainframe? A mainframe is, I'm afraid, whatever the users decide to call a mainframe. I've worked in several shops where the end-users all called the box in the computer room "the mainframe", whether the "mainframe" was a real IBM behemoth or just a small workstation. I think the most accurate assesment of "mainframe" is "the computer in the computer room that the end-users aren't allowed to touch." This thread has appeared on alt.folklore.computers several times in the past, I think one of my previous contributions was "You know it's a mainframe if you've ever lost an oscilliscope inside of it." Tim. ###### Message-ID: <394F8FC4.9E46A57F@cwnet.com> Date: Tue, 20 Jun 2000 08:37:40 -0700 From: Dale DePriest X-Mailer: Mozilla 4.73 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <394F3A96.534E7003@trailing-edge.com> <8ins7k$3ai$1@news.igs.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 205.162.108.38 X-Trace: 20 Jun 2000 08:43:50 -0700, 205.162.108.38 Lines: 44 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!feed2.newsfeeds.com!newsfeeds.com!newspeer.cwnet.com!news2.cwnet.com!205.162.108.38 Xref: chonsp.franklin.ch alt.folklore.computers:58107 I think mini's are gone. Sun calls their company Sun microsystems and they are the closest thing to mini's today under this definition. I think the definition of mainframes is still valid however. It may also need to be made by a "mainframe manufacturer" such as IBM and UniSys or Cray. Dale donald tees wrote: > > Tim Shoppa wrote in message <394F3A96.534E7003@trailing-edge.com>... > >Mike Buckler wrote: > >> > >> After searching whatis.com, webopedia.com, Google and Altavista, the only > >> definition I could find was "a computer that supports hundreds of > >> simultaneous users". Is that it? (Is that all there is, my friend? Then > let's > >> keep dancing. Let's break out the booze and have a ball... but I digress) > >> If so does that mean that any modern box with a multiuser OS is a > mainframe? > > > >A mainframe is, I'm afraid, whatever the users decide to call a > >mainframe. I've worked in several shops where the end-users all > >called the box in the computer room "the mainframe", whether the > >"mainframe" was a real IBM behemoth or just a small workstation. > >I think the most accurate assesment of "mainframe" is "the computer > >in the computer room that the end-users aren't allowed to touch." > > > >This thread has appeared on alt.folklore.computers several times in > >the past, I think one of my previous contributions was "You know it's > >a mainframe if you've ever lost an oscilliscope inside of it." > > > >Tim. > > Sounds like the argument over what a "mini-computer" was twenty years back. > I think the most useful definition then was > "over $100,000 is a mainframe, under $100,000" is a mini". Perhaps under > $10,000 is a micro. -- For GPS data see: Joe -- http://joe.mehaffey.com Peter -- http://www.vancouver-webpages.com/peter/ Karen -- http://www.gpsy.com/gpsinfo/ Dale -- http://users.cwnet.com/dalede ###### From: "W Stiefer" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 20 Jun 2000 11:31:01 -0500 Organization: Illuminati Online Lines: 56 Message-ID: <8io6dn$imm$1@hiram.io.com> References: <394F3A96.534E7003@trailing-edge.com> <8ins7k$3ai$1@news.igs.net> <394F8FC4.9E46A57F@cwnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Trace: hiram.io.com 961518840 19158 208.2.105.40 (20 Jun 2000 16:34:00 GMT) X-Complaints-To: abuse@io.com X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!feed1.news.rcn.net!rcn!korova.insync.net!solomon.io.com!hiram.io.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58123 would "server" apply? Dale DePriest wrote in message = news:394F8FC4.9E46A57F@cwnet.com... > I think mini's are gone. Sun calls their company Sun microsystems and > they are the closest thing to mini's today under this definition. I > think the definition of mainframes is still valid however. It may = also > need to be made by a "mainframe manufacturer" such as IBM and UniSys = or > Cray. >=20 > Dale >=20 > donald tees wrote: > >=20 > > Tim Shoppa wrote in message <394F3A96.534E7003@trailing-edge.com>... > > >Mike Buckler wrote: > > >> > > >> After searching whatis.com, webopedia.com, Google and Altavista, = the only > > >> definition I could find was "a computer that supports hundreds of > > >> simultaneous users". Is that it? (Is that all there is, my = friend? Then > > let's > > >> keep dancing. Let's break out the booze and have a ball... but I = digress) > > >> If so does that mean that any modern box with a multiuser OS is a > > mainframe? > > > > > >A mainframe is, I'm afraid, whatever the users decide to call a > > >mainframe. I've worked in several shops where the end-users all > > >called the box in the computer room "the mainframe", whether the > > >"mainframe" was a real IBM behemoth or just a small workstation. > > >I think the most accurate assesment of "mainframe" is "the computer > > >in the computer room that the end-users aren't allowed to touch." > > > > > >This thread has appeared on alt.folklore.computers several times in > > >the past, I think one of my previous contributions was "You know = it's > > >a mainframe if you've ever lost an oscilliscope inside of it." > > > > > >Tim. > >=20 > > Sounds like the argument over what a "mini-computer" was twenty = years back. > > I think the most useful definition then was > > "over $100,000 is a mainframe, under $100,000" is a mini". Perhaps = under > > $10,000 is a micro. >=20 > --=20 > For GPS data see: Joe -- http://joe.mehaffey.com > Peter -- http://www.vancouver-webpages.com/peter/ > Karen -- http://www.gpsy.com/gpsinfo/ > Dale -- http://users.cwnet.com/dalede ###### From: "W Stiefer" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 20 Jun 2000 11:32:54 -0500 Organization: Illuminati Online Lines: 56 Message-ID: <8io6do$imm$2@hiram.io.com> References: <394F3A96.534E7003@trailing-edge.com> <8ins7k$3ai$1@news.igs.net> <394F8FC4.9E46A57F@cwnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Trace: hiram.io.com 961518841 19158 208.2.105.40 (20 Jun 2000 16:34:01 GMT) X-Complaints-To: abuse@io.com X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!portc.blue.aol.com.MISMATCH!portc03.blue.aol.com!nntp2.giganews.com!nntp.giganews.com!solomon.io.com!hiram.io.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58124 How about "servee"? Dale DePriest wrote in message = news:394F8FC4.9E46A57F@cwnet.com... > I think mini's are gone. Sun calls their company Sun microsystems and > they are the closest thing to mini's today under this definition. I > think the definition of mainframes is still valid however. It may = also > need to be made by a "mainframe manufacturer" such as IBM and UniSys = or > Cray. >=20 > Dale >=20 > donald tees wrote: > >=20 > > Tim Shoppa wrote in message <394F3A96.534E7003@trailing-edge.com>... > > >Mike Buckler wrote: > > >> > > >> After searching whatis.com, webopedia.com, Google and Altavista, = the only > > >> definition I could find was "a computer that supports hundreds of > > >> simultaneous users". Is that it? (Is that all there is, my = friend? Then > > let's > > >> keep dancing. Let's break out the booze and have a ball... but I = digress) > > >> If so does that mean that any modern box with a multiuser OS is a > > mainframe? > > > > > >A mainframe is, I'm afraid, whatever the users decide to call a > > >mainframe. I've worked in several shops where the end-users all > > >called the box in the computer room "the mainframe", whether the > > >"mainframe" was a real IBM behemoth or just a small workstation. > > >I think the most accurate assesment of "mainframe" is "the computer > > >in the computer room that the end-users aren't allowed to touch." > > > > > >This thread has appeared on alt.folklore.computers several times in > > >the past, I think one of my previous contributions was "You know = it's > > >a mainframe if you've ever lost an oscilliscope inside of it." > > > > > >Tim. > >=20 > > Sounds like the argument over what a "mini-computer" was twenty = years back. > > I think the most useful definition then was > > "over $100,000 is a mainframe, under $100,000" is a mini". Perhaps = under > > $10,000 is a micro. >=20 > --=20 > For GPS data see: Joe -- http://joe.mehaffey.com > Peter -- http://www.vancouver-webpages.com/peter/ > Karen -- http://www.gpsy.com/gpsinfo/ > Dale -- http://users.cwnet.com/dalede ###### From: David Scheidt Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 20 Jun 2000 23:03:20 GMT Organization: EnterAct Corp. Lines: 18 Message-ID: <8iot7o$163c$1@news.enteract.com> References: <394F3A96.534E7003@trailing-edge.com> <8ins7k$3ai$1@news.igs.net> <394F8FC4.9E46A57F@cwnet.com> NNTP-Posting-Host: shell-1.enteract.com X-Trace: news.enteract.com 961542200 39020 207.229.143.40 (20 Jun 2000 23:03:20 GMT) X-Complaints-To: abuse@enteract.com NNTP-Posting-Date: 20 Jun 2000 23:03:20 GMT User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (FreeBSD/3.3-STABLE (i386)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!news.maxwell.syr.edu!newsfeed.enteract.com!betanews.enteract.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58227 Dale DePriest wrote: : I think mini's are gone. Sun calls their company Sun microsystems and : they are the closest thing to mini's today under this definition. I : think the definition of mainframes is still valid however. It may also : need to be made by a "mainframe manufacturer" such as IBM and UniSys or : Cray. Of course, Sun make a huge range of machines, from things that are pretty micro -- their low-end graphics workstations, say -- to things that I'd have no problem calling mainframes, like an Ultra E10000. They run the same OS, and the same user applications, more or less. Interestingly, this was one of IBMs goals with the 360. -- dscheidt@enteract.com You could look at our current gun laws as a way of ensuring that British armed criminals are really *motivated*. -- Phil Edwards ###### From: Shez Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 21 Jun 2000 07:45:03 +0100 Organization: the Last Stop Cafe Message-ID: References: Reply-To: Shez NNTP-Posting-Host: xerez.demon.co.uk X-NNTP-Posting-Host: xerez.demon.co.uk:193.237.22.174 X-Trace: news.demon.co.uk 961610012 nnrp-11:23828 NO-IDENT xerez.demon.co.uk:193.237.22.174 X-Complaints-To: abuse@demon.net MIME-Version: 1.0 Content-Type: text/plain;charset=us-ascii;format=flowed X-Newsreader: Turnpike Integrated Version 5.01 S Lines: 19 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!dispose.news.demon.net!demon!news.demon.co.uk!demon!xerez.demon.co.uk!news Xref: chonsp.franklin.ch alt.folklore.computers:58238 Mike Buckler of writes: >After searching whatis.com, webopedia.com, Google and Altavista, the only >definition I could find was "a computer that supports hundreds of >simultaneous users". As I understand it, originally a mainframe was just that - a big metal frame for rack-mounting components, in the way that still happens in some areas of electronics. A mainframe computer was a computer whose components were rack mounted in a main frame. -Shez. -- ______________________________________________________ Experience is the worst teacher. It always gives the test first and the instruction afterward. ______________________________________________________ Take a break at the Last Stop Cafe: http://www.xerez.demon.co.uk/ Use PGP: my key is at http://www.xerez.demon.co.uk/pgp/Shez.asc ###### From: "donald tees" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 21 Jun 2000 09:35:19 -0400 Organization: IGS - Information Gateway Services Lines: 32 Message-ID: <8ins7k$3ai$1@news.igs.net> References: <394F3A96.534E7003@trailing-edge.com> NNTP-Posting-Host: ttyc00.kw.igs.net X-Trace: news.igs.net 961508404 3410 216.58.99.128 (20 Jun 2000 13:40:04 GMT) X-Complaints-To: abuse@igs.net NNTP-Posting-Date: 20 Jun 2000 13:40:04 GMT X-Newsreader: Microsoft Outlook Express 4.72.3110.1 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!torn!news.uunet.ca!nntp.igs.net!news.igs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58121 Tim Shoppa wrote in message <394F3A96.534E7003@trailing-edge.com>... >Mike Buckler wrote: >> >> After searching whatis.com, webopedia.com, Google and Altavista, the only >> definition I could find was "a computer that supports hundreds of >> simultaneous users". Is that it? (Is that all there is, my friend? Then let's >> keep dancing. Let's break out the booze and have a ball... but I digress) >> If so does that mean that any modern box with a multiuser OS is a mainframe? > >A mainframe is, I'm afraid, whatever the users decide to call a >mainframe. I've worked in several shops where the end-users all >called the box in the computer room "the mainframe", whether the >"mainframe" was a real IBM behemoth or just a small workstation. >I think the most accurate assesment of "mainframe" is "the computer >in the computer room that the end-users aren't allowed to touch." > >This thread has appeared on alt.folklore.computers several times in >the past, I think one of my previous contributions was "You know it's >a mainframe if you've ever lost an oscilliscope inside of it." > >Tim. Sounds like the argument over what a "mini-computer" was twenty years back. I think the most useful definition then was "over $100,000 is a mainframe, under $100,000" is a mini". Perhaps under $10,000 is a micro. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 21 Jun 2000 08:23:54 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <394F3A96.534E7003@trailing-edge.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.146.116 X-Original-NNTP-Posting-Host: 207.148.146.116 X-Trace: 21 Jun 2000 08:23:55 -0700, 207.148.146.116 Lines: 38 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newspump.monmouth.com!newspeer.monmouth.com!nntp.cadvision.com!news.cadvision.com!207.148.146.116 Xref: chonsp.franklin.ch alt.folklore.computers:58260 On Tue, 20 Jun 2000 09:34:14 -0400, Tim Shoppa wrote: >Mike Buckler wrote: >> >> After searching whatis.com, webopedia.com, Google and Altavista, the only >> definition I could find was "a computer that supports hundreds of >> simultaneous users". Is that it? (Is that all there is, my friend? Then let's >> keep dancing. Let's break out the booze and have a ball... but I digress) >> If so does that mean that any modern box with a multiuser OS is a mainframe? > >A mainframe is, I'm afraid, whatever the users decide to call a >mainframe. I've worked in several shops where the end-users all >called the box in the computer room "the mainframe", whether the >"mainframe" was a real IBM behemoth or just a small workstation. >I think the most accurate assesment of "mainframe" is "the computer >in the computer room that the end-users aren't allowed to touch." > >This thread has appeared on alt.folklore.computers several times in >the past, I think one of my previous contributions was "You know it's >a mainframe if you've ever lost an oscilliscope inside of it." > >Tim. The one distinguishing characteristic I have noticed on mainframe systems is the presence of multiple I/O channels to which any I/O interface could be attached. As compared to the minicomputer CPU and I/O interfaces on the backplane, or the microcomputer CPU and interfaces on the motherboard. Perhaps this might be a better taxonomy. My MF experience was IBM but I have seen comments implying similar architectures on DEC-10s, Univac 1100s, HIS/GE, etc. Anyone care to compare/contrast across MF manufacturer lines? Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: "donald tees" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 21 Jun 2000 12:49:22 -0400 Organization: IGS - Information Gateway Services Lines: 58 Message-ID: <8io7jf$92t$1@news.igs.net> References: <394F3A96.534E7003@trailing-edge.com> <8ins7k$3ai$1@news.igs.net> <394F8FC4.9E46A57F@cwnet.com> <8io6do$imm$2@hiram.io.com> NNTP-Posting-Host: ttyd11.kw.igs.net X-Trace: news.igs.net 961520047 9309 216.58.99.177 (20 Jun 2000 16:54:07 GMT) X-Complaints-To: abuse@igs.net NNTP-Posting-Date: 20 Jun 2000 16:54:07 GMT X-Newsreader: Microsoft Outlook Express 4.72.3110.1 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!prairie.attcanada.net!172.31.25.105!rockie.attcanada.net!newsfeed.attcanada.net!216.58.1.11!nntp.igs.net!news.igs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58114 Maybe we should take a "servey". W Stiefer wrote in message <8io6do$imm$2@hiram.io.com>... How about "servee"? Dale DePriest wrote in message news:394F8FC4.9E46A57F@cwnet.com... > I think mini's are gone. Sun calls their company Sun microsystems and > they are the closest thing to mini's today under this definition. I > think the definition of mainframes is still valid however. It may also > need to be made by a "mainframe manufacturer" such as IBM and UniSys or > Cray. > > Dale > > donald tees wrote: > > > > Tim Shoppa wrote in message <394F3A96.534E7003@trailing-edge.com>... > > >Mike Buckler wrote: > > >> > > >> After searching whatis.com, webopedia.com, Google and Altavista, the only > > >> definition I could find was "a computer that supports hundreds of > > >> simultaneous users". Is that it? (Is that all there is, my friend? Then > > let's > > >> keep dancing. Let's break out the booze and have a ball... but I digress) > > >> If so does that mean that any modern box with a multiuser OS is a > > mainframe? > > > > > >A mainframe is, I'm afraid, whatever the users decide to call a > > >mainframe. I've worked in several shops where the end-users all > > >called the box in the computer room "the mainframe", whether the > > >"mainframe" was a real IBM behemoth or just a small workstation. > > >I think the most accurate assesment of "mainframe" is "the computer > > >in the computer room that the end-users aren't allowed to touch." > > > > > >This thread has appeared on alt.folklore.computers several times in > > >the past, I think one of my previous contributions was "You know it's > > >a mainframe if you've ever lost an oscilliscope inside of it." > > > > > >Tim. > > > > Sounds like the argument over what a "mini-computer" was twenty years back. > > I think the most useful definition then was > > "over $100,000 is a mainframe, under $100,000" is a mini". Perhaps under > > $10,000 is a micro. > > -- > For GPS data see: Joe -- http://joe.mehaffey.com > Peter -- http://www.vancouver-webpages.com/peter/ > Karen -- http://www.gpsy.com/gpsinfo/ > Dale -- http://users.cwnet.com/dalede ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 21 Jun 2000 19:41:31 GMT Organization: teabag Message-ID: <8ir5pb$h83$2@teabag.demon.co.uk> References: NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 961617090 nnrp-13:22547 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 21 Jun 2000 19:41:31 GMT X-Newsreader: knews 1.0b.0 Lines: 18 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!blackbush.xlink.net!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58299 In article , Shez writes: > As I understand it, originally a mainframe was just that - a big metal > frame for rack-mounting components, in the way that still happens in > some areas of electronics. A mainframe computer was a computer whose > components were rack mounted in a main frame. By the look of it, IBM's latest series seem to have gone back to that from the behemoths that they used to be; the current offerings look as if they could be a standard rack covered by some designer gear. Of course over the years people have categorised mainframes depending on such diverse subjects as overall cost, floor space, environmental considerations (ie water cooling & three phase power), IO/processing power ratios, what the marketing team wants to call it and so forth, so a definitive answer seems to be elusive. Chris. ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 22 Jun 2000 12:23:40 GMT Organization: The MITRE Corporation Lines: 14 Message-ID: <8it0gc$t98$1@top.mitre.org> References: <8ir5pb$h83$2@teabag.demon.co.uk> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 961676620 29992 128.29.251.13 (22 Jun 2000 12:23:40 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 22 Jun 2000 12:23:40 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!news.voicenet.com!uunet!ffx.uu.net!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:58274 cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) writes: >Of course over the years people have categorised mainframes depending >on such diverse subjects as overall cost, floor space, environmental >considerations (ie water cooling & three phase power), IO/processing >power ratios, what the marketing team wants to call it and so forth, >so a definitive answer seems to be elusive. In other words, a "mainframe" is defined as whatever *I* choose to call a "mainframe". YMMV (Your Mainframe May Vary) Joe Morris ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 22 Jun 2000 12:40:24 GMT Organization: IBM Austin Lines: 32 Message-ID: <8it1fo$1iea$1@ausnews.austin.ibm.com> References: <8ir5pb$h83$2@teabag.demon.co.uk> <8it0gc$t98$1@top.mitre.org> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!152.163.239.131!portc03.blue.aol.com!newsjunkie.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news2atm.raleigh.ibm.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58323 In <8it0gc$t98$1@top.mitre.org>, jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) writes: >cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) writes: > >>Of course over the years people have categorised mainframes depending >>on such diverse subjects as overall cost, floor space, environmental >>considerations (ie water cooling & three phase power), IO/processing >>power ratios, what the marketing team wants to call it and so forth, >>so a definitive answer seems to be elusive. > >In other words, a "mainframe" is defined as whatever *I* choose to >call a "mainframe". > >YMMV (Your Mainframe May Vary) > >Joe Morris And, of course, for some real obfuscation, consider the IBM S/370 machines. I think most people acknowledge them as a mainframe. But, does it apply to the entire product line? I think everyone would say that the S/370 models 3135, 3145, 3155, 3165, 3158, 3168, (usually referred to as the 135, 145, 155, 165, 158, and 168) and 303x were mainframes. But, what about the 43xx machines? Were they mainframes? They had the same architecture. What about the 9370 series of machines? They ran the same operating systems. What about the P/370 machines? Let's see, if a S/370 model 168 fell on me, I'd be squished flat. But, I've hefted a dozen P/370s in one hand. :-) Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### Message-ID: <395227EE.2E828310@earthlink.net> From: jchausler X-Mailer: Mozilla 4.72 [en] (Win95; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 37 Date: Thu, 22 Jun 2000 14:53:33 GMT NNTP-Posting-Host: 63.15.50.15 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 961685613 63.15.50.15 (Thu, 22 Jun 2000 07:53:33 PDT) NNTP-Posting-Date: Thu, 22 Jun 2000 07:53:33 PDT Organization: EarthLink Inc. -- http://www.EarthLink.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!logbridge.uoregon.edu!cyclone-west.rr.com!news.rr.com|news-west.rr.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58286 Shez wrote: > Mike Buckler of writes: > >After searching whatis.com, webopedia.com, Google and Altavista, the only > >definition I could find was "a computer that supports hundreds of > >simultaneous users". > > As I understand it, originally a mainframe was just that - a big metal > frame for rack-mounting components, in the way that still happens in > some areas of electronics. A mainframe computer was a computer whose > components were rack mounted in a main frame. I believe the original UNIVAC started as "room" built of a metal framework on and in which were mounted the various modules and mercury memory "tanks". Most other early computers I've ever seen or seen pictures of (and that's all I've seen of early UNIVACs) seemed to be more built of classic "relay racks" some open but more closed. I don't recall any others which were actually "rooms" (although many filled large rooms). Second generation machines frequently did not use standard 19 inch rack mount configurations and instead had some form of "custom framework" for supporting their modules. A company here in western NY, Mercury Aircraft, build a lot of these custom structures for IBM. I believe they're still in business. I recall disassembling a UNIVAC 1108 processor chassis further than it had ever been disassembled since being built in order to get pieces small enough to fit in an elevator (it had been installed by cutting a hole in the roof and dropping it in with a crane :-) It was definitely custom sized units. Chris AN GETTO$;DUMP;RUN,ALGOL,TAPE $$ ###### From: David Scheidt Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 22 Jun 2000 16:29:25 GMT Organization: EnterAct Corp. Lines: 21 Message-ID: <8itet5$2ccm$1@news.enteract.com> References: <394F3A96.534E7003@trailing-edge.com> NNTP-Posting-Host: shell-2.enteract.com X-Trace: news.enteract.com 961691365 78230 207.229.143.41 (22 Jun 2000 16:29:25 GMT) X-Complaints-To: abuse@enteract.com NNTP-Posting-Date: 22 Jun 2000 16:29:25 GMT User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (FreeBSD/3.3-STABLE (i386)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!feeder.via.net!newsfeed.enteract.com!betanews.enteract.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58305 Brian Inglis wrote: : The one distinguishing characteristic I have noticed on mainframe : systems is the presence of multiple I/O channels to which any I/O : interface could be attached. As compared to the minicomputer CPU : and I/O interfaces on the backplane, or the microcomputer CPU and : interfaces on the motherboard. Perhaps this might be a better : taxonomy. My MF experience was IBM but I have seen comments : implying similar architectures on DEC-10s, Univac 1100s, HIS/GE, : etc. Anyone care to compare/contrast across MF manufacturer : lines? Of course, this is becomming less true. There are now Intel based PCs with multiple PCI buses attached to the bus nexus. Big minis have had this for a long time, too. -- dscheidt@enteract.com But I simply can't find it anywhere in me to imagine that someone might want to stick sharp pasta spikes down into his penis. -D.M. Procida ###### From: "Simon Bowring" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Thu, 22 Jun 2000 17:29:33 +0100 (BST) Organization: MPC Data Limited Distribution: World Message-ID: References: Reply-To: "Simon Bowring" NNTP-Posting-Host: bath.mpc-data.co.uk X-NNTP-Posting-Host: mpc-data.demon.co.uk:158.152.55.245 X-Trace: news.demon.co.uk 961694404 nnrp-02:21153 NO-IDENT mpc-data.demon.co.uk:158.152.55.245 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Newsreader: PMINews 2.00.1205 For OS/2 Lines: 52 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!mpc-data.demon.co.uk!burton.mpc-data.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58290 Yes, *MOST* definately - IBM have documents on their site about this, but I can't find them right now. Modern Mainframes generally provide features that are not *ALL* found on other machines: 1. A very high level of fault tolerance and very low down-time. 2. Very high utilisation of resources such as CPU and disk (tends to use these resources much more efficiently than say Unix boxes, which tend to be "idle" and "peak out" more often) 3. Supports a very large number of users compared to the CPUs power [Probably many 10s or 100s of time the number of users that a a typical Unix box with the same power CPU will support). Your desktop PC probably has more CPU power than a 1980s mainframe, but it cannot service 100s or 1000s of users! 4. Very strong managability and tracability (peripherals produce lots of management data etc). 5. Very good backwards compatibility and long-term support (people commonly run 10-20 year old machines which are still supported but not manufactured!). Likewise 20 year old programs run unchanged on new kit. 6. Very high I/O bandwidth (compared to the CPU speed) to transfer data between CPU and disk and to users and other units etc 7. Very sophisticated VMs allowing the machine to be "partitioned" into mutiple virtual machines upon which you can run different operating systems, or test the next release etc without prejudicing your existing workload etc. This is really aspects of 1 and 4! 8. Very good expandability (clustering, extra CPUs, disk channels etc) Now many of the features listed above are gradually finding their way into Unixes and even things like NT, ***BUT*** they are *years* behind, and while these other platforms play catch-up, the mainframe marches forwards also. Currently only "mainframes" deliver the real reliability, manageability, efficiency and preservation of investment (the "legacy" word) that are (or should be) required for true long term "mission critical" 24x7 operation! IMO all arguments based on cost or CPU power are irrelevent; you get very expensive/powerful Super Computers which tend to run a Unix varient but do not exhibit all (or any of) the attributes listed above and are not usually called mainframes (and usually require special programming techniques to expoit their power). BTW: I'm fairly sure that only IBM or IBM compatible mainframes are still in production today - I'm sure someone will prove me wrong...! Simon Bowring ###### From: "Simon Bowring" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Thu, 22 Jun 2000 18:21:42 +0100 (BST) Organization: MPC Data Limited Distribution: World Message-ID: References: Reply-To: "Simon Bowring" NNTP-Posting-Host: bath.mpc-data.co.uk X-NNTP-Posting-Host: mpc-data.demon.co.uk:158.152.55.245 X-Trace: news.demon.co.uk 961698016 nnrp-14:1783 NO-IDENT mpc-data.demon.co.uk:158.152.55.245 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Newsreader: PMINews 2.00.1205 For OS/2 Lines: 27 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!dispose.news.demon.net!demon!news.demon.co.uk!demon!mpc-data.demon.co.uk!burton.mpc-data.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58376 >Even IBM appears to be using the phrase "enterprise server" (whatever that >means). So what's the problem, does a mainframe have to look like the cooling >jacket of a nuclear reactor? Can nothing be sold unless "enterprise" is in >the product name? I think you'll find the term is most apt (especially when compared with "mainframe"). An enterprise server is, uhm, a "server" (one of which) can service your entire "enterprise" (= business or organisation in IBM speak). Not too innappropriate! And as for docs that indicate the differences between mainframes and other systems, I haven't been able to find a definition, but if you're bored browse some of the docs at: http://www9.s390.ibm.com/marketing/position.html A comparsion with Unix machines is made at: http://www9.s390.ibm.com/marketing/gf225038.html Simon Bowring ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 22 Jun 2000 17:34:50 GMT Organization: teabag Message-ID: <8itinq$tl$1@teabag.demon.co.uk> References: NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 961695358 nnrp-11:29427 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 22 Jun 2000 17:34:50 GMT X-Newsreader: knews 1.0b.0 Lines: 14 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58302 In article , "Simon Bowring" writes: > BTW: I'm fairly sure that only IBM or IBM compatible mainframes are > still in production today - I'm sure someone will prove me > wrong...! Glad to be of assistance. :) Off the top of my head I can think of ICL, which share many of the attributes of IBM systems (even EBCDIC!) but are completely different beasts, and Compaq who're still making DEC's Alpha/VMS based mainframes (whether or not the latter are properly classed as mainframes seems to be more an argument of semantics rather than capability) Chris. ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 39 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 22 Jun 2000 21:45:15 GMT NNTP-Posting-Host: 209.63.29.106 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961710315 209.63.29.106 (Thu, 22 Jun 2000 15:45:15 MDT) NNTP-Posting-Date: Thu, 22 Jun 2000 15:45:15 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58421 "Simon Bowring" writes: > 1. A very high level of fault tolerance and very low down-time. close to 10 years ago i remember something about mainframe memory have 64/80 ecc (detect 16 bit errors and correct 15 bit errors) and there was work looking at reed-soloman error correcting codes. lot of current PC memory has fake parity ... although on some machines you can get 8/10 ecc (detect 2 bit errors and correct single bit errors). in the long distant past there was some documentation on distinction between calling something 195/370 (even if it wasn't quite all of 370) vis-a-vis 195/360 ... the distinction was the "370" had a lot more instruction retry for soft errors ... supposedly going from something like a couple hrs mean time between soft error failure (given the total number of components in a large 195 & the failure rate per component) to weeks or months mean time between soft error failure. there are also service related issues ... like discussed in mainframe thread http://www.garlic.com/~lynn/99.html#137 where every error on every machine is logged and evaluated. The reference in the thread was a particular scenerio involving huge concern about there being 15 total errors (of a particularly kind) over a 12 month period across several thousand installed machines ... not per machine over 12 month period .. but 15 total aggregate errors for all machines in 12 month period (many non-mainframe don't even bother to collect every error that has occured on every machine built). misc. references to prior "mainframe" threads here in alt.folklore.computers over the past 5-6 years: http://www.garlic.com/~lynn/subindex.html#mainframe -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8ins7k$3ai$1@news.igs.net> <394F8FC4.9E46A57F@cwnet.com> <8io6dn$imm$1@hiram.io.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 39 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Thu, 22 Jun 2000 21:06:44 EDT Date: Fri, 23 Jun 2000 01:06:44 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news-peer.gip.net!news.gsl.net!gip.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58455 In article <8io6dn$imm$1@hiram.io.com>, W Stiefer wrote: >would "server" apply? A "server" is a program that provides services by answering incoming requests from a variable number of requestors, known as "clients", by accepting requests at some well-known location from a plurality of other locations. For example, a web server is a server. So is a fileserver --- whether it's a Win98 box with shares or a Sun Starfire Enterprise 10000 serving a thousand clients with NFS. At home, I run an X11R6 server on the diskless Sun 3/60 in my bedroom; it provides the service of sending mouse and keyboard events to programs on the Linux box in the basement that request them, while allowing them to draw on the screen of the Sun. The Sun boots by requesting its IP address from a RARP server built into the kernel of the Linux box, followed by fetching a netboot proram from a TFTP server running on the Linux box, followed by obtaining boot parameters from a bootparamd server running on the Linux box (mediated by a portmapping server running on the selfsame Linux box), followed by connecting to a filesystem mount server "mountd" running on the Linux box (mediated by the portmapper), followed by connecting to a filesystem export server "nfsd" running on the Linux box (again mediated by the portmapper), followed by running some programs it finds on that disk, running through the process again from the bootparamd step, and running the X server --- which then connects to an XDMCP server running on the Linux box, which connects as a client to the X server. In microkernel systems, the process that emulates Unix (by providing Unix system call services to other processes, typically running on the same machine) is also called a server. So the answer is "no". And you should know better than to quote a huge article and add one line of your stupidity to it. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <394F3A96.534E7003@trailing-edge.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 17 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Thu, 22 Jun 2000 21:09:11 EDT Date: Fri, 23 Jun 2000 01:09:11 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58459 In article , Brian Inglis wrote: >The one distinguishing characteristic I have noticed on mainframe >systems is the presence of multiple I/O channels to which any I/O >interface could be attached. As compared to the minicomputer CPU >and I/O interfaces on the backplane, or the microcomputer CPU and >interfaces on the motherboard. Perhaps this might be a better >taxonomy. So the SGI Octanes and Visual Workstations, with their crossbar I/O architecture, are mainframes? How about a Beowulf with an Ethernet switch as a backplane? -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Distribution: World Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 187 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Thu, 22 Jun 2000 22:02:03 EDT Date: Fri, 23 Jun 2000 02:02:03 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news-peer.gip.net!news.gsl.net!gip.net!howland.erols.net!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58470 In article , Simon Bowring wrote: >Modern Mainframes generally provide features that are not *ALL* found >on other machines: This is a more interesting definition than most on this thread. I really appreciate you posting it. >1. A very high level of fault tolerance and very low down-time. This is, of course, a matter of degree; clustering approaches like Tandem's are going to bring this to the PC world. >2. Very high utilisation of resources such as CPU and disk (tends > to use these resources much more efficiently than say Unix boxes, > which tend to be "idle" and "peak out" more often) This is more an issue of how you use your machine than how you design it; if you run mostly long-running batch jobs with predictable I/O patterns, you can get this on any kind of machine. >3. Supports a very large number of users compared to the CPUs power > [Probably many 10s or 100s of time the number of users that a > a typical Unix box with the same power CPU will support). Your > desktop PC probably has more CPU power than a 1980s mainframe, > but it cannot service 100s or 1000s of users! My desktop PC at home is a K6-2/450. I've heard it can saturate 100Mbps Ethernet with minimal-size packets, and that it can handle hundreds of HTTP requests per second --- much more than enough to support thousands of HTTP users, given that even heavy HTTP users normally perform only a few requests per minute. I haven't tested either assertion. My Sun-3 at home is capable of supporting ten or so concurrent users connected via telnet; I'm sure my PC could support hundreds if I tweaked a couple of kernel parameters. My work PC is a 750MHz Coppermine PIII, capable of considerably more. 1980s and even 1970s mainframes could handle hundreds or thousands of users only by not running a whole process for every user. Instead, a single TP monitor process would talk to hundreds or thousands of 3270s or similar terminals, sending things very similar to web forms and receiving responses very similar to web form submissions. [Am I full of nonsense here? I was born in 1976 and have never used a mainframe.] Now my machine can do the same thing. I think it would be really cool to graft something like a $40 Nokia cellphone onto a $1 USB interface for a low-cost terminal; you could hook hundreds of these guys to a commodity PC for $50-$80 per port. >4. Very strong managability and tracability (peripherals produce > lots of management data etc). If I understand what you're talking about, this is coming to the PC world too, although it's not widely used yet; IBM drives, among others, include something called S.M.A.R.T., which reports error and near-error statistics; even $10 Ethernet cards have long supported error statistics gathering; now every Intel motherboard includes sensors that report fan RPM, motherboard temperature, and CPU temperature; and some motherboards even include the ability to log ECC-corrected errors. If you're running Linux, tools for grabbing, summarizing, and reporting on all this info are available. >5. Very good backwards compatibility and long-term support (people > commonly run 10-20 year old machines which are still supported > but not manufactured!). Likewise 20 year old programs run unchanged > on new kit. Although I'm trying not to be too much of a free-software zealot, the Sun-3 in my bedroom was manufactured in November, 1988; it's not supported by Sun, but it's certainly well-supported. [By the way, it was originally used as a database server, among other things; it was called "thor", and was at CMU. It was presumably thor.cs.cmu.edu, although I don't know that. Its previous owner, Frank Skelly, told me that the first implementation of SNMPv1 was developed partly on that machine, and that it was the first non-VAX incarnation of thor at CMU. Does anyone in this newsgroup remember this machine? :)] I don't know about 20-year-old programs, but I have downloaded and recompiled 10-year-old programs and run them without problems. The only 20-year-old program I've tried to run is Visicalc 1.0 for DOS, which I believe worked unchanged in dosemu. >6. Very high I/O bandwidth (compared to the CPU speed) to transfer > data between CPU and disk and to users and other units etc Can you give me some numbers here? I'm thinking of things like the SGI Visual Workstation and like Beowulfs, which use fairly high-bandwidth crossbar connections. I note that, according to Lynn Wheeler's past posts, this attribute has become steadily less pronounced over the years. >7. Very sophisticated VMs allowing the machine to be "partitioned" > into mutiple virtual machines upon which you can run different > operating systems, or test the next release etc without prejudicing > your existing workload etc. This is really aspects of 1 and 4! Well, AFAIK, only IBM and compatible mainframes offered this, so you can't really cite it as a distinguishing feature of mainframes in general; but you can do this to some extent today on a PC with VMWare. This lets you run, say, Windows NT in a window on your Linux machine. There's also a free-software emulator, called "Bochs", that lets you do the same thing, but much more slowly, and there's a project to replace VMWare with free software called "plex86". >8. Very good expandability (clustering, extra CPUs, disk channels etc) This is something most PCs don't have; clustering, USB, FireWire, and Fibre Channel[0] are probably the way up. >Now many of the features listed above are gradually finding their way >into Unixes and even things like NT, ***BUT*** they are *years* behind, >and while these other platforms play catch-up, the mainframe marches >forwards also. So what distinguishes mainframes from other machines *today*? And what will distinguish them tomorrow? >Currently only "mainframes" deliver the real reliability, manageability, >efficiency and preservation of investment (the "legacy" word) that are >(or should be) required for true long term "mission critical" 24x7 >operation! That's certainly true in some circumstances, but it's rather an unjustifiably absolute statement, don't you think? There's a whole spectrum in here of cost of downtime for "mission critical 24x7 operation", from $100 per month or so up to hundreds of millions of dollars per minute, along with various kinds of tolerance for scheduled downtime (if you're running a bank's machine, you can take it down for scheduled maintenance for hours every night if you have to, while if you're running a phone switch in a CO, taking it down for more than a few seconds will irritate your customers.) Further, whether or not mainframes are good for preservation of investment depends largely on where your investment is now. If your "investment" is in MCSE exam preparation courses for your staff, well, they may be fairly useless now and completely obsolete in three years, but if you switch to an S/390, they'll be useless immediately! Manageability is equally nonscalar. For running a phone switch, a 3B1 is probably much more manageable than an S/390; for routing network packets, you'd likely be much better off with a SuperStack or something than an S/390; for various embedded tasks, you'll almost certainly prefer a tiny stripped-down environment that's much simpler to manage than Unix or MVS; and for your Web server, you might as well run Linux on a PC and only administer Linux as run Linux on your S/390 and have to deal with the funky mainframe hardware and environment, no? "Efficiency" is pretty hard to define. What do you mean by it? >IMO all arguments based on cost or CPU power are irrelevant; Agreed. >you get very expensive/powerful Super Computers which tend to run a >Unix varient but do not exhibit all (or any of) the attributes >listed above and are not usually called mainframes (and usually >require special programming techniques to expoit their power). Yep. Although they usually require quite a bit of these attributes these days; take a look at the Tera^H^H^H^HCray MTA architecture, e.g. >BTW: I'm fairly sure that only IBM or IBM compatible mainframes are > still in production today - I'm sure someone will prove me > wrong...! Well, there are Honeywell Bull and Burroughs mainframes from Unisys. And, although I don't know much about them, I think Tandem^H^H^H^H^H^HCompaq's NonStop machines run a rather different OS than IBM's. And doesn't HP make mainframes anymore? [0] Despite being Americans, they changed the spelling to "Fibre", ostensibly so people wouldn't think it was limited to fiber optics. Is that the stupidest naming decision you've ever heard of or what? -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: optcamel@ix.netcom.com (Howard and Kelly Lute) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Thu, 22 Jun 2000 19:39:20 -0700 Organization: Optimum Camel Companies, Ltd. Lines: 31 Message-ID: References: NNTP-Posting-Host: cf.df.a8.49 X-Server-Date: 23 Jun 2000 02:44:12 GMT X-Newsreader: MicroPlanet Gravity v2.12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!firehose.mindspring.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58452 In article , news@xerez.demon.co.uk says... > Mike Buckler of writes: > >After searching whatis.com, webopedia.com, Google and Altavista, the only > >definition I could find was "a computer that supports hundreds of > >simultaneous users". > > As I understand it, originally a mainframe was just that - a big metal > frame for rack-mounting components, in the way that still happens in > some areas of electronics. A mainframe computer was a computer whose > components were rack mounted in a main frame. > > -Shez. > You are largely correct. The 360, 370 series of IBM Mainframes (Models 30, 40, 50, 65, 75, 85, 95) all used a matrix of 10" X 12" boards INDIVIDUALLY wired (via 3 and 4 layer sandwich) in a large door like frame that held between 12 and 20 of these boards, each board then held up to 104 individual boards. They were/are amazing machines, some of the most complex machines ever devised. I was fortunate to get to service all of these during my career as a field engineer for IBM and a third party service vendor Comma Corp (later bought by Control Data and ruined). -- friend, --------------------- Howard and Kelly Lute He: Electronics Instructor, Terrible Mechanic, Worse Plumber She: Patient eBay: kell EBay Rating: 660 LPFM Site: http://pw1.netcom.com/~optcamel/fmradio.htm Try this site BEFORE you buy -->http://www.epinions.com ###### From: optcamel@ix.netcom.com (Howard and Kelly Lute) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Thu, 22 Jun 2000 19:41:25 -0700 Organization: Optimum Camel Companies, Ltd. Lines: 47 Message-ID: References: <8ir5pb$h83$2@teabag.demon.co.uk> <8it0gc$t98$1@top.mitre.org> <8it1fo$1iea$1@ausnews.austin.ibm.com> NNTP-Posting-Host: cf.df.a8.49 X-Server-Date: 23 Jun 2000 02:46:09 GMT X-Newsreader: MicroPlanet Gravity v2.12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!blackbush.xlink.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!netnews.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!firehose.mindspring.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58444 In article <8it1fo$1iea$1@ausnews.austin.ibm.com>, glass2@glass2.lexington.ibm.com says... > In <8it0gc$t98$1@top.mitre.org>, jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) writes: > >cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) writes: > > > >>Of course over the years people have categorised mainframes depending > >>on such diverse subjects as overall cost, floor space, environmental > >>considerations (ie water cooling & three phase power), IO/processing > >>power ratios, what the marketing team wants to call it and so forth, > >>so a definitive answer seems to be elusive. > > > >In other words, a "mainframe" is defined as whatever *I* choose to > >call a "mainframe". > > > >YMMV (Your Mainframe May Vary) > > > >Joe Morris > > And, of course, for some real obfuscation, consider the IBM S/370 > machines. I think most people acknowledge them as a mainframe. > But, does it apply to the entire product line? I think everyone > would say that the S/370 models 3135, 3145, 3155, 3165, 3158, 3168, > (usually referred to as the 135, 145, 155, 165, 158, and 168) and > 303x were mainframes. But, what about the 43xx machines? Were they > mainframes? They had the same architecture. What about the 9370 > series of machines? They ran the same operating systems. What > about the P/370 machines? Let's see, if a S/370 model 168 fell on > me, I'd be squished flat. But, I've hefted a dozen P/370s in one > hand. :-) > > Dave > > P.S. Standard Disclaimer: I work for them, but I don't speak for them. > > I'd say all of the earlier machines were mainframes...the sys/360 and 370 series...the 4XXX series was too but they were certainly different from their predecessors. -- friend, --------------------- Howard and Kelly Lute He: Electronics Instructor, Terrible Mechanic, Worse Plumber She: Patient eBay: kell EBay Rating: 660 LPFM Site: http://pw1.netcom.com/~optcamel/fmradio.htm Try this site BEFORE you buy -->http://www.epinions.com ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 23 Jun 2000 03:55:27 GMT Organization: The PACSIBM SIG BBS Lines: 17 Message-ID: <8iun3f$3p1@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!205.252.116.205!howland.erols.net!news-xfer.netaxs.com.MISMATCH!news-xfer.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58467 In reading the 1958 proposal, the term "mainframe" was used as two words and to two the CPU, back then spelled as central processing unit. In those days, the machine wasn't a single box, but many boxes; and the some of the units could and did work independently off line on separate tasks (ie card-to-tape, tape-to-print, tape-to-tape, etc.). I think in that context, the word "central" in CPU has more meaning than we think of today, because there were other "processing units". In an earlier post, someone message mainframes of the 70s couldn't handle lots of users without not doing anything else. The load of course depends on the specific model and workload of terminals and batch jobs, but even low end 1960s machines like the S/360-40 could support ongoing batch and on-line terminal processing. Do not under estimate the I/O structure of S/360-390. It was and is extremely powerful. ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 23 Jun 2000 12:38:47 GMT Organization: IBM Austin Lines: 51 Message-ID: <8ivlon$1adm$1@ausnews.austin.ibm.com> References: <8iun3f$3p1@netaxs.com> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!znr.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news2atm.raleigh.ibm.com!tornews.torolab.ibm.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58430 In <8iun3f$3p1@netaxs.com>, lwinson@bbs.cpcn.com (lwin) writes: >In reading the 1958 proposal, the term "mainframe" was used as two words >and to two the CPU, back then spelled as central processing unit. In >those days, the machine wasn't a single box, but many boxes; and the >some of the units could and did work independently off line on separate >tasks (ie card-to-tape, tape-to-print, tape-to-tape, etc.). > >I think in that context, the word "central" in CPU has more meaning than >we think of today, because there were other "processing units". > >In an earlier post, someone message mainframes of the 70s couldn't >handle lots of users without not doing anything else. The load of >course depends on the specific model and workload of terminals and batch >jobs, but even low end 1960s machines like the S/360-40 could support >ongoing batch and on-line terminal processing. > >Do not under estimate the I/O structure of S/360-390. It was and is >extremely powerful. Also, don't forget that the early mainframes were constrained by very limited memory sizes. Here's a summary from a table in appendix D of the book 'Programming the IBM 360' by Clarence B. Germain (1967): 360 MODELS # Core Size 25 30 40 50 65 75 85 91-95 44 67 20 B 4K......X.............................................X. C 8K......X...X.........................................X. D 16K......X...X...X.....................................X. E 32K......X...X...X.............................X......... F 64K..........X...X...X.........................X......... G 128K..............X...X...X.....................X......... H 256K..............X...X...X...X.................X..X...... I 512K..................X...X...X...X................X...... J 1024K..................X...X...X...X.....X..........X...... K 2048K..............................X.....X..........X...... L 4096K..............................X.....X................. 2361 1024K..................X...X...X........................... 2361 2048K..................X...X...X........................... CD 24K...................................................... DE 48K...................................................... Cycle time(uS) .9 1.5 2.5 2.0 .75 .75 1.04 .75-.125 1.0 .75 3.6 per _ bytes 2 1 2 4 8 8 16 8 4 8 1 Of course, remember that the S/360 machines did not support virtual memory (with the exception of the model 67). Thus, what you had was all you got! Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### Message-ID: <39539BD3.D1BAD5A8@cwnet.com> Date: Fri, 23 Jun 2000 10:18:11 -0700 From: Dale DePriest X-Mailer: Mozilla 4.73 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 209.160.185.159 X-Trace: 23 Jun 2000 10:24:40 -0700, 209.160.185.159 Lines: 71 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!128.230.129.106!news.maxwell.syr.edu!newspeer1.nac.net!newspeer.cwnet.com!news2.cwnet.com!209.160.185.159 Xref: chonsp.franklin.ch alt.folklore.computers:58336 Interesting but not quite accurate. For the model 50 and up you could get extended memory in a box. For ibm it came in 2Meg chucks of 8 microsecend memory but from Ampex is was 2 microsecond memory (actually the same speed as model 50 internal memory). You could add several boxes if you wished and had the money. Internally the Ampex boxes used submicrosecond memory but by the time you got through the interface and cabling we could be only about 1.8 microseconds out of the beast. This memory box, by the way, would have qualified as a mainframe by most of the definitions related to cost, weight, and size put forth in this thread. Dale glass2@glass2.lexington.ibm.com wrote: > > In <8iun3f$3p1@netaxs.com>, lwinson@bbs.cpcn.com (lwin) writes: > >In reading the 1958 proposal, the term "mainframe" was used as two words > >and to two the CPU, back then spelled as central processing unit. In > >those days, the machine wasn't a single box, but many boxes; and the > >some of the units could and did work independently off line on separate > >tasks (ie card-to-tape, tape-to-print, tape-to-tape, etc.). > > > >I think in that context, the word "central" in CPU has more meaning than > >we think of today, because there were other "processing units". > > > >In an earlier post, someone message mainframes of the 70s couldn't > >handle lots of users without not doing anything else. The load of > >course depends on the specific model and workload of terminals and batch > >jobs, but even low end 1960s machines like the S/360-40 could support > >ongoing batch and on-line terminal processing. > > > >Do not under estimate the I/O structure of S/360-390. It was and is > >extremely powerful. > > Also, don't forget that the early mainframes were constrained by very > limited memory sizes. Here's a summary from a table in appendix D of > the book 'Programming the IBM 360' by Clarence B. Germain (1967): > > 360 MODELS > # Core Size 25 30 40 50 65 75 85 91-95 44 67 20 > B 4K......X.............................................X. > C 8K......X...X.........................................X. > D 16K......X...X...X.....................................X. > E 32K......X...X...X.............................X......... > F 64K..........X...X...X.........................X......... > G 128K..............X...X...X.....................X......... > H 256K..............X...X...X...X.................X..X...... > I 512K..................X...X...X...X................X...... > J 1024K..................X...X...X...X.....X..........X...... > K 2048K..............................X.....X..........X...... > L 4096K..............................X.....X................. > 2361 1024K..................X...X...X........................... > 2361 2048K..................X...X...X........................... > CD 24K...................................................... > DE 48K...................................................... > > Cycle time(uS) .9 1.5 2.5 2.0 .75 .75 1.04 .75-.125 1.0 .75 3.6 > per _ bytes 2 1 2 4 8 8 16 8 4 8 1 > > Of course, remember that the S/360 machines did not support virtual memory > (with the exception of the model 67). Thus, what you had was all you got! > > Dave > > P.S. Standard Disclaimer: I work for them, but I don't speak for them. -- For GPS data see: Joe -- http://joe.mehaffey.com Peter -- http://www.vancouver-webpages.com/peter/ Karen -- http://www.gpsy.com/gpsinfo/ Dale -- http://users.cwnet.com/dalede ###### From: Eric Sosman Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Fri, 23 Jun 2000 13:41:37 -0400 Organization: Sun Microsystems Lines: 12 Message-ID: <3953A151.CB044626@east.sun.com> References: NNTP-Posting-Host: tardis.east.sun.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!blackbush.xlink.net!newsfeed.germany.net!news.vas-net.net!newsfeed.gamma.ru!Gamma.RU!easynet-melon!easynet.net!btnet-feed2!btnet!carbon.eu.sun.com!new-usenet.uk.sun.com!eastnews1.east.sun.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58348 Howard and Kelly Lute wrote: > > You are largely correct. The 360, 370 series of IBM Mainframes (Models > 30, 40, 50, 65, 75, 85, 95) [...] I can understand omitting the Model 20 from this list, but why leave out Models 44 and 67? Well, okay, maybe the 44 is disqualified by its reduced instruction set (and by the rotary dial which controlled floating-point precision!), but the 67 was an impressive chunk of metal. -- Eric.Sosman@east.sun.com ###### From: optcamel@ix.netcom.com (Howard and Kelly Lute) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Fri, 23 Jun 2000 12:13:56 -0700 Organization: Optimum Camel Companies, Ltd. Lines: 63 Message-ID: References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> NNTP-Posting-Host: cf.df.a8.78 X-Server-Date: 23 Jun 2000 19:25:25 GMT X-Newsreader: MicroPlanet Gravity v2.12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!firehose.mindspring.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58447 In article <8ivlon$1adm$1@ausnews.austin.ibm.com>, glass2@glass2.lexington.ibm.com says... > In <8iun3f$3p1@netaxs.com>, lwinson@bbs.cpcn.com (lwin) writes: > >In reading the 1958 proposal, the term "mainframe" was used as two words > >and to two the CPU, back then spelled as central processing unit. In > >those days, the machine wasn't a single box, but many boxes; and the > >some of the units could and did work independently off line on separate > >tasks (ie card-to-tape, tape-to-print, tape-to-tape, etc.). > > > >I think in that context, the word "central" in CPU has more meaning than > >we think of today, because there were other "processing units". > > > >In an earlier post, someone message mainframes of the 70s couldn't > >handle lots of users without not doing anything else. The load of > >course depends on the specific model and workload of terminals and batch > >jobs, but even low end 1960s machines like the S/360-40 could support > >ongoing batch and on-line terminal processing. > > > >Do not under estimate the I/O structure of S/360-390. It was and is > >extremely powerful. > > Also, don't forget that the early mainframes were constrained by very > limited memory sizes. Here's a summary from a table in appendix D of > the book 'Programming the IBM 360' by Clarence B. Germain (1967): > > 360 MODELS > # Core Size 25 30 40 50 65 75 85 91-95 44 67 20 > B 4K......X.............................................X. > C 8K......X...X.........................................X. > D 16K......X...X...X.....................................X. > E 32K......X...X...X.............................X......... > F 64K..........X...X...X.........................X......... > G 128K..............X...X...X.....................X......... > H 256K..............X...X...X...X.................X..X...... > I 512K..................X...X...X...X................X...... > J 1024K..................X...X...X...X.....X..........X...... > K 2048K..............................X.....X..........X...... > L 4096K..............................X.....X................. > 2361 1024K..................X...X...X........................... > 2361 2048K..................X...X...X........................... > CD 24K...................................................... > DE 48K...................................................... > > Cycle time(uS) .9 1.5 2.5 2.0 .75 .75 1.04 .75-.125 1.0 .75 3.6 > per _ bytes 2 1 2 4 8 8 16 8 4 8 1 > > Of course, remember that the S/360 machines did not support virtual memory > (with the exception of the model 67). Thus, what you had was all you got! > > Dave > > P.S. Standard Disclaimer: I work for them, but I don't speak for them. LCS (Large Core Storage) could be added to Mod 50's and up and could add as much as 32MB (Two boxes) to each MF. -- friend, --------------------- Howard and Kelly Lute He: Electronics Instructor, Terrible Mechanic, Worse Plumber She: Patient eBay: kell EBay Rating: 660 LPFM Site: http://pw1.netcom.com/~optcamel/fmradio.htm Try this site BEFORE you buy -->http://www.epinions.com ###### From: optcamel@ix.netcom.com (Howard and Kelly Lute) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Fri, 23 Jun 2000 12:15:58 -0700 Organization: Optimum Camel Companies, Ltd. Lines: 24 Message-ID: References: <3953A151.CB044626@east.sun.com> NNTP-Posting-Host: cf.df.a8.78 X-Server-Date: 23 Jun 2000 19:27:27 GMT X-Newsreader: MicroPlanet Gravity v2.12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!firehose.mindspring.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58440 In article <3953A151.CB044626@east.sun.com>, Eric.Sosman@east.sun.com says... > Howard and Kelly Lute wrote: > > > > You are largely correct. The 360, 370 series of IBM Mainframes (Models > > 30, 40, 50, 65, 75, 85, 95) [...] > > I can understand omitting the Model 20 from this list, but why > leave out Models 44 and 67? Well, okay, maybe the 44 is disqualified > by its reduced instruction set (and by the rotary dial which controlled > floating-point precision!), but the 67 was an impressive chunk of metal. > > Yeah I worked on 67's, 65, 50's, mod 75...all in SF. Fun machines with LOTS of tools to help troubleshoot the ol' girls. Big, Fast, Fun. -- friend, --------------------- Howard and Kelly Lute He: Electronics Instructor, Terrible Mechanic, Worse Plumber She: Patient eBay: kell EBay Rating: 660 LPFM Site: http://pw1.netcom.com/~optcamel/fmradio.htm Try this site BEFORE you buy -->http://www.epinions.com ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 28 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Fri, 23 Jun 2000 15:47:23 EDT Date: Fri, 23 Jun 2000 19:47:23 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!newscore.gigabell.net!newsfeed00.sul.t-online.de!t-online.de!newsfeed.wirehub.nl!newsfeed.icl.net!feed2.onemain.com!feed1.onemain.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58462 In article <8iun3f$3p1@netaxs.com>, lwin wrote: >In an earlier post, someone message mainframes of the 70s couldn't >handle lots of users without not doing anything else. The load of >course depends on the specific model and workload of terminals and batch >jobs, but even low end 1960s machines like the S/360-40 could support >ongoing batch and on-line terminal processing. I don't know if you're referring to my post or not, but what I said is that mainframes of the 70s and 80s supported hundreds or thousands of users by connecting many users to a single process, rather than by creating one or more processes per user. I'm not sure what a "process" maps to in the mainframe world, but I'm thinking of something including at least one stack, a set of memory mappings, and some information in the supervisor about things like open files^H^H^H^H^Hdatasets. Does such an entity exist in MVS? What do you call it? >Do not under estimate the I/O structure of S/360-390. It was and is >extremely powerful. I'm interested in hearing more details. I'm in no position to estimate it at all, let alone underestimate it :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 23 Jun 2000 20:07:01 GMT Organization: The PACSIBM SIG BBS Lines: 32 Message-ID: <8j0g15$ev0@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!news-xfer.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58460 > I'm not sure what a "process" maps to in the mainframe world, but I'm > thinking of something including at least one stack, a set of memory > mappings, and some information in the supervisor about things like open > files^H^H^H^H^Hdatasets. Does such an entity exist in MVS? What do > you call it? Yes, almost since the beginning. Called either multiprogramming or multi-tasking. On S/360 principal operating systems, the partitions or regions were all independent from each other and work could function simultaneously without impacting other regions. There were two main o/s's--DOS and OS. OS was more complex, intended for bigger machines, and handled more work and options. OS evolved into MVS. S/360 hardware protected the partitions. A storage protection key was built in the machine to prevent bad access. S/360 hardware had two states "supervisor" and "problem (application)". Certain functions had to be done by the supervisor only. The supervisor kept track of the various partitions and what was going on. There was a system of interupts issued by programs and I/O and the supervisor tracked them. A principal online terminal manager program, CICS, allows each terminal to be running a separate program accessing separate data files. I feel strongly about this since certain _contemporary_ hardware and software manufacturers, that while extremely popular today, don't seem to have grasped some of the basic S/360 principles implemented 35 years ago. ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 23 Jun 2000 20:47:41 GMT Organization: Neosoft (using Airnews.net!) Lines: 16 Message-ID: X-Orig-Message-ID: References: <8iun3f$3p1@netaxs.com> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Fri Jun 23 15:47:41 2000 NNTP-Posting-Host: !Y@%21k-XBFH"87 (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!HSNX.atgi.net!cabal11.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58375 On Fri, 23 Jun 2000 19:47:23 GMT, Kragen Sitaker wrote: >I'm not sure what a "process" maps to in the mainframe world, but I'm >thinking of something including at least one stack, a set of memory >mappings, and some information in the supervisor about things like open >files^H^H^H^H^Hdatasets. Does such an entity exist in MVS? What do >you call it? Well, kinda sorta. The 370 architecture doesn't have a stack per se (yes, there's a linkage stack, but that was a 370/XA addition). A job (or started task, or TSO user) occupies an address space, which corresponds to a set of memory mappings and I/O definitions (what JCL is really good for). An address space can have multiple tasks, each representing a different running program; each TCB (task control block) is dispatchable on a different CPU in the configuration. Does that answer your question, or just confuse you more? :-) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 36 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Fri, 23 Jun 2000 22:56:24 EDT Date: Sat, 24 Jun 2000 02:56:24 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58473 In article , Jay Maynard wrote: >On Fri, 23 Jun 2000 19:47:23 GMT, Kragen Sitaker wrote: >>I'm not sure what a "process" maps to in the mainframe world, but I'm >>thinking of something including at least one stack, a set of memory >>mappings, and some information in the supervisor about things like open >>files^H^H^H^H^Hdatasets. Does such an entity exist in MVS? What do >>you call it? > >Well, kinda sorta. The 370 architecture doesn't have a stack per se (yes, >there's a linkage stack, but that was a 370/XA addition). A job (or started >task, or TSO user) occupies an address space, which corresponds to a set of >memory mappings and I/O definitions (what JCL is really good for). An >address space can have multiple tasks, each representing a different running >program; each TCB (task control block) is dispatchable on a different CPU in >the configuration. > >Does that answer your question, or just confuse you more? :-) It definitely answers it; the job/started task/TSO user maps closely to a Win32 or Solaris "process", and the tasks map closely to threads. So now my question is: how many users could a 1970s mainframe reasonably support with TSO? More than 200? How about a 1980s mainframe? I know that today's S/390s can run 40,000+ Linux VMs. My original assertion was that 1970s and 1980s mainframes were able to support hundreds or thousands of users by using a single job that served many users. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j0g15$ev0@netaxs.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 36 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Fri, 23 Jun 2000 23:01:26 EDT Date: Sat, 24 Jun 2000 03:01:26 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58471 In article <8j0g15$ev0@netaxs.com>, lwin wrote: >. . . the partitions or regions were >all independent from each other and work could function simultaneously >without impacting other regions . . . > >S/360 hardware protected the partitions. A storage protection key >was built in the machine to prevent bad access. > >S/360 hardware had two states "supervisor" and "problem (application)". >Certain functions had to be done by the supervisor only. The supervisor >kept track of the various partitions and what was going on. There >was a system of interupts issued by programs and I/O and the supervisor >tracked them. > >I feel strongly about this since certain _contemporary_ hardware >and software manufacturers, that while extremely popular today, don't >seem to have grasped some of the basic S/360 principles implemented >35 years ago. Are you referring to the principles described in what I've quoted above? Which hardware and software are you thinking of? Unix and Windows NT (and their supporting hardware) conform to the description you have listed above (except when bugs strike, of course), although many processors and OSes used in embedded systems don't. >A principal online terminal manager program, CICS, allows each terminal >to be running a separate program accessing separate data files. CICS ran (and runs) in one partition with a single memory map, so a bug in the program running on one terminal can crash the programs on the other terminals, right? -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 24 Jun 2000 13:30:23 GMT Organization: Neosoft (using Airnews.net!) Lines: 66 Message-ID: <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net> X-Orig-Message-ID: References: <8iun3f$3p1@netaxs.com> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sat Jun 24 08:30:23 2000 NNTP-Posting-Host: !b1=<1k-Wj#$\L[ (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58480 On Sat, 24 Jun 2000 02:56:24 GMT, Kragen Sitaker wrote: >It definitely answers it; the job/started task/TSO user maps closely to >a Win32 or Solaris "process", and the tasks map closely to threads. Pretty much, though a TCB is a bit more expensive to create than most threads, and an address space is extremely expensive to create, definitely unliks a Unix process. >So now my question is: how many users could a 1970s mainframe >reasonably support with TSO? More than 200? How about a 1980s >mainframe? The systems I worked on in the 80s tended to be the larger end of the IBM spectrum, and would support 50-70 TSO users at once - not so much due to limitations of the system's ability to support them, but rather because that was as many users who needed TSO. In the typical mainframe shop, programmers, operators, and administrators (systems types) were the only ones that had TSO. The real work of the enterprise was done with applications running under CICS or IMS/DC, and users logged on to that. I don't know how many logged-on TSO users a typical 3081/3084-class system would support...because nobody ever ran them that way. >My original assertion was that 1970s and 1980s mainframes were able to >support hundreds or thousands of users by using a single job that >served many users. This is definitely the case, but, OTOH, they also provided the facilities that made that possible. Unix was never intended to carry that kind of a load, and so nobody ever wrote a CICS-style TP monitor for it. Because of that, every user has an address space, and every user has his own amount of overhead, and so supporting thousands of users takes more resources. Mainframes did what they did because they had to; even at the end of the 80s, 256 MB was a huge system, but would support tens of thousands of users without breathing hard. To answer a question in another post, yes, CICS was a single address space. Not only that, but it ran all applications under a single TCB, and only used separate TCBs for specific functions (the first, for example, was a separate TCB to do VSAM I/O). CICS applications have to be reentrant between CICS system calls, and are only loaded once into memory and shared. CICS multitasking is cooperative, with a runaway task timer that ABENDs a task that doesn't issue a CICS call within a configurable length of time (typically 10 seconds). As CICS systems became too large to fit in the 8-10 MB of available address space in MVS/370, shops would break them up into multiple regions, each in a separate MVS started task, along application boundaries; even that grew too restrictive, and facilities were added to allow CICS regions to communicate requests for resources to other CICS regions in the same MVS instance (with cross-memory calls), or on other MVS instances (via VTAM). All of this left CICS vulnerable to an application stomping on critical control blocks; in practice, however, this was rare for shops that used high-level languages and the CICS command-level interface, both of which hid the details of control blocks and low-level addressing from the programmer's ability to screw up. I don't know when you could use PL/I or COBOL for CICS, but the command-level interface wasn't available until 1978 or so (with CICS 1.3). This doesn't mean it never happened, and the layout of control blocks in CICS's memory wasn't exactly designed for robustness (critical task-related control blocks were located next to application work areas in such a way that overrunning them did Bad Things), and when it did, debugging could be a real bitch; however, nearly every shop that ran CICS had production and test regions so applications programmers could do development and debugging without breaking the critical work of the system. ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 24 Jun 2000 14:55:33 GMT Organization: The MITRE Corporation Lines: 39 Message-ID: <8j2i55$f1a$1@top.mitre.org> References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 961858533 15402 128.29.251.13 (24 Jun 2000 14:55:33 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 24 Jun 2000 14:55:33 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:58476 glass2@glass2.lexington.ibm.com writes: [extracted:] >360 MODELS ># Core Size 25 30 40 50 65 75 85 91-95 44 67 20 >G 128K..............X...X...X.....................X......... >H 256K..............X...X...X...X.................X..X...... >I 512K..................X...X...X...X................X...... >J 1024K..................X...X...X...X.....X..........X...... I don't recall ever hearing of a mod 65 with 128 KB of memory. On that model you could attach 1, 2, 3, or 4 boxes of 256 KB (IIRC the memory boxes were machine type 2365). Note that a 768 KB machine (or any S/360 model) would be called a model IH; in general, letters represented power-of-two memory size, and they were combined when the machine had a combination that was not a power of 2. >K 2048K..............................X.....X..........X...... Some models with 2 MB memory were documented as "JJ" instead of K, but I'm not sure why. >2361 1024K..................X...X...X........................... >2361 2048K..................X...X...X........................... These aren't S/360 models; they are separate outboard memory boxes known as "Large Core Storage", or LCS. They were (comparatively) inexpensive because while the 2365 memory had a cycle time of 600 nsec or so, the LCS box cycle time was 8 usec. We used LCS on our model 65; one of the local mods I wrote was to modify IPLTXT (the bootstrap loader) to make it stick infrequently-used nucleus modules into slow memory, while keeping high-usage routines in regular (fast) memory. Joe Morris ###### From: optcamel@ix.netcom.com (Howard and Kelly Lute) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 24 Jun 2000 08:28:48 -0700 Organization: Optimum Camel Companies, Ltd. Lines: 53 Message-ID: References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> NNTP-Posting-Host: cf.df.a8.7d X-Server-Date: 24 Jun 2000 15:37:56 GMT X-Newsreader: MicroPlanet Gravity v2.12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!firehose.mindspring.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58491 In article <8j2i55$f1a$1@top.mitre.org>, jcmorris@jmorris-pc.MITRE.ORG says... > glass2@glass2.lexington.ibm.com writes: > > [extracted:] > >360 MODELS > ># Core Size 25 30 40 50 65 75 85 91-95 44 67 20 > >G 128K..............X...X...X.....................X......... > >H 256K..............X...X...X...X.................X..X...... > >I 512K..................X...X...X...X................X...... > >J 1024K..................X...X...X...X.....X..........X...... > > I don't recall ever hearing of a mod 65 with 128 KB of memory. On > that model you could attach 1, 2, 3, or 4 boxes of 256 KB (IIRC the > memory boxes were machine type 2365). > > Note that a 768 KB machine (or any S/360 model) would be called a > model IH; in general, letters represented power-of-two memory > size, and they were combined when the machine had a combination > that was not a power of 2. > > >K 2048K..............................X.....X..........X...... > > Some models with 2 MB memory were documented as "JJ" instead of K, > but I'm not sure why. > > > >2361 1024K..................X...X...X........................... > >2361 2048K..................X...X...X........................... > > These aren't S/360 models; they are separate outboard memory boxes > known as "Large Core Storage", or LCS. They were (comparatively) > inexpensive because while the 2365 memory had a cycle time of 600 nsec > or so, the LCS box cycle time was 8 usec. > > We used LCS on our model 65; one of the local mods I wrote was to > modify IPLTXT (the bootstrap loader) to make it stick infrequently-used > nucleus modules into slow memory, while keeping high-usage routines > in regular (fast) memory. > > Joe Morris > Smallest I ever saw was a Mod65 with 512...usually they were full up with a meg. -- friend, --------------------- Howard and Kelly Lute He: Electronics Instructor, Terrible Mechanic, Worse Plumber She: Patient eBay: kell EBay Rating: 660 LPFM Site: http://pw1.netcom.com/~optcamel/fmradio.htm Try this site BEFORE you buy -->http://www.epinions.com ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 68 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 24 Jun 2000 15:37:27 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961861047 209.63.28.138 (Sat, 24 Jun 2000 09:37:27 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 09:37:27 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58486 kragen@dnaco.net (Kragen Sitaker) writes: > So now my question is: how many users could a 1970s mainframe > reasonably support with TSO? More than 200? How about a 1980s > mainframe? in the case of something like CICS & IMS (light-weight sub-monitor) it was pssible to have something like 30,000-50,000 connected terminal users in the early '80s. This put strains on other parts of the system, like if system ever crashed & rebooted (say a major power outage). The SNA session setup code for terminal connection was a large monster. Having 30,000 people/terminals trying to connect simultaneously would drive SNA session setup into page thrashing on most machines. I think at the 20,000 terminal number the (simultaneou reconnect) elapsed time was out around 90 minutes (for really large machines, with really large real memories ... at least for mid '80s). this turned out to a problem for IMS hot-standby since even tho the dbms was replicated in case of failure & was instantly took-over, terminal session re-establishment would take (relatively) forever in large configurations. misc. refs: http://www.garlic.com/~lynn/95.html#13 http://www.garlic.com/~lynn/98.html#35a http://www.garlic.com/~lynn/98.html#40 http://www.garlic.com/~lynn/99.html#71 part of the following refs was targeted at maintain session state replication that was critical for infrastructures like ims hot-standby http://www.garlic.com/~lynn/99.html#67 http://www.garlic.com/~lynn/99.html#70 in the case of VM/CMS, CMS users tend to be pretty heavy weight since they are both an address space & thread (as well as the type of work performed). there is the 68/83 comparision that I've posted here several times, from: http://www.garlic.com/~lynn/93.html#31 which basically showed that between 68 & 83 the processer and memory capacity increased by factor of 50-60 times ... but the number of concurrent CMS users typically only increased by a factor of four times. the basic conclusion was that typical disk farm thruput between 68 & 83 only increased by a factor of four ... which was the correlation factor with the number of concurrent users. It was necessary to start changing disk i/o paradigms in order compensate for the relatively different rates of change in the different technologies. other large complex vm/cms configurations: http://www.garlic.com/~lynn/2000c.html#30 http://www.garlic.com/~lynn/2000c.html#31 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 24 Jun 2000 15:51:57 GMT Organization: teabag Message-ID: <8j2let$v7u$1@teabag.demon.co.uk> References: <8iun3f$3p1@netaxs.com> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 961863491 nnrp-13:16973 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 24 Jun 2000 15:51:57 GMT X-Newsreader: knews 1.0b.0 Lines: 14 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!newscore.gigabell.net!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58484 In article , Anne & Lynn Wheeler writes: > in the case of VM/CMS, CMS users tend to be pretty heavy weight since > they are both an address space & thread (as well as the type of work > performed). Hmm, I still wonder why Philips went from Vista to Profs for their email/diary/noticeboard system... Vista was a nice low-load package that ran under MVS, Profs was a high-load package that needed a separate VM for every user; IMHO Vista was also the better system, being much more tightly integrated, easier to use and much, much better with system resources. Chris. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 24 Jun 2000 10:54:55 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: <3vt8lskh5u5gd3vaqav3d7827d7g9sc12s@4ax.com> References: <394F3A96.534E7003@trailing-edge.com> <8itet5$2ccm$1@news.enteract.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.138.150 X-Original-NNTP-Posting-Host: 207.148.138.150 X-Trace: 24 Jun 2000 10:54:56 -0700, 207.148.138.150 Lines: 27 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news-feed.fnsi.net!news.idt.net!nntp.cadvision.com!news.cadvision.com!207.148.138.150 Xref: chonsp.franklin.ch alt.folklore.computers:58493 On 22 Jun 2000 16:29:25 GMT, David Scheidt wrote: >Brian Inglis wrote: > >: The one distinguishing characteristic I have noticed on mainframe >: systems is the presence of multiple I/O channels to which any I/O >: interface could be attached. As compared to the minicomputer CPU >: and I/O interfaces on the backplane, or the microcomputer CPU and >: interfaces on the motherboard. Perhaps this might be a better >: taxonomy. My MF experience was IBM but I have seen comments >: implying similar architectures on DEC-10s, Univac 1100s, HIS/GE, >: etc. Anyone care to compare/contrast across MF manufacturer >: lines? > >Of course, this is becomming less true. There are now Intel based PCs with >multiple PCI buses attached to the bus nexus. Big minis have had this for a >long time, too. How many dozen interfaces can you run simultaneously on each system? If you have a fixed number of interfaces, it ain't no mainframe. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 24 Jun 2000 10:54:57 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <394F3A96.534E7003@trailing-edge.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.138.150 X-Original-NNTP-Posting-Host: 207.148.138.150 X-Trace: 24 Jun 2000 10:54:58 -0700, 207.148.138.150 Lines: 30 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!nntp.cadvision.com!news.cadvision.com!207.148.138.150 Xref: chonsp.franklin.ch alt.folklore.computers:58474 On Fri, 23 Jun 2000 01:09:11 GMT, kragen@dnaco.net (Kragen Sitaker) wrote: >In article , >Brian Inglis wrote: >>The one distinguishing characteristic I have noticed on mainframe >>systems is the presence of multiple I/O channels to which any I/O >>interface could be attached. As compared to the minicomputer CPU >>and I/O interfaces on the backplane, or the microcomputer CPU and >>interfaces on the motherboard. Perhaps this might be a better >>taxonomy. > >So the SGI Octanes and Visual Workstations, with their crossbar I/O >architecture, are mainframes? How about a Beowulf with an Ethernet >switch as a backplane? How many dozen simultaneous transfers can you have at once, and what's the aggregate thruput in GB/s? I should perhaps have emphasized the *ANY I/O interface* and note that multiple channels could be connected to high speed interfaces for increased performance and reliability. Can your system be configured to handle the data from all transponders on a satellite at once? In both directions? Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <8j2let$v7u$1@teabag.demon.co.uk> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 59 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 24 Jun 2000 17:51:24 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961869084 209.63.28.138 (Sat, 24 Jun 2000 11:51:24 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 11:51:24 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58487 cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) writes: > Hmm, I still wonder why Philips went from Vista to Profs for their > email/diary/noticeboard system... Vista was a nice low-load package > that ran under MVS, Profs was a high-load package that needed a > separate VM for every user; IMHO Vista was also the better system, > being much more tightly integrated, easier to use and much, much > better with system resources. > > Chris. the core of PROFS was VMSG ... with a lot of stuff ... like the PFKEY converntion and misc. other stuff that didn't perform very well ... wrapped around it). the source code that the PROFS group used for email was VMSG version 0.9 (effectively a test version before general release ... although in this case general release was internal corporate ... not for "customers" ... but since the internal network at the time was larger than the arpanet/internet ... still a reasonably large population ... and it was nearly all VM/CMS). there was actually a dispute with the PROFS group where they picked up the assembler source code ... claiming it wasn't VMSG. However, it turned out that the primary author of VMSG had placed his initials in an administrative field on every piece of email handled. His initials could be found in every email originated by VMSG and every email originated by PROFS (including those in customer shops anywhere in the world). My small contribution to VMSG was a little of the CC & mailing list code; at one point I kept a small nickname file with 25,000 or so names ... before we did the corporate telephone book stuff (there was an incident where i accidently invoked the whole nickname file). Another thing that was done was the internal telephone book support. There had been a project with dedicated data processing center and 40-50 people that would provide an "internal telephone book" client/server support function (annual budget of something like $5m/annum). A couple of us designed an infrastructure that had a goal of being able to respond to a telephone book lookup in .25 secs elapsed (with a employee population of several hundred thousand) ... and that it would take less than 1/2 time person for keeping the data current ... and no dedicated data processing services. This function was also made available in PROFS (PROFS group wrappered PROFS around several feature/functions ... not just VMSG/email). with regard to assembler code & VM/CMS thread over in comp.arch ... VMSG was all 360 assembler while the phone book stuff was mostly PLS. random refs: http://www.garlic.com/~lynn/2000c.html#40 http://www.garlic.com/~lynn/2000c.html#41 http://www.garlic.com/~lynn/2000c.html#42 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 26 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 24 Jun 2000 19:41:13 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961875673 209.63.28.138 (Sat, 24 Jun 2000 13:41:13 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 13:41:13 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!logbridge.uoregon.edu!cyclone-west.rr.com!news.rr.com|news-west.rr.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58528 Anne & Lynn Wheeler writes: > part of the following refs was targeted at maintain session state > replication that was critical for infrastructures like ims hot-standby > > http://www.garlic.com/~lynn/99.html#67 > http://www.garlic.com/~lynn/99.html#70 in the above reference with 60,000 "logged" in terminals and 2700 terminal messages per second (early '80s actual mainframe configurations) ... there was second order information gathered from real-live environment to feed into the HONE 3705 configurator ... effectively half the terminal 2700 msgs/sec were input that avg. about 125 bytes per message and half the terminal 2700 msgs/sec were output that avg. about 500 bytes per message. configurations today would possibly be expected to handle thruput well over an order of magnitude larger. misc. HONE configurator ref: http://www.garlic.com/~lynn/2000c.html#30 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: "GerardS" Newsgroups: alt.folklore.computers References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 14 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Message-ID: Date: Sat, 24 Jun 2000 15:43:26 -0500 NNTP-Posting-Host: 208.149.16.231 X-Trace: newsfeed.slurp.net 961878862 208.149.16.231 (Sat, 24 Jun 2000 15:34:22 CDT) NNTP-Posting-Date: Sat, 24 Jun 2000 15:34:22 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.slurp.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58525 |"Joe Morris" wrote: |> glass2@glass2.lexington.ibm.com writes: ---[snipped--- | >K 2048K..............................X.....X..........X...... | Some models with 2 MB memory were documented as "JJ" instead of K, | but I'm not sure why. ---[more snipping]--- _________________________ Perhaps a number (say, 2065) with a suffixed K would be confused with 2065K (kilo-whatever)... Gerard S. ###### Newsgroups: alt.folklore.computers From: jsaum@world.std.com (Jim Saum) Subject: Re: Does the word "mainframe" still have a meaning? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Sat, 24 Jun 2000 20:45:22 GMT References: <8iun3f$3p1@netaxs.com> <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net> NNTP-Posting-Host: ppp0b185.std.com Organization: Software Tool & Die X-Newsreader: MT-NewsWatcher 2.4.4 Lines: 30 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!logbridge.uoregon.edu!news.umass.edu!world!jsaum Xref: chonsp.franklin.ch alt.folklore.computers:58500 In article <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net>, jmaynard@conmicro.cx wrote: > ... I don't know when you could use PL/I or COBOL for CICS, >but the command-level interface wasn't available until 1978 or so (with CICS >1.3). ... CICS/OS (and, IIRC, CICS/DOS) supported PL/I and COBOL macro-level programming in 1972 and possibly a little before then. (The PL/I Optimizing Compiler came out in 1971.) Since macro-level programming was full of pointers and direct access to CICS's control blocks, even in COBOL or PL/I it didn't feel very high-level. In many ways macro-level CICS programming in S/360-370 assembler was cleaner and more readable; as a bonus, assembler programs didn't have to go through a preprocessor prior to compilation. Also, in the early 70's there were still a lot of 360s and small 370s with very tight memory environments, where the fixed costs of COBOL or PL/I libraries were unacceptable. In the early 70's, most CICS programming was done in assembler. By the end of the 1970's, most new CICS programming was done in COBOL. When CICS dropped support for macro-level apps in the early 90's, 3rd-party software suppliers began offering emulators to allow legacy macro-level apps to run in a command-level-only environment. Such tools are still being advertised in Enterprise Systems Journal, so I guess some shops are still running some macro-level CICS apps. - Jim Saum ###### Newsgroups: alt.folklore.computers From: Terry Kennedy Subject: Re: Does the word "mainframe" still have a meaning? X-Complaints-To: Email abuse@spc.edu if this posting is inappropriate User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (BSD/OS/4.1 (i386)) NNTP-Posting-Host: gate.tmk.com Organization: St. Peter's College, US Message-ID: References: <394F3A96.534E7003@trailing-edge.com> X-Trace: spcuna.spc.edu 961879525 29703 terry [204.141.35.61] Date: Sat, 24 Jun 2000 20:45:26 GMT Lines: 12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!europa.netcrusader.net!4.1.16.34!cpk-news-hub1.bbnplanet.com!crtntx1-snh1.gtei.net!news.gtei.net!dfw-peer.news.verio.net!ord-feed.news.verio.net!news.verio.net!news.spc.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58505 Brian Inglis writes: > How many dozen simultaneous transfers can you have at once, and > what's the aggregate thruput in GB/s? Was my 370/138 a mainframe? Old-style IBM channels were 3MB/sec max, as I recall. ESCON has sped that up some. In reference to another post - is a Sun connected to a SAN with 6TB of disk a mainframe? Terry Kennedy http://www.tmk.com terry@tmk.com Jersey City, NJ USA ###### Newsgroups: alt.folklore.computers From: jsaum@world.std.com (Jim Saum) Subject: Re: Does the word "mainframe" still have a meaning? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Sat, 24 Jun 2000 21:09:29 GMT References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> NNTP-Posting-Host: ppp0b185.std.com Organization: Software Tool & Die X-Newsreader: MT-NewsWatcher 2.4.4 Lines: 34 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!152.163.239.131!portc03.blue.aol.com!peerfeed.news.psi.net!uunet!ffx.uu.net!world!jsaum Xref: chonsp.franklin.ch alt.folklore.computers:58502 In article <8j2i55$f1a$1@top.mitre.org>, jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) wrote: >These aren't S/360 models; they are separate outboard memory boxes >known as "Large Core Storage", or LCS. They were (comparatively) >inexpensive because while the 2365 memory had a cycle time of 600 nsec >or so, the LCS box cycle time was 8 usec. > >We used LCS on our model 65; one of the local mods I wrote was to >modify IPLTXT (the bootstrap loader) to make it stick infrequently-used >nucleus modules into slow memory, while keeping high-usage routines >in regular (fast) memory. At least one non-IBM vendor (Ampex) also made LCS boxes for 360's. Since even the largest (mod 75) of the original five 360 models had a standard core size of only 1MB, it was only when installations started adding LCS boxes to their systems that any 360's actually had 16MB, which was all the 24-bit address space allowed. (Such configurations weren't possible at all on the small 360 models, which had fewer than 24 address bits; e.g., the 360/40 had 18.) Around 1970 in the bug lists from IBM was a report saying one of the operating systems (DOS?) was unable to IPL ("Initial Program Load" = mainframe-ese for "boot") on a 16MB machine. The OS would loop during IPL clearing memory, relying on the addressing exception interrupt (like a bus error in some desktop systems) to exit the loop and determine how much memory was present. But the 360 architecture specified that if and only if there was exactly 16MB of memory installed, the memory addressing should wrap around to zero instead of causing an addressing exception, so in this one case the technique failed. - Jim Saum ###### From: "GerardS" Newsgroups: alt.folklore.computers References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 37 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Message-ID: <8t955.1097$E82.4485@newsfeed.slurp.net> Date: Sat, 24 Jun 2000 16:19:35 -0500 NNTP-Posting-Host: 208.149.16.139 X-Trace: newsfeed.slurp.net 961881028 208.149.16.139 (Sat, 24 Jun 2000 16:10:28 CDT) NNTP-Posting-Date: Sat, 24 Jun 2000 16:10:28 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.slurp.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58523 |"Jim Saum" wrote: |> (Joe Morris) wrote: |>These aren't S/360 models; they are separate outboard memory boxes |>known as "Large Core Storage", or LCS. They were (comparatively) |>inexpensive because while the 2365 memory had a cycle time of 600 nsec |>or so, the LCS box cycle time was 8 usec. |>We used LCS on our model 65; one of the local mods I wrote was to |>modify IPLTXT (the bootstrap loader) to make it stick infrequently-used |>nucleus modules into slow memory, while keeping high-usage routines |>in regular (fast) memory. | At least one non-IBM vendor (Ampex) also made LCS boxes for 360's. | | Since even the largest (mod 75) of the original five 360 models had a | standard core size of only 1MB, it was only when installations started | adding LCS boxes to their systems that any 360's actually had 16MB, | which was all the 24-bit address space allowed. (Such configurations | weren't possible at all on the small 360 models, which had fewer than | 24 address bits; e.g., the 360/40 had 18.) | | Around 1970 in the bug lists from IBM was a report saying one of the | operating systems (DOS?) was unable to IPL ("Initial Program Load" = | mainframe-ese for "boot") on a 16MB machine. The OS would loop during | IPL clearing memory, relying on the addressing exception interrupt | (like a bus error in some desktop systems) to exit the loop and | determine how much memory was present. But the 360 architecture | specified that if and only if there was exactly 16MB of memory | installed, the memory addressing should wrap around to zero instead of | causing an addressing exception, so in this one case the technique | failed. | - Jim Saum _____________________________ Yes, this is true. When VM (CP/67) first came out, some (if not most) operating systems couldn't handle 16 Meg of "real" memory, but of course, under CP, you just gave the guest 16Meg - 4K. Gerard S. ###### Newsgroups: alt.folklore.computers From: jsaum@world.std.com (Jim Saum) Subject: Re: Does the word "mainframe" still have a meaning? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Sat, 24 Jun 2000 21:36:17 GMT References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> <8t955.1097$E82.4485@newsfeed.slurp.net> NNTP-Posting-Host: ppp0b185.std.com Organization: Software Tool & Die X-Newsreader: MT-NewsWatcher 2.4.4 Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.icl.net!join.news.pipex.net!pipex!tank.news.pipex.net!pipex!ams.uu.net!ffx.uu.net!world!jsaum Xref: chonsp.franklin.ch alt.folklore.computers:58501 In article <8t955.1097$E82.4485@newsfeed.slurp.net>, "GerardS" wrote: >Yes, this is true. When VM (CP/67) first came out, some (if not most) >operating systems couldn't handle 16 Meg of "real" memory, but >of course, under CP, you just gave the guest 16Meg - 4K. Which makes me wonder. I know that by the mid-70's IBM developers were using VM/370 (introduced in 1972) a lot for testing their target OS's, but CP/67 was available well before that, so why didn't these 16MB problems get fixed earlier? Because CP/67 was an unofficial (as opposed to TSS), below-the-radar thing with all the hard feelings and politics that contrast engendered, or perhaps because the developer sites didn't have mod 67's to run it? - Jim Saum ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> <8t955.1097$E82.4485@newsfeed.slurp.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 63 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 24 Jun 2000 21:46:31 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961883191 209.63.28.138 (Sat, 24 Jun 2000 15:46:31 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 15:46:31 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58521 "GerardS" writes: > Yes, this is true. When VM (CP/67) first came out, some (if not most) > operating systems couldn't handle 16 Meg of "real" memory, but > of course, under CP, you just gave the guest 16Meg - 4K. > > Gerard S. a slightly different problem was somebody had originally modified the VM/370 boot image setup routine to use MVCL (aka move character long) with length of 16mbytes (as part of the conversion from CP/67 to VM/370). The MVCL insturction could clear memory and give back a residual length equal to 16mbyte minus actual memory (& the starting address of the field). multiple byte 360 instructions and the majority of the 370 instructions were defined as testing the starting and ending address boundaries and not executing or nullifying the instruction if any of the address portions violated something (memory not available, protected, page fault, etc ... also instruction was interruptable for i/o and other types of interrupts ... storing the updated values for the portion of the operation not yet performance). MVCL (& CLCL, aka compare logical character long) 370 instuction was defined to incrementally perform the operation and update address pointers and residual length of unexecuted portion. MVCL also allowed definition of padding character if target length was longer than origin length. In any case, 370/125 (and 370/115) as originally shipped to customers mis-implemented the MVCL (& CLCL) instructions ... prechecking the ending address (start+length) and not executing the instruction at all if there was a problem (like exceeding available real storage). This resulted in a failure of the boot image setup routine (i.e. VM/370 could be booted on a 125 machine if the boot image had been built on a non-125 machine ... it just wasn't possible to build on a kernel on a 370/125 ... until they shipped a hardware fix for the bug). Possibly the largest used application of 16mbyte virtual address space under CP/67 was numerous APL applications (after cambridge had ported APL/360 to CMS/APL). All APL/360 installations that I was aware of (at the time) limited APL workspaces to at most 64kbytes ... and commonly 32kbytes. With CMS/APL there was something of an explosion in financial modeling applications, performance modeling applications, configuration modeling applications, etc (and to some extent gave rise to some of the origins of data processing capacity planning). This gave rise to large use of CMS/APL by corporate financial planning people, product planning people, product forcasting people and competitive analysis people (filling at least the niche served by a lot of spreadsheet applications that are seen today). It also formed the basis for the majority of the applications delivered to field/customer support people on HONE. misc ref: http://www.garlic.com/~lynn/2000c.html#30 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> <8t955.1097$E82.4485@newsfeed.slurp.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 43 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 24 Jun 2000 22:09:48 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961884588 209.63.28.138 (Sat, 24 Jun 2000 16:09:48 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 16:09:48 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58532 jsaum@world.std.com (Jim Saum) writes: > Which makes me wonder. I know that by the mid-70's IBM developers were > using VM/370 (introduced in 1972) a lot for testing their target OS's, > but CP/67 was available well before that, so why didn't these 16MB > problems get fixed earlier? Because CP/67 was an unofficial (as > opposed to TSS), below-the-radar thing with all the hard feelings and > politics that contrast engendered, or perhaps because the developer > sites didn't have mod 67's to run it? while CP/67 was used extensively for product development (program edit, compile, etc) and simple regression testing of the operating system; very few people thot of using it for memory-size regression testing (& they didn't really believe their target customers would be doing a lot of production work under CP/67). Regression testing that was done was with typically sized (virtual) memory configurations expected at customer locations (not running CP). The VM/370 work for ECPS and vs/1 handshaking (basically originally targetted for 370/138 & 370/148 market place; individual installations had done earlier "CP sensitivity" work on MVT & VS/1, but it wasn't being shipped in products) was one of the first major efforts for production operation of operating systems under CP (and could also be considered an precursor to existing mainframe LPAR support). random refs: http://www.garlic.com/~lynn/94.html#21 http://www.garlic.com/~lynn/94.html#27 http://www.garlic.com/~lynn/94.html#28 http://www.garlic.com/~lynn/94.html#35 http://www.garlic.com/~lynn/97.html#26 http://www.garlic.com/~lynn/99.html#74 http://www.garlic.com/~lynn/2000.html#8 http://www.garlic.com/~lynn/2000.html#63 http://www.garlic.com/~lynn/2000.html#86 http://www.garlic.com/~lynn/2000b.html#50 http://www.garlic.com/~lynn/2000b.html#52 http://www.garlic.com/~lynn/2000b.html#61 http://www.garlic.com/~lynn/2000b.html#62 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 24 Jun 2000 23:48:20 GMT Organization: Neosoft (using Airnews.net!) Lines: 16 Message-ID: X-Orig-Message-ID: References: <8iun3f$3p1@netaxs.com> <8ivlon$1adm$1@ausnews.austin.ibm.com> <8j2i55$f1a$1@top.mitre.org> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sat Jun 24 18:48:20 2000 NNTP-Posting-Host: !dM$s1k-Vk*"T4p (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58511 On Sat, 24 Jun 2000 21:09:29 GMT, Jim Saum wrote: >Around 1970 in the bug lists from IBM was a report saying one of the >operating systems (DOS?) was unable to IPL ("Initial Program Load" = >mainframe-ese for "boot") on a 16MB machine. The OS would loop during >IPL clearing memory, relying on the addressing exception interrupt >(like a bus error in some desktop systems) to exit the loop and >determine how much memory was present. But the 360 architecture >specified that if and only if there was exactly 16MB of memory >installed, the memory addressing should wrap around to zero instead of >causing an addressing exception, so in this one case the technique >failed. This was true for OS/360 21.8...I found it out the hard way, playing with Hercules. The copy of 21.8 I have also won't run TSO with more than 8 MB of real memory present; TCAS dies horribly. Do you know if anyone out there had run TSO in larger systems, and how if so? ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 00:54:27 GMT Organization: The PACSIBM SIG BBS Lines: 11 Message-ID: <8j3l83$6mp@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsfeed.fast.net!yellow.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58539 > CICS ran (and runs) in one partition with a single memory map, so a bug > in the program running on one terminal can crash the programs on the > other terminals, right? No. CICS isolates each program and terminal. If there is a bug in a program, that particular program will fail, but other users will not be affected. The user executing the program will see the program abend, but NOT his CICS session. ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 01:03:28 GMT Organization: The PACSIBM SIG BBS Lines: 38 Message-ID: <8j3lp0$84l@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!fu-berlin.de!news.maxwell.syr.edu!news-xfer.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58538 > So now my question is: how many users could a 1970s mainframe > reasonably support with TSO? More than 200? How about a 1980s > mainframe? In the 1970s we were supporting 1,000 terminals on our machine. It really depends on the model, memory, peripherals, and workload. The S/360-S/370 line ran the gamut from small to powerful processors. The amount of real memory would vary performance. The kind of disks-- older slow ones or newer fast ones would affect performance. A system where users were continuously banging away at their terminal, doing transactions that had complex database updates would support fewer users than one with occassional terminal usage and simple inquiries. In the 1970s TSO was a resource hog. Many users competiting on-line editors, such as ADR's ROSCOE, for programmer development. > My original assertion was that 1970s and 1980s mainframes were able to > support hundreds or thousands of users by using a single job that > served many users. From the terminal user's point of view, whether they were supported by their own dedicated job or were one of many sharing a single job, didn't matter. Different online applications worked differently. Many places had multiple CICS regions to spread the load out. Systems in those days did require serious tuning effort to allocate real vs. virtual memory, and other CPU and DASD resources and task switching priorities. There were a great many tradeoffs and configuration options which depended on the nature of the workload being put through the machine. ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 01:07:30 GMT Organization: The PACSIBM SIG BBS Lines: 5 Message-ID: <8j3m0i$8t9@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news-peer.gip.net!news.gsl.net!gip.net!feeder.qis.net!yellow.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58540 What is "Vista"? I still use PROFS as my email system at work, even though most people now use their PC "Groupwise". I find PROFS (officevision) far more reliable and easier to use. ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 02:08:46 GMT Organization: Neosoft (using Airnews.net!) Lines: 12 Message-ID: <55E2821F92DE89ED.35495102DD55329E.979A24FB048DF463@lp.airnews.net> X-Orig-Message-ID: References: <8j3l83$6mp@netaxs.com> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sat Jun 24 21:08:46 2000 NNTP-Posting-Host: !bt^\1k-Xg%dCY_ (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!209.50.235.254!europa.netcrusader.net!205.252.116.205!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal1.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58508 On 25 Jun 2000 00:54:27 GMT, lwin wrote: >CICS isolates each program and terminal. If there is a bug in a program, >that particular program will fail, but other users will not be affected. >The user executing the program will see the program abend, but NOT his >CICS session. Not in the sense a Unix person is thinking, it doesn't. CICS programs in a region execute in the same address space, with no hardware memory protection from others. CICS is good about blowing out just the failing program, but it is still possible for a CICS application to take the whole region down. (Well, unless subsystem storage protection is in use; even then, a CICS application can trash another application in the same region.) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 102 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 24 Jun 2000 22:20:33 EDT Date: Sun, 25 Jun 2000 02:20:33 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58541 In article , Brian Inglis wrote: >On Fri, 23 Jun 2000 01:09:11 GMT, kragen@dnaco.net (Kragen >Sitaker) wrote: >>In article , >>Brian Inglis wrote: >>>systems is the presence of multiple I/O channels to which any I/O >>>interface could be attached. >> >>So the SGI Octanes and Visual Workstations, with their crossbar I/O >>architecture, are mainframes? How about a Beowulf with an Ethernet >>switch as a backplane? > >How many dozen simultaneous transfers can you have at once, and >what's the aggregate thruput in GB/s? On the SGI machines, I think only a few --- fewer than ten. (Origin 2000s are another story, of course.) Fast Ethernet switches are up in the several-dozen range (three dozen simultaneous transfers at 100 megabits on a 36-port switch, some of which are stackable so you can scale up to two or three switches) and can typically handle full-duplex communication on all ports at once, so you have a total aggregate throughput of 3.6 gigabits on a 36-port switch. Some people build Beowulfs with Gigabit Ethernet or Myrinet for a backplane; Myrinet scales better, so you can get many tens of 1.28Gbps ports on a fabric. The jargon is "network" and "bandwidth" rather than "backplane" and "throughput", but that's what it is. >I should perhaps have emphasized the *ANY I/O interface* and note >that multiple channels could be connected to high speed >interfaces for increased performance and reliability. Yes, people do that in the Beowulf world, too; the first Beowulf had it, in fact. They call it "channel bonding". They usually only use up to two channels per node. You might protest that each node is actually a whole computer, not just a disk or tape drive, but that's true of S/390 machines' peripherals too these days :) >Can your system be configured to handle the data from all >transponders on a satellite at once? In both directions? Of course, but it depends on the satellite. But you shouldn't say "your system"; the only Beowulf I myself own is a currently non-operational little four-node thing I bought for $2500. (It's non-operational because I haven't gotten around to installing software on it.) To be concrete, Landsat 7 has a 150-megabit downlink to ground stations (plus some piddly little extra downlinks and an uplink that, added together, aren't significant compared to the 150-megabitter.) 150 megabits is 18.75 megabytes per second. The nice new $175 IBM hard drives I bought for the Beo claim they can do 22 megabytes per second on a sustained basis. My experience is that I typically get about 20% of my nominal disk bandwidth (on non-Beowulf systems; maybe I could fix this); this suggests I could handle this aggregate transfer rate with five such disks, which means five nodes. Except that the disks themselves are only 20GB. So my five disks will only last me 5300 seconds' worth of Landsat data, or a bit less than an hour and a half. I'd need to be shuffling the data off to tape or some other inexpensive medium rather frequently. So I actually need twice that aggregate bandwidth (unless I shuffle the data off to tape as soon as I get it in RAM, in which case I don't need disks at all). So I need ten nodes, costing roughly $6500. So I estimate I'd need a ten-node Beowulf to set up a Landsat ground station. The tape drives I'd be writing data onto would probably cost considerably more than the computer and network hardware; ten DLT-7000s, which can each theoretically archive about 5 megabytes per second (40 megabits), would cost me $30,000 or so. Considerably cheaper than a DCR-SI drive, which IIRC is what NASA uses for this purpose, and much cheaper per MBps than any other archival storage media I know of, but still more expensive than the computer itself. Probably cheaper than the non-computational downlink hardware, still. (A significant amount of processing has to happen on all the data, but it is easily parallelizable and it can easily be done at more than 1.9 megabytes per second per machine, which is what each node would have to do on average in this ten-node situation.) (I understand that, in the past, satellite telemetry data was often recorded on analog tapes in real-time, then played back at slower speeds for computer processing.) Actually, though, the tapes cost about fifty cents a gigabyte, and I'd be handling a gigabyte every minute or so that the bird was overhead. I suspect that the costs from buying and storing the tapes would make the computer hardware costs sink into insignificance. What kind of satellite are you thinking of? -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: Erno Palonheimo Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 05:35:18 +0300 Organization: Helsinki University of Technology, Finland Lines: 33 Message-ID: References: NNTP-Posting-Host: vipunen.hut.fi X-Trace: nntp.hut.fi 961900519 5858 130.233.249.7 (25 Jun 2000 02:35:19 GMT) X-Complaints-To: usenet@nntp.hut.fi NNTP-Posting-Date: 25 Jun 2000 02:35:19 GMT X-Newsreader: Gnus v5.3/Emacs 19.33 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed1.swip.net!swipnet!newsfeed2.funet.fi!newsfeeds.funet.fi!nntp.hut.fi!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58522 kragen@dnaco.net (Kragen Sitaker) writes: > Some people build Beowulfs with Gigabit Ethernet or Myrinet for a > backplane; Myrinet scales better, so you can get many tens of 1.28Gbps > ports on a fabric. > > The jargon is "network" and "bandwidth" rather than "backplane" and > "throughput", but that's what it is. It's a network, it's not a backplane. Get a recent large multi-processor Alphaserver, there's a backplane with more speed than gigabit Ethernet can ever achieve. Next? > Yes, people do that in the Beowulf world, too; the first Beowulf had > it, in fact. They call it "channel bonding". They usually only use up > to two channels per node. Beowulf is made up of glue and well-chewed bubble gum. It's not usable in production systems. I'd like to say same about Linux in general, but it has gained quite some commercial support and special hardware vendors, so ok, now it's possible to run it with support contracts on hardware and software, but still it is an ugly resource hog designed by a crew with some witty ones (Linus himself, Alan Cox et al) and a lot of not-so-witty-ones (such as the idiots who decided to write a kernel-level httpd & nfsd). I'm sorry about this sounding a flame, but I have my opinions, and I've heard the critic about two-hundred something times during last seven years. This is just my $0.02. -- --- Erno Palonheimo -- Helsinki University of Technology ----- esp@cc.hut.fi -- Administrative Manouvers in the Dark ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 118 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 24 Jun 2000 23:13:53 EDT Date: Sun, 25 Jun 2000 03:13:53 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58542 In article <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net>, Jay Maynard wrote: >On Sat, 24 Jun 2000 02:56:24 GMT, Kragen Sitaker wrote: >>It definitely answers it; the job/started task/TSO user maps closely to >>a Win32 or Solaris "process", and the tasks map closely to threads. > >Pretty much, though a TCB is a bit more expensive to create than most >threads, and an address space is extremely expensive to create, definitely >unliks a Unix process. Out of curiosity, why was this? >In the typical mainframe shop, programmers, operators, and administrators >(systems types) were the only ones that had TSO. The real work of the >enterprise was done with applications running under CICS or IMS/DC, and >users logged on to that. > >I don't know how many logged-on TSO users a typical 3081/3084-class system >would support...because nobody ever ran them that way. How much memory did each TSO user typically consume? And how fast are switches between address spaces? Those numbers should give an upper estimate. >>My original assertion was that 1970s and 1980s mainframes were able to >>support hundreds or thousands of users by using a single job that >>served many users. > >This is definitely the case, but, OTOH, they also provided the facilities >that made that possible. Unix was never intended to carry that kind of a >load, and so nobody ever wrote a CICS-style TP monitor for it. There are, actually, CICS-style TP monitors for Unixes. I believe Transarc's Encina and BEA's Tuxedo are such creatures. I don't think it requires a heck of a lot of OS support to write a CICS-style TP monitor; you just need to be able to get event notifications for I/O, right? >Because of >that, every user has an address space [on Unix], and every user has >his own amount of overhead, and so supporting thousands of users takes >more resources. Mainframes did what they did because they had to; even >at the end of the 80s, 256 MB was a huge system, but would support tens >of thousands of users without breathing hard. It looks like Apache needs about 200K-300K of physical RAM per Apache child on my machine; this suggests that a 256MB machine could support around 1000 concurrent connections. (I only have 128MB.) If each user submits a request every 15 seconds, and the request takes a full second to complete, then 1000 concurrent connections means 15,000 concurrent users. According to http://www.caucho.com/articles/benchmark.html, on a PII-266, Apache on Linux can serve 300 requests per second with various dynamic page generation techniques, which would add up to about 4500 concurrent users. So "tens of thousands" is quite reasonable (if the users are a little less strenuously interactive than the users postulated above, or if you're using something faster than a PII-266), and the limitation is likely to be something other than memory. >To answer a question in another post, yes, CICS was a single address space. >Not only that, but it ran all applications under a single TCB, and only used >separate TCBs for specific functions (the first, for example, was a separate >TCB to do VSAM I/O). CICS applications have to be reentrant between CICS >system calls, and are only loaded once into memory and shared. CICS >multitasking is cooperative, with a runaway task timer that ABENDs a task >that doesn't issue a CICS call within a configurable length of time >(typically 10 seconds). Thanks for the detailed explanation. I've programmed systems like this (DU MOO --- I'm such an Internet boy); it requires a certain unusual kind of awareness. You have to carefully consider each loop and recursive procedure call to figure out whether it can push you over your time limit, and insert "yield" calls if it can. >As CICS systems became too large to fit in the 8-10 >MB of available address space in MVS/370, shops would break them up into >multiple regions, each in a separate MVS started task, along application >boundaries; So you could have up to 10 MB in each started task, but the machine could physically accomodate much more? That seems weird. Didn't the 370 have 32-bit byte-based addressing? > even that grew too restrictive, and facilities were added to >allow CICS regions to communicate requests for resources to other CICS >regions in the same MVS instance (with cross-memory calls), or on other MVS >instances (via VTAM). An "MVS instance" is MVS running in an LPAR, I take it? (My understanding of most of this mainframe stuff is pretty vague.) >All of this left CICS vulnerable to an application stomping on critical >control blocks; in practice, however, this was rare for shops that used >high-level languages and the CICS command-level interface, both of which hid >the details of control blocks and low-level addressing from the programmer's >ability to screw up. I don't know when you could use PL/I or COBOL for CICS, >but the command-level interface wasn't available until 1978 or so (with CICS >1.3). This doesn't mean it never happened, and the layout of control blocks >in CICS's memory wasn't exactly designed for robustness (critical >task-related control blocks were located next to application work areas in >such a way that overrunning them did Bad Things), and when it did, debugging >could be a real bitch; however, nearly every shop that ran CICS had >production and test regions so applications programmers could do development >and debugging without breaking the critical work of the system. Interesting. I threw that bit in about applications stomping on each other to disambiguate, not to imply it was a problem. It's interesting to hear that it was. I thought most CICS code was in COBOL and therefore didn't need to worry about that kind of problem. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8iun3f$3p1@netaxs.com> <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 33 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 25 Jun 2000 03:37:51 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 961904271 209.63.28.138 (Sat, 24 Jun 2000 21:37:51 MDT) NNTP-Posting-Date: Sat, 24 Jun 2000 21:37:51 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58531 kragen@dnaco.net (Kragen Sitaker) writes: > I don't think it requires a heck of a lot of OS support to write a > CICS-style TP monitor; you just need to be able to get event > notifications for I/O, right? reference earlier post regarding 30,000-60,000 CICS users in early to mid 80s ... well under 100mip processing power ... doing 1000+ transactions per second. as mentioned there was problem with vtam session establishment if system cycled and all users attempted to re-establish their sessions concurrently (i.e. vtam session establishment page thrashing ... 20,000 simultaneous connection requests degrading to 90 minutes elapsed time ... it wasn't a problem with support 30,000-60,000 concurrent sessions ... it was a problem trying to perform 20,000+ simultaneous session establishments). this is somewhat analogous (but different) to http web server problems in the mid-90s ... most TCP implementations had linear finwait lists (i.e. prior to http ... telnet, ftp, etc. had long running sessions ... even some larger configurations with 5,000 concurrent sessions ... number of session terminations in the finwait interval was very small ... so number of items on the finwait list was trivial). tcp has minimum 7 packet exchange ... with dangling finwaits for session termination. typical http might add a couple more packets to a tcp session but web servers doing 50-100 http hits per second sometimes built finwait list with a couple thousand entries and 98% of the available cpu sometimes were lost to running finwait list. does anybody remember when netscape added ftp20.netscape.com, what it was, and why (compared to ftp1.netscape.com thru ftp?.netscape.com)? -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 03:44:52 GMT Organization: Neosoft (using Airnews.net!) Lines: 82 Message-ID: X-Orig-Message-ID: References: <8iun3f$3p1@netaxs.com> <805DBD7F7FA551E7.1F7E0935980F3FC9.077568064983AD11@lp.airnews.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sat Jun 24 22:44:52 2000 NNTP-Posting-Host: ![8(!1k-X?7^;Wm (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!205.252.116.205!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58506 On Sun, 25 Jun 2000 03:13:53 GMT, Kragen Sitaker wrote: >>Pretty much, though a TCB is a bit more expensive to create than most >>threads, and an address space is extremely expensive to create, definitely >>unliks a Unix process. >Out of curiosity, why was this? Creating an address space involves setting up a lot of control blocks, address mapping tables, and so on. The tradeoff was made early on that, since creating address spaces was done much, much less often than other operations, it was acceptable to make that slow in order to make other things fast. >>I don't know how many logged-on TSO users a typical 3081/3084-class system >>would support...because nobody ever ran them that way. >How much memory did each TSO user typically consume? And how fast are >switches between address spaces? Those numbers should give an upper estimate. Each TSO user had an address space of 4 to 8 MB, plus a few hundres K of overhead. The working set (amount of resident storage in use at any given moment) tended to run in the 2 meg range on a lightly loadded system, down to a few hundred K if the system was paging fairly heavily. I can't speak well to the address space switch time; while switching the machine from one address space to another takes loading 16 general purpose registers, some number of control registers (definitely the one that defines the base of the segment table), and the 4 floating point registers, there's more associated overhead. >I don't think it requires a heck of a lot of OS support to write a >CICS-style TP monitor; you just need to be able to get event >notifications for I/O, right? Pretty much. It comes back to the I/O architecture: CICS was built for efficiently handling I/O operations on behalf of transaction programs. >So "tens of thousands" is quite reasonable (if the users are a little >less strenuously interactive than the users postulated above, or if >you're using something faster than a PII-266), and the limitation is >likely to be something other than memory. The last is definitely true. I suspect you'd run out of disk bandwidth first. >Thanks for the detailed explanation. I've programmed systems like this >(DU MOO --- I'm such an Internet boy); it requires a certain unusual >kind of awareness. You have to carefully consider each loop and >recursive procedure call to figure out whether it can push you over >your time limit, and insert "yield" calls if it can. Right, though the design of CICS tended to make that kind of thing rare. A CICS application is usually structured as a series of programs; each program handles one or a small number of input transactions, does something with the data supplied, generates an output screen, and exits. Information that needed to be propagated from one program to the next in a series of transactions was stored in special temporary storage facilities provided for the purpose. >So you could have up to 10 MB in each started task, but the machine >could physically accomodate much more? That seems weird. Didn't the >370 have 32-bit byte-based addressing? Not in the days of MVS/370. The original 360/370 architecture had 24-bit addressing. The 370/XA (eXtended Architecture) had two major changes: 31-bit addressing (the high order bit is used as a mode bit to select 31- or 24-bit addresses), and a major restructuring of the I/O facilities. >An "MVS instance" is MVS running in an LPAR, I take it? (My >understanding of most of this mainframe stuff is pretty vague.) I use that terminology to cover a multitude of sins: LPARs, VM guest systems, different physical machines. There was no way for the average application to tell the difference, and they all used the same facilities to communicate with other MVS instances. >Interesting. I threw that bit in about applications stomping on each >other to disambiguate, not to imply it was a problem. It's interesting >to hear that it was. I thought most CICS code was in COBOL and >therefore didn't need to worry about that kind of problem. By the time CICS 1.4.1 and 1.5 became common, in the 1980 timeframe, most applications were written in COBOL command level, and the problems largely went away. When dealing with macro level programs, however, it was all too common... ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 74 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 24 Jun 2000 23:46:57 EDT Date: Sun, 25 Jun 2000 03:46:57 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58544 In article , Erno Palonheimo wrote: >kragen@dnaco.net (Kragen Sitaker) writes: >> Some people build Beowulfs with Gigabit Ethernet or Myrinet for a >> backplane; Myrinet scales better, so you can get many tens of 1.28Gbps >> ports on a fabric. >> >> The jargon is "network" and "bandwidth" rather than "backplane" and >> "throughput", but that's what it is. > >It's a network, it's not a backplane. Get a recent large >multi-processor Alphaserver, there's a backplane with more speed than >gigabit Ethernet can ever achieve. Next? The technical summary of the 32-processor AlphaServer GS320 (http://www.compaq.com/alphaserver/download/gs_techsummary.pdf) suggests that you are correct; each "quad building block" provides four processors and four memory modules with an aggregate bandwidth of 6.4 gigabytes per second, or 51.2 gigabits per second. A GS320 consists of eight of these building blocks. For what it's worth, it's probably more accurate to call today's Alphaserver and Starfire backplanes networks or crossbars instead of backplanes. :) It's true that there are still companies out there building high-speed interconnects that offer considerably more bandwidth per processor than off-the-shelf GBE parts. I'm not claiming that Beowulfs are preferable for all purposes; I'm claiming that they're well into the realm of 1980s "mainframe" architecture, at least if I/O architecture is the only consideration. (There's also the point that you get an order of magnitude nicer price-to-performance ratio than with traditional supercomputers, and an even bigger advantage over mainframes, but that wasn't what I was talking about.) >> Yes, people do that in the Beowulf world, too; the first Beowulf had >> it, in fact. They call it "channel bonding". They usually only use up >> to two channels per node. > >Beowulf is made up of glue and well-chewed bubble gum. It's not usable >in production systems. This depends on what your system is producing. ;) I'd be upset if my bank or stockbroker told me they did all their OLTP on a Beowulf this year. Next year, the year after --- who can say? But right now, I don't think the manageability is there. What do you mean by "production systems"? > I'd like to say same about Linux in general, >but it has gained quite some commercial support and special hardware >vendors, so ok, now it's possible to run it with support contracts on >hardware and software, but still it is an ugly resource hog "Ugly" is hard to argue with --- I don't agree with it, although I think Plan9 is nicer, but de gustibus non est disputandum. In what sense is it a resource hog? >I'm sorry about this sounding a flame, but I have my opinions, and >I've heard the critic about two-hundred something times during last >seven years. This is just my $0.02. I'm interested to hear your opinions in more detail, although I'm starting to think we should move it to comp.os.linux.advocacy. :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: Erno Palonheimo Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 07:45:48 +0300 Organization: Helsinki University of Technology, Finland Lines: 71 Message-ID: References: NNTP-Posting-Host: vipunen.hut.fi X-Trace: nntp.hut.fi 961908348 7141 130.233.249.7 (25 Jun 2000 04:45:48 GMT) X-Complaints-To: usenet@nntp.hut.fi NNTP-Posting-Date: 25 Jun 2000 04:45:48 GMT X-Newsreader: Gnus v5.3/Emacs 19.33 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!news.algonet.se!algonet!newsfeed1.swip.net!swipnet!newsfeed3.funet.fi!newsfeeds.funet.fi!nntp.hut.fi!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58530 kragen@dnaco.net (Kragen Sitaker) writes: > For what it's worth, it's probably more accurate to call today's > Alphaserver and Starfire backplanes networks or crossbars instead of > backplanes. :) Ok, they are crossbar switches, but still, it's not Ether{net,not}working :-) > It's true that there are still companies out there building high-speed > interconnects that offer considerably more bandwidth per processor than > off-the-shelf GBE parts. I'm not claiming that Beowulfs are preferable > for all purposes; I'm claiming that they're well into the realm of > 1980s "mainframe" architecture, at least if I/O architecture is the > only consideration. They aren't even that because of the lack of the disk bandwidth. And, still, I'd like to see something like the IBM channel architecture on the modern PC platform. Or, actually, I'd like to see that on RS6000 :-) > >Beowulf is made up of glue and well-chewed bubble gum. It's not usable > >in production systems. > > This depends on what your system is producing. ;) Beowulf, or what Beowulf is today, is supposed to be good at scientific calculations. I've seen an institute (not the one I work at as my .signature implies :-) going towards Linux/Beowulf but instead of that they decided to go the Tru64/ASE way. I still dislike Tru64, but even that is more stable, usable and scalable than Linux is today. > I'd be upset if my bank or stockbroker told me they did all their OLTP > on a Beowulf this year. Next year, the year after --- who can say? I'd stay away from businesses using that. Beowulf has quite some good ideas, but they're all implemented The Wrong Way... Sheesh, do people ever learn? Take a look at multiprocessing code of any commercial unix implementation, leave alone the mainframe operating systems, and rethink. > What do you mean by "production systems"? The "production systems" I have to affect in a way or another include one of the largest NNTP servers in *.fi, hut.fi www server, and all the systems of the local university administration (not the CS admin computers, but the ones that are used to pay the salaries etc etc). > In what sense is it a resource hog? Take a look at the VM. At home, I have a MicroVAX II with 13 megabytes of memory. If Linux ran on VAX architecture, it'd be a lot slower than the variant of BSD I currently run because of the brain-damaged kludge-over-kludge-over-kludge-x86-specific optimizations. Linux isn't a multiplatform OS, and I don't want to see it ported over to more architectures before it gets some serious sanity checking done. > I'm interested to hear your opinions in more detail, although I'm > starting to think we should move it to comp.os.linux.advocacy. :) I'm not interested in fighting against 10.000 Linux-smoking glueheads about a matter that I know I've got some more clue than the average Linux advocate. An average Linux advocate hasn't ever ran a Unix-based production system with over 500 simultaneous active users. So, let's keep it here, or move it to alt.flame (if it still exists) :-) I'm sorry about my quite incomprehensible English, but I hope you can cope with it. -- --- Erno Palonheimo -- Helsinki University of Technology ----- esp@cc.hut.fi -- Administrative Manouvers in the Dark ###### From: mbeattie@sable.ox.ac.uk (Malcolm Beattie) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 14:39:18 GMT Organization: Oxford University, England Lines: 32 Message-ID: <8j55im$9db$1@news.ox.ac.uk> References: <394F3A96.534E7003@trailing-edge.com> NNTP-Posting-Host: sable.ox.ac.uk X-Trace: news.ox.ac.uk 961943958 9643 163.1.2.4 (25 Jun 2000 14:39:18 GMT) X-Complaints-To: newsmaster@ox.ac.uk NNTP-Posting-Date: 25 Jun 2000 14:39:18 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!logbridge.uoregon.edu!server3.netnews.ja.net!news.ox.ac.uk!sable.ox.ac.uk!mbeattie Xref: chonsp.franklin.ch alt.folklore.computers:58551 In article , Kragen Sitaker wrote: >In article , >Brian Inglis wrote: >>The one distinguishing characteristic I have noticed on mainframe >>systems is the presence of multiple I/O channels to which any I/O >>interface could be attached. As compared to the minicomputer CPU >>and I/O interfaces on the backplane, or the microcomputer CPU and >>interfaces on the motherboard. Perhaps this might be a better >>taxonomy. > >So the SGI Octanes and Visual Workstations, with their crossbar I/O >architecture, are mainframes? How about a Beowulf with an Ethernet >switch as a backplane? A modern mainframe running VM provides 500 Mbytes/sec of bandwidth between its guests via IUCV which appears to the guests somewhat like a virtual ethernet interface (well, a network interface anyway). That would be, um, pricey to implement with an ethernet-based interconnect. On the I/O side, you can have up to 38 FICON (each 100 Mbytes/sec) channels or 256 ESCON channels (each ~16 Mbytes/sec). To even try to compete with that, you have a much smaller range of non-mainframe systems (e.g. Sun E10K) in the same ballpark. I/O multipathing is another indicator of mainframe-like design and that's more mature for S/390 than for E10K. --Malcolm -- Malcolm Beattie Oxford University Computing Services "I permitted that as a demonstration of futility" --Grey Roger ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 80 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 25 Jun 2000 17:46:47 EDT Date: Sun, 25 Jun 2000 21:46:47 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58604 In article , Erno Palonheimo wrote: >kragen@dnaco.net (Kragen Sitaker) writes: >> It's true that there are still companies out there building high-speed >> interconnects that offer considerably more bandwidth per processor than >> off-the-shelf GBE parts. I'm not claiming that Beowulfs are preferable >> for all purposes; I'm claiming that they're well into the realm of >> 1980s "mainframe" architecture, at least if I/O architecture is the >> only consideration. > >They aren't even that because of the lack of the disk bandwidth. And, >still, I'd like to see something like the IBM channel architecture on >the modern PC platform. Or, actually, I'd like to see that on RS6000 :-) What do you mean by "lack of disk bandwidth"? By the simple expedient of putting a $100 motherboard+CPU, a $40 case, and a $40 FE card on a disk, you can attach it directly to your Ethernet switch; some people do this, in fact. This means that all of your bandwidth can be used as disk bandwidth. What is the IBM channel architecture like? >> I'd be upset if my bank or stockbroker told me they did all their OLTP >> on a Beowulf this year. Next year, the year after --- who can say? > >I'd stay away from businesses using that. Beowulf has quite some good >ideas, but they're all implemented The Wrong Way... Sheesh, do people >ever learn? Take a look at multiprocessing code of any commercial unix >implementation, leave alone the mainframe operating systems, and >rethink. Multiprocessing code of commercial Unix implementations is for big SMPs. Big SMPs have a poor price/performance ratio and don't scale well. [Cray MTA excepted.] (Although ccNUMA is an interesting option; they're integrating ccNUMA support into Linux 2.4.) What ideas are implemented The Wrong Way? It's not as if Beowulf is a single coherent software package; just about every idea that is applicable has multiple competing implementations done in different ways. >> In what sense is it a resource hog? > >Take a look at the VM. At home, I have a MicroVAX II with 13 megabytes >of memory. If Linux ran on VAX architecture, it'd be a lot slower than >the variant of BSD I currently run because of the brain-damaged >kludge-over-kludge-over-kludge-x86-specific optimizations. Can you be specific? I don't know much about Linux's VM. >Linux isn't >a multiplatform OS, and I don't want to see it ported over to more >architectures before it gets some serious sanity checking done. I'm sure that if you rewrote the VM subsystem so it performed better on most platforms, you would find many willing users. There aren't a lot of architectures left for it to be ported to; AS/400, VAX, old mainframes, and various MMUless machines are about it, I think. >I'm not interested in fighting against 10.000 Linux-smoking glueheads >about a matter that I know I've got some more clue than the average >Linux advocate. An average Linux advocate hasn't ever ran a Unix-based >production system with over 500 simultaneous active users. An average Linux advocate is also not in a position to argue about kludges in the virtual memory system. :) >I'm sorry about my quite incomprehensible English, but I hope you can >cope with it. Your English is quite comprehensible. :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 25 Jun 2000 22:17:13 GMT Organization: Neosoft (using Airnews.net!) Lines: 8 Message-ID: <0929C64EC49FCCE8.CC2D5FC333B27421.0F70A1BEACA570F9@lp.airnews.net> X-Orig-Message-ID: References: Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Sun Jun 25 17:17:14 2000 NNTP-Posting-Host: !ci0(1k-YG(=K$g (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.axxsys.net!HSNX.atgi.net!cabal11.airnews.net!news.airnews.net!cabal12.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal1.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58556 On Sun, 25 Jun 2000 21:46:47 GMT, Kragen Sitaker wrote: >There aren't a lot of architectures left for it to be ported to; >AS/400, VAX, old mainframes, and various MMUless machines are about it, >I think. I think his complaint is analogous to the old "You can write FORTRAN in any language if you try hard enough"...in this case, you can implement i386 memory management on any hardware if you try hard enough. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 45 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 25 Jun 2000 18:19:53 EDT Date: Sun, 25 Jun 2000 22:19:53 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58606 In article <8j55im$9db$1@news.ox.ac.uk>, Malcolm Beattie wrote: >In article , >Kragen Sitaker wrote: >>In article , >>Brian Inglis wrote: >>>The one distinguishing characteristic I have noticed on mainframe >>>systems is the presence of multiple I/O channels to which any I/O >>>interface could be attached. As compared to the minicomputer CPU >>>and I/O interfaces on the backplane, or the microcomputer CPU and >>>interfaces on the motherboard. Perhaps this might be a better >>>taxonomy. >> >>So the SGI Octanes and Visual Workstations, with their crossbar I/O >>architecture, are mainframes? How about a Beowulf with an Ethernet >>switch as a backplane? > >A modern mainframe running VM provides 500 Mbytes/sec of bandwidth >between its guests via IUCV which appears to the guests somewhat >like a virtual ethernet interface (well, a network interface anyway). >That would be, um, pricey to implement with an ethernet-based >interconnect. With off-the-shelf CPUs, it's worse than that --- they tend to saturate at well under a gigabit talking to GBE NICs. Four gigabits is out of the question, I think. The interface between CPU and RAM on a PC using PC100 SDRAM is slower than that! > On the I/O side, you can have up to 38 FICON (each >100 Mbytes/sec) channels or 256 ESCON channels (each ~16 Mbytes/sec). >To even try to compete with that, you have a much smaller range of >non-mainframe systems (e.g. Sun E10K) in the same ballpark. I/O >multipathing is another indicator of mainframe-like design and >that's more mature for S/390 than for E10K. So at any given time, you can tell whether something is a mainframe by looking at its I/O capabilities; a real mainframe provides at least a factor of four bandwidth advantage over its contemporary competitors. Is that what you are saying? If so, is there a definition that doesn't depend on comparisons to other contemporary machines? -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <0929C64EC49FCCE8.CC2D5FC333B27421.0F70A1BEACA570F9@lp.airnews.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 19 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 25 Jun 2000 18:23:12 EDT Date: Sun, 25 Jun 2000 22:23:12 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58607 In article <0929C64EC49FCCE8.CC2D5FC333B27421.0F70A1BEACA570F9@lp.airnews.net>, Jay Maynard wrote: >On Sun, 25 Jun 2000 21:46:47 GMT, Kragen Sitaker wrote: >>There aren't a lot of architectures left for it to be ported to; >>AS/400, VAX, old mainframes, and various MMUless machines are about it, >>I think. > >I think his complaint is analogous to the old "You can write FORTRAN in any >language if you try hard enough"...in this case, you can implement i386 >memory management on any hardware if you try hard enough. Right --- I thought he was saying that he didn't want Linux to be be ported to any more architectures because the VM system was so kludgy. I was just pointing out that it's almost too late for that :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: mbeattie@sable.ox.ac.uk (Malcolm Beattie) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 26 Jun 2000 10:34:33 GMT Organization: Oxford University, England Lines: 124 Message-ID: <8j7bjp$6me$1@news.ox.ac.uk> References: NNTP-Posting-Host: sable.ox.ac.uk X-Trace: news.ox.ac.uk 962015673 6862 163.1.2.4 (26 Jun 2000 10:34:33 GMT) X-Complaints-To: newsmaster@ox.ac.uk NNTP-Posting-Date: 26 Jun 2000 10:34:33 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!tank.news.pipex.net!pipex!server1.netnews.ja.net!server2.netnews.ja.net!news.ox.ac.uk!sable.ox.ac.uk!mbeattie Xref: chonsp.franklin.ch alt.folklore.computers:58554 In article , Kragen Sitaker wrote: >In article , >Erno Palonheimo wrote: >>kragen@dnaco.net (Kragen Sitaker) writes: >>> It's true that there are still companies out there building high-speed >>> interconnects that offer considerably more bandwidth per processor than >>> off-the-shelf GBE parts. I'm not claiming that Beowulfs are preferable >>> for all purposes; I'm claiming that they're well into the realm of >>> 1980s "mainframe" architecture, at least if I/O architecture is the >>> only consideration. >> >>They aren't even that because of the lack of the disk bandwidth. And, >>still, I'd like to see something like the IBM channel architecture on >>the modern PC platform. Or, actually, I'd like to see that on RS6000 :-) > >What do you mean by "lack of disk bandwidth"? By the simple expedient >of putting a $100 motherboard+CPU, a $40 case, and a $40 FE card on a >disk, you can attach it directly to your Ethernet switch; some people >do this, in fact. This means that all of your bandwidth can be used as >disk bandwidth. Think of it like this: you have CPU, memory and I/O devices and you have a large variety of potential ways to interconnect them. If your system requirements are specific enough, then you could hope to design your interconnections so that the links which didn't require high bandwidth were cheap and the links which didn't need to be reliable were less robust. In the Beowulf case, you design a system where the software you're going to run on it parallelises along CPU/memory lines with inter-node bandwidth of 10Mbytes/sec sufficient (for fast ethernet). For some things that works fine. Case in point: the Linux IMAP/WING webmail 15-node mail cluster I run here with 750 GB disk. I have nothing against that design where it's suitable. Another way of splitting up a system into smaller parts is where you need not too massive disk bandwidth, where I/O latency (seek time) is almost completely unimportant and where bandwidth to other systems is very low: a typical PC. The problem with both of those (and with any other way of splitting up a system) is that in some cases you can't split it up in such a way that it satisifes all the work you want to do on it. It's at that point that you end up with multiple servers of different designs which is just the start of your maintenance problems. Then over the years, you find that one server actually needs higher bandwidth to another one and that you want to run one more service that needs the big disk space on server foo, the large memory on server bar and the multipathing for robustness that you bought for server baz. There are people around who, instead of going that route, decide to buy a single machine which has sufficient internal bandwidth and is sufficiently extensible in plugging in many, many multipath-able I/O devices that, however there future services split up, the big box can cope. An appropriate motto, already in use in the security community for one design there, is Put all your eggs in one basket...but watch it very, very closely. On the internal bandwidth side, with S/390 each group of 3 CPUs (well, their L2 caches) have their own two 2.6 Gbytes/sec buses. That was last generation (G5), I think, so it may or may not have gone up since. I have a feeling the L2 cache on each CPU is 8 MB but I'm not sure. >>I'd stay away from businesses using that. Beowulf has quite some good >>ideas, but they're all implemented The Wrong Way... Sheesh, do people >>ever learn? Take a look at multiprocessing code of any commercial unix >>implementation, leave alone the mainframe operating systems, and >>rethink. Er, yes, what's your point? Nicely designed and maintainable locking, low latency drivers for commodity hardware that you can buy cheap and pile high. Looks fine to me for a number of design niches. >>Take a look at the VM. At home, I have a MicroVAX II with 13 megabytes >>of memory. If Linux ran on VAX architecture, it'd be a lot slower than >>the variant of BSD I currently run because of the brain-damaged >>kludge-over-kludge-over-kludge-x86-specific optimizations. > >Can you be specific? I don't know much about Linux's VM. I do. Not only do *I* think it's mostly nice and clean (swapping aside but that is due for and is getting a rewrite during 2.5 to a more BSDish design) but at least one of the authors of the FreeBSD VM does too. The Linux VM design is not suitable for the VAX architecture. Putting it the other way around: the VAX architecture is not suitable for Linux because the VAX made design trade-offs that clash with Linux's design trade-offs. Given the current crop of hardware designs that Linux works well with, which includes S/390 so you can't claim that all architectures around when the VAX was designed would not be suitable for Linux. As for x86-specific optimisations: the arch independent tree does not contain x86-specific optimisations of course. Funnily enough, the arch/i386 tree does contain optimisations but that's hardly surprising is it? >>Linux isn't >>a multiplatform OS, and I don't want to see it ported over to more >>architectures before it gets some serious sanity checking done. Just keep your eyes closed then. I really don't care what you say you want to see except insofar as readers of a.f.c may take your unsubstantiated, untrue claims as gospel. >>I'm not interested in fighting against 10.000 Linux-smoking glueheads >>about a matter that I know I've got some more clue than the average >>Linux advocate. But not that much more, from the sound of it. >> An average Linux advocate hasn't ever ran a Unix-based >>production system with over 500 simultaneous active users. I have. I do. I hack the Linux kernel. Care to make any specific claims now? >>I'm sorry about my quite incomprehensible English, but I hope you can >>cope with it. > >Your English is quite comprehensible. :) His English is comprehensible but his content isn't :-) --Malcolm -- Malcolm Beattie Oxford University Computing Services "I permitted that as a demonstration of futility" --Grey Roger ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 26 Jun 2000 13:22:17 GMT Organization: IBM Austin Lines: 31 Message-ID: <8j7le9$13v4$1@ausnews.austin.ibm.com> References: <8j3l83$6mp@netaxs.com> <55E2821F92DE89ED.35495102DD55329E.979A24FB048DF463@lp.airnews.net> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!colt.net!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!znr.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news2atm.raleigh.ibm.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58593 In <55E2821F92DE89ED.35495102DD55329E.979A24FB048DF463@lp.airnews.net>, jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: >On 25 Jun 2000 00:54:27 GMT, lwin wrote: >>CICS isolates each program and terminal. If there is a bug in a program, >>that particular program will fail, but other users will not be affected. >>The user executing the program will see the program abend, but NOT his >>CICS session. > >Not in the sense a Unix person is thinking, it doesn't. CICS programs in a >region execute in the same address space, with no hardware memory protection >from others. CICS is good about blowing out just the failing program, but it >is still possible for a CICS application to take the whole region down. >(Well, unless subsystem storage protection is in use; even then, a CICS >application can trash another application in the same region.) Well, of course, there have been some enhancements in this area to prevent one rogue application from taking the region down. CICS and the hardware now supports running user transactions in a different storage key than the CICS system itself (key 9 for user transactions versus key 8 for CICS itself). Additionally, parts of CICS are loaded into key-0 (I think) read-only storage. And, since the user transactions are running key-9, they can't modify sections of CICS itself. And, finally, CICS can now run user transactions in a separate address space, well, sort of a separate address space, by using the subspace group facility. For more information about these various protection features, check the CICS manuals for information on STORAGE PROTECTION, and TRANSACTION ISOLATION. Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 26 Jun 2000 13:24:52 GMT Organization: IBM Austin Lines: 24 Message-ID: <8j7lj4$13v4$2@ausnews.austin.ibm.com> References: <8iun3f$3p1@netaxs.com> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!znr.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news2atm.raleigh.ibm.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58594 In , jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: >On Fri, 23 Jun 2000 19:47:23 GMT, Kragen Sitaker wrote: >>I'm not sure what a "process" maps to in the mainframe world, but I'm >>thinking of something including at least one stack, a set of memory >>mappings, and some information in the supervisor about things like open >>files^H^H^H^H^Hdatasets. Does such an entity exist in MVS? What do >>you call it? > >Well, kinda sorta. The 370 architecture doesn't have a stack per se (yes, >there's a linkage stack, but that was a 370/XA addition). A job (or started >task, or TSO user) occupies an address space, which corresponds to a set of >memory mappings and I/O definitions (what JCL is really good for). An >address space can have multiple tasks, each representing a different running >program; each TCB (task control block) is dispatchable on a different CPU in >the configuration. > >Does that answer your question, or just confuse you more? :-) Well, actually, the hardware linkage stack was added in the ESA architecture. Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### Newsgroups: alt.folklore.computers Date: Mon, 26 Jun 2000 10:19:34 -0400 From: Charlton Wilbur X-Sender: cwwilbur@lessing.oit.umass.edu Subject: Re: Does the word "mainframe" still have a meaning? In-Reply-To: Message-ID: References: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII NNTP-Posting-Host: lessing.oit.umass.edu X-Trace: 26 Jun 2000 10:19:37 -0500, lessing.oit.umass.edu Organization: University of Massachusetts, Amherst Lines: 29 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.cwix.com!news.umass.edu!oit.umass.edu!lessing.oit.umass.edu!cwwilbur Xref: chonsp.franklin.ch alt.folklore.computers:58560 On 25 Jun 2000, Erno Palonheimo wrote: > Beowulf is made up of glue and well-chewed bubble gum. It's not usable > in production systems. I'd like to say same about Linux in general, > but it has gained quite some commercial support and special hardware > vendors, so ok, now it's possible to run it with support contracts on > hardware and software, but still it is an ugly resource hog designed > by a crew with some witty ones (Linus himself, Alan Cox et al) and a > lot of not-so-witty-ones (such as the idiots who decided to write a > kernel-level httpd & nfsd). But the beauty of the Linux approach is that you can adapt a different nfsd if you care that strongly, or write one from scratch, or pay someone to write one for you. If it's better, you can make a case for it, and it will replace what's there. Linux is not a win in the long run because it's magically technically superior. In fact, it's frequently technically inferior, especially in development releases. Linux is a win in the long run because its design philosophy and programmer culture allow for people to make improvements on it and then release the improvements to the world at large. Charlton -- Charlton Wilbur | Wer bin ich, waer' ich deine Wille nicht? University of Massachusetts | -- Bruennhilde cwwilbur@music.umass.edu | 1832 - 1841 - 1976 - 1992 - 2001 ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Fri, 30 Jun 2000 20:24:32 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> References: <394F3A96.534E7003@trailing-edge.com> <8j55im$9db$1@news.ox.ac.uk> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.142.89 X-Original-NNTP-Posting-Host: 207.148.142.89 X-Trace: 30 Jun 2000 20:24:35 -0700, 207.148.142.89 Lines: 59 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!nntp.cadvision.com!news.cadvision.com!207.148.142.89 Xref: chonsp.franklin.ch alt.folklore.computers:58913 On 25 Jun 2000 14:39:18 GMT, mbeattie@sable.ox.ac.uk (Malcolm Beattie) wrote: >In article , >Kragen Sitaker wrote: >>In article , >>Brian Inglis wrote: >>>The one distinguishing characteristic I have noticed on mainframe >>>systems is the presence of multiple I/O channels to which any I/O >>>interface could be attached. As compared to the minicomputer CPU >>>and I/O interfaces on the backplane, or the microcomputer CPU and >>>interfaces on the motherboard. Perhaps this might be a better >>>taxonomy. >> >>So the SGI Octanes and Visual Workstations, with their crossbar I/O >>architecture, are mainframes? How about a Beowulf with an Ethernet >>switch as a backplane? > >A modern mainframe running VM provides 500 Mbytes/sec of bandwidth >between its guests via IUCV which appears to the guests somewhat >like a virtual ethernet interface (well, a network interface anyway). >That would be, um, pricey to implement with an ethernet-based >interconnect. On the I/O side, you can have up to 38 FICON (each >100 Mbytes/sec) channels or 256 ESCON channels (each ~16 Mbytes/sec). >To even try to compete with that, you have a much smaller range of >non-mainframe systems (e.g. Sun E10K) in the same ballpark. I/O >multipathing is another indicator of mainframe-like design and >that's more mature for S/390 than for E10K. > >--Malcolm Somewhat unfair comparison, as IUCV is internal to the system. But your numbers on channels are interesting and new to me: higher than I would have expected IBM to have reached by now, given their slow movement from 3 to 4.5MBps. Thanks for the update on S/390 I/O capabilities, which I have not looked at in over 8 years, when ESCON was promised. I'll have to check out the big iron sections of the web soon, and update myself. You're preaching to the converted here -- but hopefully some others as well as myself will be educated. I slid myself out of VM systems programming as it became obvious that mainframes were going to go out of fashion in this region, and I wanted to stay here. I have an MVS sysprog friend who has managed to hang on until now, but he sees a very low probability of any mainframes left in town in a couple of years, and a high probability of the bureaux running them going out of business within that time. Other areas of this country seem to be going the same way, except for a few very large businesses and governments. Anyone care to comment what the prospects are like for mainframes in their region in the near future? Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Fri, 30 Jun 2000 20:24:34 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <8j3l83$6mp@netaxs.com> <55E2821F92DE89ED.35495102DD55329E.979A24FB048DF463@lp.airnews.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.142.89 X-Original-NNTP-Posting-Host: 207.148.142.89 X-Trace: 30 Jun 2000 20:24:37 -0700, 207.148.142.89 Lines: 30 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.icl.net!newsfeed.skycache.com!Cidera!209.50.235.254!europa.netcrusader.net!205.231.236.10!newspeer.monmouth.com!nntp.cadvision.com!news.cadvision.com!207.148.142.89 Xref: chonsp.franklin.ch alt.folklore.computers:58826 On 25 Jun 2000 02:08:46 GMT, jmaynard@thebrain.conmicro.cx (Jay Maynard) wrote: >On 25 Jun 2000 00:54:27 GMT, lwin wrote: >>CICS isolates each program and terminal. If there is a bug in a program, >>that particular program will fail, but other users will not be affected. >>The user executing the program will see the program abend, but NOT his >>CICS session. > >Not in the sense a Unix person is thinking, it doesn't. CICS programs in a >region execute in the same address space, with no hardware memory protection >from others. CICS is good about blowing out just the failing program, but it >is still possible for a CICS application to take the whole region down. >(Well, unless subsystem storage protection is in use; even then, a CICS >application can trash another application in the same region.) One more reason TABLEs/arrays were abhorred in COBOL. Application programmers never did latch on to the idea of bounds checking, and I assume the compiler and runtime libraries didn't help, from the number of times I had to read dumps from the test CICS partition/region. Got smart quickly about scanning the source code for TABLEs and emailing pointed suggestions and bounds checking code snippets to programmers. Got out of that area about the time both CICS (~V1.5) and the applications were starting to stabilize. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### Message-ID: <395D6E84.DB233C22@cmc.com> Date: Fri, 30 Jun 2000 21:07:32 -0700 From: Lars Poulsen X-Mailer: Mozilla 4.73 [en] (Win98; I) X-Accept-Language: da,en MIME-Version: 1.0 Newsgroups: alt.folklore.computers To: Brian.Inglis@SystematicSw.ab.ca Subject: Re: Does the word "mainframe" still have a meaning? References: <394F3A96.534E7003@trailing-edge.com> <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 33 NNTP-Posting-Host: 207.154.90.88 X-Trace: azure.impulse.net 962424406 194 207.154.90.88 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.netcologne.de!sienna.impulse.net!azure.impulse.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58899 Brian Inglis wrote: > ... I slid myself out of > VM systems programming as it became obvious that mainframes were > going to go out of fashion in this region, and I wanted to stay > here. > I have an MVS sysprog friend who has managed to hang on until > now, but he sees a very low probability of any mainframes left in > town in a couple of years, and a high probability of the bureaux > running them going out of business within that time. Other areas > of this country seem to be going the same way, except for a few > very large businesses and governments. My take on why the mainframes are dropping out: The load that needed a 370/158 with MVS and CICS today can often be handled by a small PC server (say dual P3 and a SCSI-RAID array) with a webserver, and if it can't, we are much better at partitioning it today than we used to be. In any case, today's S/390 requires a much heavier load to justify. There is a place for it in the headquarters of a large corporation, or in a centralized government operation. But if Canada with its tradition of good government has a scarcity of those applications, it can only be worse everywhere else. In "the old days" every installation needed two operators, at least one systems programmer, a business analyst and two applications programmers, so to achieve reasonable economics, the needs of several businesses were aggregated at a service bureau. Today, a business will buy one or two smaller machines, contract customization and maintenance of a business software package out to a consultant, and avoid the bureau. -- / Lars Poulsen - http://www.cmc.com/lars - lars@cmc.com 125 South Ontare Road, Santa Barbara, CA 93105 - +1-805-569-5277 ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 51 Message-ID: <8pe75.8551$I96.261411@news-east.usenetserver.com> X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 01 Jul 2000 00:25:40 EDT Date: Sat, 01 Jul 2000 04:25:40 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58924 In article <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com>, Brian Inglis wrote: >On 25 Jun 2000 14:39:18 GMT, mbeattie@sable.ox.ac.uk (Malcolm >Beattie) wrote: >>In article , >>Kragen Sitaker wrote: >>>So the SGI Octanes and Visual Workstations, with their crossbar I/O >>>architecture, are mainframes? How about a Beowulf with an Ethernet >>>switch as a backplane? >> >>A modern mainframe running VM provides 500 Mbytes/sec of bandwidth >>between its guests via IUCV > >Somewhat unfair comparison, as IUCV is internal to the system. I was positing Beowulfs connected via Ethernet as "mainframes", if the term is defined by high-bandwidth backplanes. The Ethernet is internal to the Beowulf, so it's a fair comparison. ;) 500 Mbytes/sec point-to-point is really impressive. >You're preaching to the converted here -- but hopefully some >others as well as myself will be educated. My point was not, "mainframes suck" but, "maybe a definition based on I/O architecture is not a perfect definition of 'mainframe', because these other systems that have I/O architectures like you describe are not really considered mainframes". >I have an MVS sysprog friend who has managed to hang on until >now, but he sees a very low probability of any mainframes left in >town in a couple of years, and a high probability of the bureaux >running them going out of business within that time. Other areas >of this country seem to be going the same way, except for a few >very large businesses and governments. > >Anyone care to comment what the prospects are like for mainframes >in their region in the near future? I'd be very surprised if they disappeared altogether. I'd actually heard IBM's mainframe revenues were on their way up; seems the Web encourages big centralized databases and big servers, so mainframes have become more useful. Also, "data warehousing" is a big fashion, or was a year or two ago. Cashews are yummy. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <395D6E84.DB233C22@cmc.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 21 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 01 Jul 2000 00:32:14 EDT Date: Sat, 01 Jul 2000 04:32:14 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58824 In article <395D6E84.DB233C22@cmc.com>, Lars Poulsen wrote: >In "the old days" every installation needed two operators, at >least one systems programmer, a business analyst and two >applications programmers, so to achieve reasonable economics, the >needs of several businesses were aggregated at a service bureau. >Today, a business will buy one or two smaller machines, contract >customization and maintenance of a business software package out >to a consultant, and avoid the bureau. . . . but now "ASPs" --- application service providers, the new name for service bureaus --- are a hot buzzword. Companies like Interliant are growing quite rapidly, despite bleeding megabucks. Even in the real world, outside of businesses, many people use huge "service bureaus" like Google, AltaVista, Yahoo, and Hotmail to provide many of their computing services. In 1994, they didn't. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Message-ID: <395D79C0.4B72BBE8@cmc.com> Date: Fri, 30 Jun 2000 21:55:28 -0700 From: Lars Poulsen X-Mailer: Mozilla 4.73 [en] (Win98; I) X-Accept-Language: da,en MIME-Version: 1.0 Newsgroups: alt.folklore.computers To: Kragen Sitaker Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <395D6E84.DB233C22@cmc.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 24 NNTP-Posting-Host: 207.154.90.88 X-Trace: azure.impulse.net 962427282 191 207.154.90.88 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!newsfeed.direct.ca!look.ca!sn-xit-03!supernews.com!sienna.impulse.net!azure.impulse.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58903 Kragen Sitaker wrote: > . . . but now "ASPs" --- application service providers, the new > name for service bureaus --- are a hot buzzword. Companies like > Interliant are growing quite rapidly, despite bleeding megabucks. The statement was that the service bureaus that provided jobs for MVS system programmers are going away. ASPs don't propose to run MVS ... Also, it will take a while before we find out if "there is a pony" in the ASP business. > Even in the real world, outside of businesses, many people use > huge "service bureaus" like Google, AltaVista, Yahoo, and Hotmail > to provide many of their computing services. And none of those charge the users for services rendered ... But indeed, some of those are good candidates for mainframes. One particular kind of profitable ASP is the web hosting facility. An S/390 with 40,000 instances of Linux makes one heck of a hosting facility. Still no MVS, though. -- / Lars Poulsen - http://www.cmc.com/lars - lars@cmc.com 125 South Ontare Road, Santa Barbara, CA 93105 - +1-805-569-5277 ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <395D6E84.DB233C22@cmc.com> <395D79C0.4B72BBE8@cmc.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 50 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 01 Jul 2000 01:13:01 EDT Date: Sat, 01 Jul 2000 05:13:01 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58930 In article <395D79C0.4B72BBE8@cmc.com>, Lars Poulsen wrote: >Kragen Sitaker wrote: >> . . . but now "ASPs" --- application service providers, the new >> name for service bureaus --- are a hot buzzword. Companies like >> Interliant are growing quite rapidly, despite bleeding megabucks. > >The statement was that the service bureaus that provided jobs for >MVS system programmers are going away. ASPs don't propose to run >MVS ... Don't they? I thought some of them did. . . >Also, it will take a while before we find out if "there is a pony" >in the ASP business. True enough. >> Even in the real world, outside of businesses, many people use >> huge "service bureaus" like Google, AltaVista, Yahoo, and Hotmail >> to provide many of their computing services. > >And none of those charge the users for services rendered ... Most don't. I use pobox.com to handle my email, though, which does. I suspect that most of the free services run by for-profit companies will either have to resort to outright villainy (as AltaVista, for example, has --- selling their users' search terms to DoubleClick! For shame!) or fail. Google is an interesting case; they make money by selling their search service to Web sites for a bit less than a penny (down to about 0.6 cents) per query, plus a bit of prostitution (selling ads), which is probably doomed in the long run. Their free service is a way to build a good brand so people (like, e.g., Yahoo) will want to buy their service. >But indeed, some of those are good candidates for mainframes. >One particular kind of profitable ASP is the web hosting facility. >An S/390 with 40,000 instances of Linux makes one heck of a hosting >facility. Still no MVS, though. I know an S/390 has run 40,000 instances of Linux, but I think none of them were running anything. My experience with the ratio of RAM required to boot Linux and to do something useful with it on PCs suggests that 1,000 to 5,000 might be a more reasonable number. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers From: nailed_barnacleSPAMFREE@hotmail.com (barnacle) Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> X-Newsreader: News Xpress 2.01 Lines: 14 Message-ID: Date: Sat, 01 Jul 2000 07:30:13 GMT NNTP-Posting-Host: 194.154.98.206 X-Complaints-To: abuse@easynet.net X-Trace: monolith.news.easynet.net 962436789 194.154.98.206 (Sat, 01 Jul 2000 08:33:09 BST) NNTP-Posting-Date: Sat, 01 Jul 2000 08:33:09 BST Organization: [posted via Easynet] Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.icl.net!diablo.theplanet.net!easynet-melon!easynet.net!monolith.news.easynet.net!egbert Xref: chonsp.franklin.ch alt.folklore.computers:58911 In article <8pe75.8551$I96.261411@news-east.usenetserver.com>, kragen@dnaco.net (Kragen Sitaker) wrote: >Cashews are yummy. but a) poisonous/corrosive before processing b) grow *outside* their fruit c) keep dying when I try to grow them in the UK Sorry. It's been a long week :) -- barnacle http://www.nailed-barnacle.co.uk ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 96 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 01 Jul 2000 15:50:14 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962466614 209.63.28.138 (Sat, 01 Jul 2000 09:50:14 MDT) NNTP-Posting-Date: Sat, 01 Jul 2000 09:50:14 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58881 kragen@dnaco.net (Kragen Sitaker) writes: > In article <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com>, > Brian Inglis wrote: > >On 25 Jun 2000 14:39:18 GMT, mbeattie@sable.ox.ac.uk (Malcolm > >Beattie) wrote: > >>In article , > >>Kragen Sitaker wrote: > >>>So the SGI Octanes and Visual Workstations, with their crossbar I/O > >>>architecture, are mainframes? How about a Beowulf with an Ethernet > >>>switch as a backplane? > >> > >>A modern mainframe running VM provides 500 Mbytes/sec of bandwidth > >>between its guests via IUCV my wife and I sat thru a s/390 (future) roadmap presentation last week. One of the foils used was something she had drawn 25 years ago when she was in POK and responsible for non-shared-memory multiprocessing (aka cluster) architecture random refs. http://www.garlic.com/~lynn/99.html#71 http://www.garlic.com/~lynn/99.html#77 http://www.garlic.com/~lynn/99.html#100 http://www.garlic.com/~lynn/2000.html#13 another part of the roadmap presentation was ficon ... this I haven't looked at in nearly 10 years. It appeared to be the work that some people were doing on FC-4 (fiber-channel standard) level mapping syncronous half-duplex channel protocol on top of asyncronous dual-simplex (i.e. dedicated fiber for transmission in each direction, aka full-duplex) fiber-channel. another bullet for the s/390 roadmap was high speed queued I/O ... which also brought back memories. I don't know if anything was done with it but at the time we were doing fiber-channel cluster scaleup ... another person and I wrote up patent application for high speed queued I/O. When I mentioned that, the person doing the presentation commented that sometimes it takes decades for technology to show up in mainframes. random refs: http://www.garlic.com/~lynn/95.html#13 http://www.garlic.com/~lynn/99.html#145 http://www.garlic.com/~lynn/2000c.html#12 http://www.garlic.com/~lynn/2000c.html#21 http://www.garlic.com/~lynn/2000c.html#22 ... misc. posting from fiber channel standard mailing list > Date: Tue, 09 Jul 91 16:52:54 PDT > To: fibre-channel-ext@Think.COM > Subject: Revisting Bi-Directional Sequences > > I've been looking at what it would take to map (CCWs) Channel Command > Words (Block Mux to Dal) onto FC-PH and have hit the same restriction > found some time ago. The problem is that CCW operations > are necessarily bi-directional in nature, and require simultaneous > bi-directional transport mechanisms to be supported within FC-PH. > > .... > > The sentiment of the working group was that FC-PH already provides a > dual-exchange solution to this problem. However, there is no way to > relate or control dual-exchanges in FC-PH currently. In addition, there > are several other reasons not to use dual-exchanges: > > 1. Operations requiring dual-exchanges may not be initiated if two > exchanges are not available. > > 2. Twice the number of exchanges are used up. This will likely result > in a costlier implementation. > > 3. Dual-exchange use will affect operation recovery. > > 4. An exchange is the wrong level (too-high) to provide support for > bi-directional transfer. > > I propose that the FCS working group fully investigate the issue of > supporting bi-directional Sequences within an Exchange as soon as > possible since FC-PH may be impacted. I don't believe that full reverse > direction Sequence support is required as I'm assuming that traffic in > the reverse direction is minimal. This is certainly true of CCW > operations and should be explored for IPI, SCSI, and other FC4s. This > limitation may result in a fairly simple change to FC-PH while providing > effective simultaneous bi-directional operation support. > > I'd like to discuss this issue in more detail next week while we're > discussing other FC-4 to FC-PH mapping issues. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 46 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sat, 01 Jul 2000 14:56:25 EDT Date: Sat, 01 Jul 2000 18:56:25 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed2.onemain.com!feed1.onemain.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58917 In article , barnacle wrote: >In article <8pe75.8551$I96.261411@news-east.usenetserver.com>, >kragen@dnaco.net (Kragen Sitaker) wrote: > >>Cashews are yummy. > > but >a) poisonous/corrosive before processing Yes, I'd heard that. I thought it was urushiol, but on looking it up, it turns out it's cardol, a similar substance. >b) grow *outside* their fruit What is the fruit like? >c) keep dying when I try to grow them in the UK You're trying to grow cashews? What kind of madman are you? :) I think you'll have to maintain a real tropical climate, which is a very expensive undertaking in the UK. For those of us who are susceptible to it, urushiol is really nasty stuff, as plant toxins go. I don't know of any other natural plant toxins that can produce such a reaction from simple skin contact, except for the cashew shell oil. Are you not susceptible to urushiol? I understand that cashew shell oil is even worse, containing cardol plus phenols (carbolic acid) and "anacardic acids", whatever those are. I just discovered that cashew shell oil has been used for wart removal in Africa for quite a while. Also, ritual scarification --- like tattoos, but without the ink :) >Sorry. It's been a long week :) So, how do you do the climate control on your tropical greenhouse? Do you, perchance, use a PDP-11? :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 73 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 02 Jul 2000 03:12:24 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962507544 209.63.28.138 (Sat, 01 Jul 2000 21:12:24 MDT) NNTP-Posting-Date: Sat, 01 Jul 2000 21:12:24 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58878 something else in the s/390 roadmap was outboard TCP protocol handler. one of the problems with one of the mainframe channel attach boxes in the 80s was that it basically was a bridge ... not a real router; the mainframe had to not only do the tcp & ip header stuff inboard, but fragmentation & MAC headers before passing it to the outboard box ... which then just had to drop the packet on enet or t/r (which the mainframe also had to handle). The box i used for rfc1044 implementation was full router ... could pass it a ip packet ... and it would handle fragmentation & MAC stuff (outboard). Later boxes ... about time of T3/NSFNET2 activity included additional outboard protocol processing (including large packets support ... fragmentation for outbound, re-assembly for inbound). Later objective was to pass a TCP packet & have it do both IP & MAC layer processing. random refs: http://www.garlic.com/~lynn/96.html#14 http://www.garlic.com/~lynn/96.html#17 http://www.garlic.com/~lynn/98.html#49 http://www.garlic.com/~lynn/98.html#50 http://www.garlic.com/~lynn/99.html#36 other work on outboard processing in 80s w/XTP (a lot of testing done w/100mbit FDDI) http://www.garlic.com/~lynn/99.html#0 http://www.garlic.com/~lynn/99.html#164 http://www.garlic.com/~lynn/99.html#207 http://www.garlic.com/~lynn/2000b.html#5 .... a reply (from somebody else) regarding CCWs on FCS. Subject: Reply to Revisiting Bi-Directional Seque Date: 10 Jul 91 16:55:14 EDT To: fc ... I guess that the basic question that you still have to answer, at least for me, is why you need a fully bidirectional Sequence (or Exchange) in FC when the OEMI (or FIPS-60 or whatever) that the CCWs run over today only operates in half-duplex mode exactly like the current FC exchange definition!! If you can explain that in some detail we might even be able to have a sensible discussion on the subject - no guarantees of course! My view at the moment is that the architecture of Fiber Channel utterly collapses if true bidirectional Sequences are required i.e. we can throw away FC-2 completely and start again. The replacement protocol would need to be tightly synchronized and would require that frames be completely processed in real-time upon reception. The reason for this time-sensitivity is that in order to be able to accurately decode the "return direction information", the receiver needs to be able to recreate the status of the link that the generator saw at the time of transmitting that information e.g. had it started to receive the Sequence that is "in flight" towards it etc. I will be presenting Mark 2 of my FC-4 Paradigm next week in Valley Forge, along with an updated FC-4 for IPI. That work has convinced me that, far from requiring bidirectional Sequences, IPI-3 CANNOT be made deadlock-free if bidirectional operation is allowed. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Newsgroups: alt.folklore.computers From: nailed_barnacleSPAMFREE@hotmail.com (barnacle) Subject: Re: Does the word "mainframe" still have a meaning? References: <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> X-Newsreader: News Xpress 2.01 Lines: 61 Message-ID: <8mB75.2576$Dr3.171179@monolith.news.easynet.net> Date: Sun, 02 Jul 2000 06:30:53 GMT NNTP-Posting-Host: 194.154.98.206 X-Complaints-To: abuse@easynet.net X-Trace: monolith.news.easynet.net 962519556 194.154.98.206 (Sun, 02 Jul 2000 07:32:36 BST) NNTP-Posting-Date: Sun, 02 Jul 2000 07:32:36 BST Organization: [posted via Easynet] Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.icl.net!colt.net!easynet-melon!easynet.net!monolith.news.easynet.net!egbert Xref: chonsp.franklin.ch alt.folklore.computers:58908 In article , kragen@dnaco.net (Kragen Sitaker) wrote: >In article , >barnacle wrote: >>In article <8pe75.8551$I96.261411@news-east.usenetserver.com>, >>kragen@dnaco.net (Kragen Sitaker) wrote: >> >>>Cashews are yummy. >> >> but >>a) poisonous/corrosive before processing > >Yes, I'd heard that. I thought it was urushiol, but on looking it up, >it turns out it's cardol, a similar substance. I didn't know the name. > >>b) grow *outside* their fruit > >What is the fruit like? It's a pseudo-fruit (apparently) with a good flavour that makes *great* fruit juices. Tends to be bitter in the ones I've tried in Brazil though, in spite of the encyclopedia entry http://www.rain-tree.com/cashew.htm Nice pictures there too. > >>c) keep dying when I try to grow them in the UK > >You're trying to grow cashews? What kind of madman are you? :) I'm an English madman, which traditionally renders me merely eccentric :) > >I think you'll have to maintain a real tropical climate, which is a >very expensive undertaking in the UK. I just have a bad habit of eating fruit and sticking the pips/stones/whatever in mud to see what sprouts...I got half a dozen to sprout on the kitchen windowsill and most lived for about two years before they noticed they didn't have proper visas and exported themselves. > >For those of us who are susceptible to it, urushiol is really nasty >stuff, as plant toxins go. I don't know of any other natural plant >toxins that can produce such a reaction from simple skin contact, >except for the cashew shell oil. Are you not susceptible to urushiol? >I understand that cashew shell oil is even worse, containing cardol >plus phenols (carbolic acid) and "anacardic acids", whatever those >are. The cunning engineer type, observes that the people who pick the fruit suffer no effects unless the shell around the nut is damaged. So I just carried the nuts like that, stuck them in mud, and waited. I don't have green fingers, even after cashews! -- barnacle http://www.nailed-barnacle.co.uk ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 03 Jul 00 09:26:55 GMT Organization: UltraNet Communications, Inc. Lines: 82 Message-ID: <8jq0ga$7pf$3@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> X-Trace: ldY/7hz8+Lw0C7oylZsL7qROngC7Ed0j+ar5n0ezdVU= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 3 Jul 2000 12:21:30 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-255-136 Xref: chonsp.franklin.ch alt.folklore.computers:58997 In article <8joosu$p0h@netaxs.com>, lwinson@bbs.cpcn.com (lwin) wrote: >> My take on why the mainframes are dropping out: The load that >> needed a 370/158 with MVS and CICS today can often be handled >> by a small PC server (say dual P3 and a SCSI-RAID array) with >> a webserver, > >Can somebody confirm this? It seems awfully optimistic to me. > >Our old S/370-158 supported a network of 500 online terminals >in a business world. The database referenced contained millions >of records. Plus daytime batch processing. > >Remember, too, the terminals were all "dumb", all formatting >and the like was handled by the mainframe. Only when the dumb terminals were the only devices available. There were "smart" terminals, too. And, as these devices were developed mainframe OS development began to support them. Where are you getting these notions from? > >I'm not sure what a "small PC server" would represent, but would >it have adequate I/O to handle that, esp with dumb cheap >terminals? That's not the key, IMO. Memory protection would be much more important. In order to do mainframe work, the computer would have to be able to prevent a user from trashing another users (or process) address space. > > >> and if it can't, we are much better at partitioning >> it today than we used to be. > >I'm not sure what "partitioning" means as far as server. What happens >when a wide network of dumb terminals serving a wide variety of >application programs call for response at the same time? You buy a system that supports a timesharing environment :-). > > >> In any case, today's S/390 requires >> a much heavier load to justify. > >Obviously that's true, but then applications have grown too. Much >more functionality is available on-line, such as archived history data, >or specialized functions. Customer records are retained on-line covering >far more detail and going further back in time. > >Before PCs, there wasn't anything else, so small applications were run >on the mainframe as well. This is just nonsense. Take a look at what people did with PDP-8s and PDP-11s. I'm sure other manufacturers have similar examples. >Today those are run on PCs. The mainframe >wasn't purchased for them, but was used none the less. And now the management of those systems is a nightmare. For instance, nstead of one full system backup, the admins now have 1000s of full system backups if they really want to take care of their stable of users. What has been happening is that the individual user has now become responsible for operations, data integrity, hardware maintenance, software maintenance, etc, etc, etc,. One of the advantages of a mainframe environment is that all of these functions were invisible to the user, allowing him/her to concentrate on their work rather than on the longevity of their work. Have you done your backups today? :-)) /BAH Subtract a hundred and four for e-mail. ###### From: lwinson@bbs.cpcn.com (lwin) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 3 Jul 2000 01:05:33 GMT Organization: The PACSIBM SIG BBS Lines: 40 Message-ID: <8joosu$p0h@netaxs.com> NNTP-Posting-Host: bbs.cpcn.com Originator: root@bbs.cpcn.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-xfer.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!bbs.cpcn.com!root Xref: chonsp.franklin.ch alt.folklore.computers:58966 > My take on why the mainframes are dropping out: The load that > needed a 370/158 with MVS and CICS today can often be handled > by a small PC server (say dual P3 and a SCSI-RAID array) with > a webserver, Can somebody confirm this? It seems awfully optimistic to me. Our old S/370-158 supported a network of 500 online terminals in a business world. The database referenced contained millions of records. Plus daytime batch processing. Remember, too, the terminals were all "dumb", all formatting and the like was handled by the mainframe. I'm not sure what a "small PC server" would represent, but would it have adequate I/O to handle that, esp with dumb cheap terminals? > and if it can't, we are much better at partitioning > it today than we used to be. I'm not sure what "partitioning" means as far as server. What happens when a wide network of dumb terminals serving a wide variety of application programs call for response at the same time? > In any case, today's S/390 requires > a much heavier load to justify. Obviously that's true, but then applications have grown too. Much more functionality is available on-line, such as archived history data, or specialized functions. Customer records are retained on-line covering far more detail and going further back in time. Before PCs, there wasn't anything else, so small applications were run on the mainframe as well. Today those are run on PCs. The mainframe wasn't purchased for them, but was used none the less. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 45 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 02 Jul 2000 23:20:06 EDT Date: Mon, 03 Jul 2000 03:20:06 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59053 In article <8joosu$p0h@netaxs.com>, lwin wrote: >> My take on why the mainframes are dropping out: The load that >> needed a 370/158 with MVS and CICS today can often be handled >> by a small PC server (say dual P3 and a SCSI-RAID array) with >> a webserver, > >Can somebody confirm this? It seems awfully optimistic to me. > >Our old S/370-158 supported a network of 500 online terminals >in a business world. The database referenced contained millions >of records. Plus daytime batch processing. A "small PC server" as described above can indeed support databases of millions of records. 500 online terminals is small potatoes. You can definitely support 500 simultaneous Web browsers, barring huge amounts of computation. I strongly suspect you could support 500 telnet sessions if you had enough RAM. 1MB per user would probably be plenty with off-the-shelf software (although, as you surely know, that is a profligate use of RAM.) >Remember, too, the terminals were all "dumb", all formatting >and the like was handled by the mainframe. > >I'm not sure what a "small PC server" would represent, but would >it have adequate I/O to handle that, esp with dumb cheap >terminals? When you say "dumb cheap terminals", what do you mean by "dumb"? A 3270, which is roughly as smart as a Web browser without JavaScript? And what do you mean by "cheap"? http://www.mticom.com/wyse.htm lists used WYSE 30 terminals each with a 90-day warranty at $130, WYSE 50s with a one-year warranty at $190 each, and color WYSE 370s with a one-year warranty for $435. Used. Used 3270s apparently cover a slightly wider range on both ends. At these prices, a PC starts to look pretty attractive as a terminal replacement :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 19 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <962598777.215926@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-68.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Mon, 3 Jul 2000 00:31:12 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 962598613 204.92.64.17 (Mon, 03 Jul 2000 00:30:13 EDT) NNTP-Posting-Date: Mon, 03 Jul 2000 00:30:13 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cyclone.bc.net!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59008 > > When you say "dumb cheap terminals", what do you mean by "dumb"? A > 3270, which is roughly as smart as a Web browser without JavaScript? > And what do you mean by "cheap"? http://www.mticom.com/wyse.htm lists > used WYSE 30 terminals each with a 90-day warranty at $130, WYSE 50s > with a one-year warranty at $190 each, and color WYSE 370s with a > one-year warranty for $435. Used. Used 3270s apparently cover a > slightly wider range on both ends. > > At these prices, a PC starts to look pretty attractive as a terminal > replacement :) IMHO and from practical experience replacing a 5251 with a pc keyboard means replacing a cufflink with a tensor bandage. Rick ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 3 Jul 2000 07:08:37 GMT Organization: Neosoft (using Airnews.net!) Lines: 25 Message-ID: <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Mon Jul 3 02:08:37 2000 NNTP-Posting-Host: !XtkM1k-W+BGfJb (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58973 On Mon, 03 Jul 2000 03:20:06 GMT, Kragen Sitaker wrote: >When you say "dumb cheap terminals", what do you mean by "dumb"? A >3270, which is roughly as smart as a Web browser without JavaScript? No. No. NO! A CUT-style 3270 (basically, anything but a 3290, 3179, or some varieties of PC emulator) is *extremely* stupid. It makes a VT100 look positively brilliant. Without the controller, it can't even process a keystroke. All of the intelligence - what little there is of it - resides in the controller. Now, when you take the terminal/controller combination together, you get what's essentially a block-mode VT100. It can handle highlight/blink/ non-displayed fields and cursor positioning, but that's about it. The keyboard-controlled cursor movement/editing functions are all processed in the controller. Many character-mode async terminals had this capability, though I don't know if the VT100 or higher does. Attribute bytes take up a space on the screen for the magic cookie. A web browser has, in addition to the ability to act like a dumb terminal, the ability to use many different fonts, versatile special formatting, handle output far in excess of one visible page and navigate through it, and the ability to figure out what's being sent and deal with it appropriately in the current environment, just off the top of my head. All of these are far beyond the 3270's capabilities. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 55 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Mon, 03 Jul 2000 04:15:46 EDT Date: Mon, 03 Jul 2000 08:15:46 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59056 In article <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net>, Jay Maynard wrote: >On Mon, 03 Jul 2000 03:20:06 GMT, Kragen Sitaker wrote: >>When you say "dumb cheap terminals", what do you mean by "dumb"? A >>3270, which is roughly as smart as a Web browser without JavaScript? > >No. No. NO! > >A CUT-style 3270 (basically, anything but a 3290, 3179, or some varieties of >PC emulator) is *extremely* stupid. It makes a VT100 look positively >brilliant. Without the controller, it can't even process a keystroke. All of >the intelligence - what little there is of it - resides in the controller. Do people use them without controllers? Do the controllers live in a separate box? >Now, when you take the terminal/controller combination together, you get >what's essentially a block-mode VT100. It can handle highlight/blink/ >non-displayed fields and cursor positioning, but that's about it. The >keyboard-controlled cursor movement/editing functions are all processed in >the controller. Many character-mode async terminals had this capability, >though I don't know if the VT100 or higher does. No, it doesn't. So you have a form on the screen that you can fill out --- without talking to the computer. Then, when you're done filling out the form, you hit SEND, and it sends the form fields to the computer. Right? This is a huge advantage for the computer (over a machine handling VT100s). Instead of handling five or so interrupts per second per terminal, it handles one every fifteen or twenty seconds at worst. Instead of having to respond within 200 milliseconds in order not to feel slow, it needs to respond within a second or two. That's why I brought up web browsers. It may be that a modern Linux box can handle 500 telnet sessions, but that is a much more impressive feat than handling 500 sessions of stuff like the previous paragraph, which a modern Linux box like the one being suggested three posts ago can handle without even noticing. >A web browser has, in addition to the ability to act like a dumb terminal, >the ability to use many different fonts, versatile special formatting, >handle output far in excess of one visible page and navigate through it, and >the ability to figure out what's being sent and deal with it appropriately >in the current environment, just off the top of my head. All of these are >far beyond the 3270's capabilities. "output far in excess of one visible page" is the only one of these that seems like a significant advantage for server performance to me. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 3 Jul 2000 13:40:20 GMT Organization: The MITRE Corporation Lines: 22 Message-ID: <8jq544$9l0$1@top.mitre.org> References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 962631620 9888 128.29.251.13 (3 Jul 2000 13:40:20 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 3 Jul 2000 13:40:20 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:58971 jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: >Exactly. This is part of the mainframe philosophy: offload work when you >can. Mainframers talk about "subsecond response time" as the Holy Grail of >tuning. If the user can make a trivial request, hit ENTER, and get his >keyboard unlocked in under a second, life is good. Back in the 1970s there were several studies (which I can't cite by name; it's been too many years) that examined the human engineering of terminal use. One relatively consistent finding was that the time between the dispatching of a request (typically by hitting the RETURN or ENTER key) and the arrival of the first response character should not exceed about 1.5 seconds because after that period the "typical" user's thought process would begin to drift to other, unrelated topics, and would require additional time to return to the current task once the output finally began to be displayed. (And please don't take that "1.5 second" figure as correct. It's what my memory serves up this morning and I think it's what the studies reported, but as I noted it's been many years since I saw the studies.) Joe Morris ###### From: byatesiii@my-deja.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 03 Jul 2000 14:20:22 GMT Organization: Deja.com - Before you buy. Lines: 25 Message-ID: <8jq7em$iii$1@nnrp1.deja.com> References: <8joosu$p0h@netaxs.com> <8jq0ga$7pf$3@bob.news.rcn.net> NNTP-Posting-Host: 4.17.20.4 X-Article-Creation-Date: Mon Jul 03 14:20:22 2000 GMT X-Http-User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows 95) X-Http-Proxy: 1.1 x68.deja.com:80 (Squid/1.1.22) for client 4.17.20.4 X-MyDeja-Info: XMYDJUIDbyatesiii Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news.man.poznan.pl!news.task.gda.pl!newsfeed00.sul.t-online.de!t-online.de!skynet.be!newsfeed.skycache.com!Cidera!152.163.239.131!portc03.blue.aol.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!news.maxwell.syr.edu!nntp2.deja.com!nnrp1.deja.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:58990 In article <8jq0ga$7pf$3@bob.news.rcn.net>, jmfbahciv@aol.com wrote: > What has been happening is that the individual user has > now become responsible for operations, data integrity, > hardware maintenance, software maintenance, etc, etc, etc,. > One of the advantages of a mainframe environment is that > all of these functions were invisible to the user, allowing > him/her to concentrate on their work rather than on the > longevity of their work. > > Have you done your backups today? :-)) > > /BAH > > Subtract a hundred and four for e-mail. > That certainly is true today. The company I work for has the attitude that "whatever is on your local drive is your responsibility", and that, I believe, is the correct opinion. You have a shared area that you share with your coworkers, and a private network drive for your own files, both which are backed up... Sent via Deja.com http://www.deja.com/ Before you buy. ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 50 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 14:51:49 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962635909 209.63.28.138 (Mon, 03 Jul 2000 08:51:49 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 08:51:49 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59021 jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: > Yes, and yes, respectively. The 3274 and 3174 controllers were separate boxes > with some amount of intelligence, capable of driving from 8 to 32 terminals > depending on the model. Some 3270s were attached directoy to dedicated > console ports on the machine, although I don't know how much of a 3x74 they > stuck in there to drive it. the (original) 3272 controllers and 3277 terminals had the logic in the heads with 3272 controller handling the channel interface. 3274's moved head & keyboard logic back into the controller for 3278s & 3279s terminals. There were 327x controllers that talked bisynch telecommunication protocol (instead of direct channel attach) ... and later SDLC telecommuncation protocol (& "LU2"). 327x had a problem with repeat key being way too slow and also if you happened to be typing at the moment the screen was written to ... the keyboard would lock up. there was a hack on the 3277 where you could patch a resister inside the 3277 keyboard and change the repeat key delay and rate (frequently used for moving the cursor around the screen in full screen operations, the resister values determined the repeat delay and rate). A slight human factors in this was that if you made it too fast, it was "higher" than the screen refresh rate i.e. in the case of repeat on cursor movement key, the cursor would continue to move after you lifted the key ... it just a little while to get used to the cursor coasting and be able to raise the key at the right moment so it stopped coasting under the correct position. For the keyboard locking, a FIFO box was built that fit where the 3277 keyboard plugged into the head (unplug the keyboard, plug in the FIFO box into the head and plug the keyboard into the FIFO box). 3278, 3279, etc moved all that logic back into the 3274 controllers (reducing per terminal cost) and it was no longer possible to improve on the human factors. I kept a modified 3277 around well into the late '80s. random urls: http://www.garlic.com/~lynn/98.html#49 http://www.garlic.com/~lynn/99.html#28 http://www.garlic.com/~lynn/99.html#69 http://www.garlic.com/~lynn/99.html#108 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> From: jata@aepiax.net (Julian Thomas) Subject: Re: Does the word "mainframe" still have a meaning? Message-ID: <3960aae8$1$wg$mr2ice@news.epix.net> X-Newsreader: MR/2 Internet Cruiser Edition for OS/2 v2.19zf/19zf Lines: 30 Date: Mon, 03 Jul 2000 15:02:02 GMT NNTP-Posting-Host: 199.224.125.59 X-Complaints-To: abuse@epix.net X-Trace: news1.epix.net 962636522 199.224.125.59 (Mon, 03 Jul 2000 11:02:02 EDT) NNTP-Posting-Date: Mon, 03 Jul 2000 11:02:02 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newspump.monmouth.com!newspeer.monmouth.com!news-xfer.epix.net!news1.epix.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59018 In <8jq544$9l0$1@top.mitre.org>, on 07/03/00 at 01:40 PM, jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) said: >Back in the 1970s there were several studies (which I can't cite by name; >it's been too many years) that examined the human engineering of terminal >use. One relatively consistent finding was that the time between the >dispatching of a request (typically by hitting the RETURN or ENTER key) >and the arrival of the first response character should not exceed about >1.5 seconds because after that period the "typical" user's thought >process would begin to drift to other, unrelated topics, and would >require additional time to return to the current task once the output >finally began to be displayed. There was one internal to IBM that was used to justify a fast link between the building housing the mainframes and the building where the engineers were doing design work on their terminals. "Fitch's Ditch" for anyone who was there. -- Julian Thomas: jt . epix @ net http://home.epix.net/~jt remove letter a for email (or switch . and @) In the beautiful Finger Lakes Wine Country of New York State! Boardmember of POSSI.org - Phoenix OS/2 Society, Inc http://www.possi.org -- -- Double your drive space! Delete Windows! ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 67 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 15:22:53 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962637773 209.63.28.138 (Mon, 03 Jul 2000 09:22:53 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 09:22:53 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!newscore.gigabell.net!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59028 jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) writes: > Back in the 1970s there were several studies (which I can't cite by > name; it's been too many years) that examined the human engineering > of terminal use. One relatively consistent finding was that the time > between the dispatching of a request (typically by hitting the RETURN > or ENTER key) and the arrival of the first response character should > not exceed about 1.5 seconds because after that period the "typical" > user's thought process would begin to drift to other, unrelated topics, > and would require additional time to return to the current task once > the output finally began to be displayed. there was a presentation at SE symposium in early 70s held at 14th st. bridge marriott in wash. it basically measured perception of "instantaneous response" threshold (i.e. couldn't tell if there was delay) ... which turned out to vary from person to person (from about .1 second to about .25 second). there was joint report written around 1980 by some YKT research people and a San Jose GPD person about effect of system delays on people productivity (we had been working on system responses at .1 second versis a lot of people saying 1.0 seconds was good enuf). The research found that there was measurably increased productivity for system response down to .2 seconds ... and then there was less good correlation (possibly explained by the 1970 research). The research did find that variable system response had bad effect on people ... productivity was effected by approx. two times the variation. If the avg. response was .5 seconds and there was specific response that was 1.5 seconds ... then the person's attention would continue to wonder 1.5-0.5=1.0 seconds after the system responded. This not only applied to interactive activity but also things like batch compiles. If a compile would nominally take a minute ... but sometimes took 5 minutes ... a person's attention would continue to wonder for several minutes after the compile finished. That report may have been contorted into supporting agendas regarding not having to support better than 1.5 second system response. However the original research about attention wondering was with regard to variability & people's expectations ... aka if people expected system to respond in 0.5 seconds and it periodically took 1.5 seconds ... then their attention started to drift after the system didn't respond at the expected time and would continue to drift for period after the system did respond. It was still possible to have productivity improvement with .2 second response if the frequency of variability was kept to a tolerable level. random refs: http://www.garlic.com/~lynn/98.html#2 http://www.garlic.com/~lynn/98.html#46 http://www.garlic.com/~lynn/2000b.html#20 http://www.garlic.com/~lynn/2000b.html#25 guy from San Jose GPD had an article in IBM Systems Journal, v20n4, 1981, "Interactive User Productivity". -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 3 Jul 2000 15:29:05 GMT Organization: Neosoft (using Airnews.net!) Lines: 15 Message-ID: X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Mon Jul 3 10:29:05 2000 NNTP-Posting-Host: !`qT31k-XM1P:3u (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:58985 On Mon, 03 Jul 2000 14:51:49 GMT, Anne & Lynn Wheeler wrote: >the (original) 3272 controllers and 3277 terminals had the logic in >the heads with 3272 controller handling the channel interface. 3274's >moved head & keyboard logic back into the controller for 3278s & 3279s >terminals. Oh, THAT was the difference between type A and type B terminals! (I mean as attached to 3274s; you could attach a 3277 to a 3274, if you declared the port to be type B. I think. It's been nearly 20 years since I looked at that.) I never knew that. I always thought they were both dumb, just differently so. Personally, I didn't like 3277s all that much; I thought the keyboards were too big and clunky. Gimme a 3278-3. (The font on the 3278-4 looked weird.) ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> <3960aae8$1$wg$mr2ice@news.epix.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 97 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 15:58:22 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962639902 209.63.28.138 (Mon, 03 Jul 2000 09:58:22 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 09:58:22 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59031 jata@aepiax.net (Julian Thomas) writes: > There was one internal to IBM that was used to justify a fast link between > the building housing the mainframes and the building where the engineers > were doing design work on their terminals. when growth at STL outgrew available space, in '81 they relocated something like 300 people from the IMS group to a rented building about half-way between the main plant site and STL. At the time, they ran a WAN using a T3 collins digital radio from bldg 12 on the main plant site to bldg. 90 (STL) ... there was a repeater tower on a hill between 12 & 90. It also ran between bldg 12 and bldg 29 (Los Gatos VLSI lab) ... again with repeater tower on the hill above the san jose dump. Microwave was put in between bldg. 12 and the rented bldg. that the IMS group moved into. Because of the response issues, "remote 3270s" were not considered adequate for the IMS developers (i.e. 3274 controllers with 32 terminals running over 19.2 SDLC link). HYPERChannel was selected for the project. Basically A220 channel attach adapters, A510 remote device adapters, and A710 link adapters driving a T1 channel thru the campus WAN facility. A HYPERChannel A510 remote device adapter simulated a 370 channel and allowed attachments of 370 controllers. A HYPERChannel A220 was a 370 control box and attached to a real 370 channel. HYPERChannel ran a protocol over a 50mbit/sec. LAN. A HYPERChannel A710 link adapter would drive T1/T2 links and act as a bridge between HYPERChannel LANs. I wrote the support at the 370 end that "packaged" up 370 device channel programs and transmitted them to a remote A510 ... where they were executed remotely in the A510 remote device adapter (driving real 370 control units). This support provided subsecond response on "local" 3270 terminals for 300 people ... multiplexed over a single T1 (1.5mbit/sec link). There was a couple bugs encountered along the way. While I could handle the speed mismatch between multiple 640kbyte/sec 3270 controllers and the 1.5mbit/sec (approx. 150kbytes/sec) link ... it turns out that A710s had a bug and weren't really "full-duplex" controllers. The A710s really only operated in "half-duplex" mode and having outboard traffic from the 370 mainframes to the remote side at the same time as inbound traffic from controllers cause problems. They eventually built a real full-duplex link adapter, the A715 to address the problems. This configuration was replicated in Boulder when they moved the IMS field support team to a bldg. on the other side of I70. Because of regulations weren't able to use microwave between the roofs of the two buildings and so instead installed T1 infrared modems on polls on the two buildings. There was concern about signal quality with the infrared modems because of rain fade. However, the worst it got was a half dozen bit errors during a white-out snow storm when nobody was able to get into work. We did have a problem with the alignment of the infrared modems as the sun progressed across the sky during the day and heating different sides of the building resulting in building uneven building expansion which affected the modem alignment. random urls: http://www.garlic.com/~lynn/94.html#23 http://www.garlic.com/~lynn/94.html#24 http://www.garlic.com/~lynn/94.html#43 http://www.garlic.com/~lynn/94.html#55 http://www.garlic.com/~lynn/96.html#14 http://www.garlic.com/~lynn/96.html#27 http://www.garlic.com/~lynn/99.html#119 turns out that there was a side-effect of remoting the channel attached 3270s with HYPERChannel. After getting the 3270 controllers removed from directly attachment to the real 370 channels ... the thruput of the overall mainframe system increased by approx. 10% (in addition to showing to measurable degradation in system response). It turns out that the 3270 controllers (which operated at 640kbbyte/sec and also had slow cchannel bus handshake overhead) were sharing channels with disk controllers (which operated at 3mbytes/sec). Routing all 3270 traffic thru the A220 channel adapter significantly reduced 3270-related channel busy ... allowing measurable additional channel capacity for doing disk i/o. more random urls: http://www.garlic.com/~lynn/93.html#27 http://www.garlic.com/~lynn/96.html#15 http://www.garlic.com/~lynn/96.html#17 http://www.garlic.com/~lynn/98.html#34 http://www.garlic.com/~lynn/98.html#49 http://www.garlic.com/~lynn/98.html#50 http://www.garlic.com/~lynn/99.html#36 http://www.garlic.com/~lynn/99.html#123 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 27 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 16:32:02 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962641922 209.63.28.138 (Mon, 03 Jul 2000 10:32:02 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 10:32:02 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59023 Anne & Lynn Wheeler writes: > guy from San Jose GPD had an article in IBM Systems Journal, v20n4, > 1981, "Interactive User Productivity". the work that went into the above was also used to show that there was measurably higher productivity with subsecond response and that "remote" 327x controllers were not adequate (i.e. 3274 sdlc over 9.6 or 19.2 could not provide subsecond response time). side-benefits of remoting local 3274 & 3272 over T1 HYPERChannel was that getting those boxes directly off local 370 channels and improving aggregate system thruput by 10-15% because of increase disk i/o thruput (because local 3274 & 3272 controllers were no longer responsible for high channel busy interference). part of the problem at the time was you could only get 16 channels on a 370. there just weren't enuf available channels to dedicate specific channels to specific device types. there was a report for customers that did come out sometime after the HYPERChannel experience that did recommend (if the customer had available channels) to separate channels with disk controller attachments and 327x controller attachments. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 21 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 16:48:33 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962642913 209.63.28.138 (Mon, 03 Jul 2000 10:48:33 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 10:48:33 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59020 Anne & Lynn Wheeler writes: > side-benefits of remoting local 3274 & 3272 over T1 HYPERChannel was > that getting those boxes directly off local 370 channels and improving > aggregate system thruput by 10-15% because of increase disk i/o > thruput (because local 3274 & 3272 controllers were no longer > responsible for high channel busy interference). another problem was that for the first couple years ... the 3274s had a sporadic problem requiring them to be rebooted (IMPL) ... initially several times a day ... but eventually maybe only once a day. we found a trick ... that if you executed a HDV/CLRIO instruction sequence in a tight loop against every device (address) on a local attached 3274 ... it would reboot itself w/o requiring manual operater intervention. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: William Hamblen Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 03 Jul 2000 12:16:28 -0500 Organization: Utterly Disorganized Lines: 15 Message-ID: <9ug1ms03s820m1p5bk5539minu4ue9v5n4@4ax.com> References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> Reply-To: william.hamblen@nashville.com X-Complaints-To: newsabuse@supernews.com Cancel-Lock: sha1:D6vNeRV0JN6HrNf4SAL+X/1OnnM= X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-NFilter: 1.2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!t-online.de!news.netcologne.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!sn-xit-03!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59044 On Mon, 03 Jul 2000 08:15:46 GMT, kragen@dnaco.net (Kragen Sitaker) wrote: >Do people use them without controllers? Do the controllers live in a >separate box? When we had 3270s in the office they (and the printer) all talked to a controller (I don't recall which one and over the course of a few years we had two different ones), which talked to the mainframe over a leased line. The lone 3270/PC we had also talked to the controller. At first we used coax but after a move we installed twisted pairs with baluns at each end. Nowadays that office is long closed and field staff dial up our network and use a Java 3270 emulator applet with a web browser to talk to the mainframe. Who'd a thunk it? ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8j55im$9db$1@news.ox.ac.uk> <72dolsgeb9neq27ddftdncbf515llrjnra@4ax.com> <8pe75.8551$I96.261411@news-east.usenetserver.com> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 87 User-Agent: Gnus/5.0805 (Gnus v5.8.5) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 17:25:24 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962645124 209.63.28.138 (Mon, 03 Jul 2000 11:25:24 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 11:25:24 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59027 Anne & Lynn Wheeler writes: > another bullet for the s/390 roadmap was high speed queued I/O > ... which also brought back memories. I don't know if anything was > done with it but at the time we were doing fiber-channel cluster > scaleup ... another person and I wrote up patent application for high > speed queued I/O. When I mentioned that, the person doing the > presentation commented that sometimes it takes decades for technology > to show up in mainframes. work on high speed queued i/o (that i was initially involved in) was with a project called VAMPS in the 1975 time-frame. The 370 115/125 were done by the Boeblingen Germany lab. It was actually a shared memory bus with up to 9 microprocessors. The 115 had all microprocessors the same ... all with different microcode loads depending on the dedicated controller function ... telecommunication, disk, etc ... and one of the microprocessors had the 370 instruction set microcode load. The 125 was identical to the 115 except the microprocessor with the 370 microload was not the same as the others ... but something that ran about 50% faster. VAMPS was to build a 370/125 where there were between two to five microprocessors running the 370 instruction set microcode load in a shared memory configuration. To simplify development for a SMP VM/370 configuraiton, I dropped most of the SMP support into microcode engine ... and ran the remaining reduced function CP kernel on a single processor. This is somewhat analogous to the current LPAR/PRSM support. The reduced function CP kernel would place virtual machines on a pending dispatch queue ... and when it had no more work it went to the microcode dispatcher. For virtual machine work that couldn't be handled in the microcode, it would place an element on the pending work queue and attempt to enter kernel mode. If another processor was already in kernel mode, the processor would just go off and look for work on the dispatch queue. Basically this was a high-speed queued dispatch support. Since I also had the disk controller microcode, created a high speed queued I/O interface for the disk interface as well. Basically the abbreivated kernel put work requests on the disk controller queue. The disk controller pulled things off the queue (potentially out of order), did the work, and put the result on a pending queue. Interrupt in the abbreviated kernel could occur ... if there wasn't already a processor executing the abbreviated kernel. This was the original high speed queued I/O interface that I worked on. When VAMPs got killed, basically a "software" version of the VAMPs implementation was built for 158/168 SMP "attached processor" machines. This didn't have the high-speed queued i/o interface ... but did have work queue for running virtual machines ... and when a processor was no longer able to handle a virtual machine it attempted to get the global kernel lock. If it failed to get the global kernel lock, it queued the work request and attempted to run some other virtual machine. random refs: http://www.garlic.com/~lynn/94.html#21 http://www.garlic.com/~lynn/94.html#27 http://www.garlic.com/~lynn/94.html#28 http://www.garlic.com/~lynn/95.html#3 http://www.garlic.com/~lynn/2000c.html#30 A little bit of the high-speed queued i/o showed up in the HYPERChannel work ... although the hyperchannel was able to do out-of-order processing very well ... but since the CCW "package" was being loaded into the remote A510/A515 remote device adapters for channel simulation ... it was possible to do a little bit with I/O package operations on the A51x random refs: http://www.garlic.com/~lynn/94.html#23 http://www.garlic.com/~lynn/94.html#24 http://www.garlic.com/~lynn/96.html#14 http://www.garlic.com/~lynn/96.html#27 The VAMPs high-speed queued i/o, the HYPERChannel experience, and the HSDT project experience (High-speed data transport) all contributed to the work on the high speed queued i/o interface for fiber channel adapters. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### From: John Hendrickx Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 3 Jul 2000 20:02:16 +0200 Organization: The University of Nijmegen, The Netherlands Lines: 10 Message-ID: References: <8joosu$p0h@netaxs.com> <8jq0ga$7pf$3@bob.news.rcn.net> NNTP-Posting-Host: catv8249.extern.kun.nl X-Trace: odysseus.uci.kun.nl 962647348 12417 131.174.118.249 (3 Jul 2000 18:02:28 GMT) X-Complaints-To: Tom.Adriaansen@uci.kun.nl NNTP-Posting-Date: 3 Jul 2000 18:02:28 GMT X-Newsreader: MicroPlanet Gravity v2.30.1784 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newsfeeds.belnet.be!news.belnet.be!news2.kpn.net!news.kpn.net!surfnet.nl!surfnet.nl!odysseus.uci.kun.nl!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59041 In article <8jq0ga$7pf$3@bob.news.rcn.net>, jmfbahciv@aol.com says... > Only when the dumb terminals were the only devices available. > There were "smart" terminals, too. And, as these devices What exactly is a "smart" terminal? All the processing is being done by the mainframe, what would the terminal add to the user experience? As a student, I considered a Tektronics 4207 graphics terminal to be more or less my personal property. It was basically the same as a 3270, but in colour (!) and if you did any graphics you could zoom in or out, it stored the graphics as vectors apparently. Would that count as "smart"? ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <8jq0ga$7pf$3@bob.news.rcn.net> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 29 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <962650532.276054@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-6.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Mon, 3 Jul 2000 14:53:43 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 962650364 204.92.64.17 (Mon, 03 Jul 2000 14:52:44 EDT) NNTP-Posting-Date: Mon, 03 Jul 2000 14:52:44 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sunqbc.risq.qc.ca!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59004 "John Hendrickx" wrote in message news:MPG.13caf391e341b2119896dc@news.uci.kun.nl... > In article <8jq0ga$7pf$3@bob.news.rcn.net>, jmfbahciv@aol.com says... > > Only when the dumb terminals were the only devices available. > > There were "smart" terminals, too. And, as these devices > > What exactly is a "smart" terminal? All the processing is being done by > the mainframe, what would the terminal add to the user experience? As a Well, that's the point actually; _not_ all the processing is done by the mainframe: terminals range from TTY type, through basic screen commands (bold, u/l, clear screen, split screen); more "intelligent" terminals have fields, scrolling, input verification for data entry at the datatype level: the brightest ones are PCs where all non-mainframe specific processing is accomplished. > student, I considered a Tektronics 4207 graphics terminal to be more or > less my personal property. It was basically the same as a 3270, but in > colour (!) and if you did any graphics you could zoom in or out, it > stored the graphics as vectors apparently. Would that count as "smart"? on the graphics oriented "idiot savante" side: imagine the mainframe load if it had to do all the work. Rick ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 03 Jul 00 22:58:36 -0800 Organization: http://extra.newsguy.com Lines: 13 Message-ID: <408.219T1831T13785252@sky.bus.com> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> NNTP-Posting-Host: p-581.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!news2 Xref: chonsp.franklin.ch alt.folklore.computers:59068 In article kragen@dnaco.net (Kragen Sitaker) writes: >Were the I/O devices capable of DMA? Say, there's another possible definition of "mainframe". You never hear the term "DMA" in a mainframe environment, simply because it's been taken for granted for decades. -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 03 Jul 00 13:39:37 -0800 Organization: http://extra.newsguy.com Lines: 49 Message-ID: <2522.219T1898T8195583@sky.bus.com> References: <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> NNTP-Posting-Host: p-357.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!208.171.248.21.MISMATCH!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!news1 Xref: chonsp.franklin.ch alt.folklore.computers:59071 In article lynn@adcomsys.net (Anne & Lynn Wheeler) writes: >jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) writes: > >> Back in the 1970s there were several studies (which I can't cite by >> name; it's been too many years) that examined the human engineering >> of terminal use. One relatively consistent finding was that the time >> between the dispatching of a request (typically by hitting the RETURN >> or ENTER key) and the arrival of the first response character should >> not exceed about 1.5 seconds because after that period the "typical" >> user's thought process would begin to drift to other, unrelated >> topics, and would require additional time to return to the current >> task once the output finally began to be displayed. I remember reading such a report in the early '80s. The test was done at a live site that added enough hardware to get the response time down below a second. After the test they decided to keep the extra hardware because it was paying for itself in increased productivity. >The research did find that variable system response had bad effect >on people ... productivity was effected by approx. two times the >variation. If the avg. response was .5 seconds and there was specific >response that was 1.5 seconds ... then the person's attention would >continue to wonder 1.5-0.5=1.0 seconds after the system responded. >This not only applied to interactive activity but also things like >batch compiles. If a compile would nominally take a minute ... but >sometimes took 5 minutes ... a person's attention would continue >to wonder for several minutes after the compile finished. > >That report may have been contorted into supporting agendas regarding >not having to support better than 1.5 second system response. However >the original research about attention wondering was with regard to >variability & people's expectations ... aka if people expected system >to respond in 0.5 seconds and it periodically took 1.5 seconds >... then their attention started to drift after the system didn't >respond at the expected time and would continue to drift for period >after the system did respond. Sperry, in their inimitable way, took their own approach to the problem. I once heard an 1100 person (I worked on the OS/3 side) say that their solution was to hold back quick responses - so in the above example you'd get a consistent 1.5-second response time. -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 3 Jul 2000 19:53:10 GMT Organization: The MITRE Corporation Lines: 14 Message-ID: <8jqqv6$etb$1@top.mitre.org> References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 962653990 15275 128.29.251.13 (3 Jul 2000 19:53:10 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 3 Jul 2000 19:53:10 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!newspeer.clara.net!news.clara.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:59076 Anne & Lynn Wheeler writes: >we found a trick ... that if you executed a HDV/CLRIO instruction >sequence in a tight loop against every device (address) on a local >attached 3274 ... it would reboot itself w/o requiring manual >operater intervention. If this was ever documented as a bug, it should have been described as a remote re-IMPL technique -- and would be another proof of the old joke that "a feature is a bug that's been documented." Out of curiosity: how was that trick discovered? Joe Morris ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> <8jqqv6$etb$1@top.mitre.org> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 55 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 21:57:48 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962661468 209.63.28.138 (Mon, 03 Jul 2000 15:57:48 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 15:57:48 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!newsfeed.yosemite.net!cyclone-transit.snfc21.pbi.net!216.218.192.242!news.he.net!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59097 jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) writes: > If this was ever documented as a bug, it should have been described as > a remote re-IMPL technique -- and would be another proof of the old > joke that "a feature is a bug that's been documented." it really was a feature; both HDV & CLRIO instructions were kernel mode only instructions ... so you wouldn't have application programs capable of doing it ... and the loop in the kernel to do it had to be specifically programmed. These were things put into the code by the engineering groups and didn't normally make it into the documentation &/or operating system (you had to go spend time with the engineers to find out little tricks they used while they may have been developing the boxes). I was doing a bullet proof i/o supervisor for the disk engineering lab. they had all these test cells (disk stuff inside steel cages ... about 4' sq & 7' high ... with combination padlocks ... inside a secure room, inside a secure building, etc). since they were stuff under development the boxes didn't alwas behave as specified in the manual ... but they still needed to do testing connected to a mainframe. A room might have 2-3 different mainframes and a dozen or so test cells that were undergoing development & test. At the time, if MVS was ipl'ed on one of the mainframes while test cells were connected and doing anything ... mean-time-between mvs crash was about 15 minutes. As a result, the guys on the test cells needed dedicated machine time ... running a small, dedicated real-time monitor. The objective of making the I/O supervisor absolutely bullet proof ... so no matter what I/O things happened ... the I/O supervisor wouldn't contribute to any sort of system failure. This would allow the various mainframes to all be run with operating systems ... performing a number of useful services while concurrently supporting testing of multiple test cells. other components had similar reboot features. It was possible to get a 303x channel director to reboot if all six channels were hit in a tight loop with CLRCH instruction. random refs: http://www.garlic.com/~lynn/95.html#3 http://www.garlic.com/~lynn/96.html#18 http://www.garlic.com/~lynn/96.html#27 http://www.garlic.com/~lynn/97.html#15 http://www.garlic.com/~lynn/99.html#31 http://www.garlic.com/~lynn/99.html#98 http://www.garlic.com/~lynn/2000.html#78 -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 82 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Mon, 03 Jul 2000 18:24:26 EDT Date: Mon, 03 Jul 2000 22:24:26 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59129 In article <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net>, Jay Maynard wrote: >On Mon, 03 Jul 2000 08:15:46 GMT, Kragen Sitaker wrote: >>Do people use them without controllers? Do the controllers live in a >>separate box? > >Yes, and yes, respectively. The 3274 and 3174 controllers were separate boxes >with some amount of intelligence, capable of driving from 8 to 32 terminals >depending on the model. Some 3270s were attached directoy to dedicated >console ports on the machine, although I don't know how much of a 3x74 they >stuck in there to drive it. So in most cases, the controller was not part of the mainframe, but for the console, it was. >>This is a huge advantage for the computer (over a machine handling >>VT100s). Instead of handling five or so interrupts per second per >>terminal, it handles one every fifteen or twenty seconds at worst. >>Instead of having to respond within 200 milliseconds in order not to >>feel slow, it needs to respond within a second or two. > >Exactly. This is part of the mainframe philosophy: offload work when you >can. Mainframers talk about "subsecond response time" as the Holy Grail of >tuning. If the user can make a trivial request, hit ENTER, and get his >keyboard unlocked in under a second, life is good. So, for an apples-to-apples comparison, an http server with 500 users is much closer to 500 3270 users than, say, 500 telnet sessions is, right? >>That's why I brought up web browsers. It may be that a modern Linux >>box can handle 500 telnet sessions, but that is a much more impressive >>feat than handling 500 sessions of stuff like the previous paragraph, >>which a modern Linux box like the one being suggested three posts ago >>can handle without even noticing. > >I'm not so sure. Can it handle those 500 sessions doing significant database >work at one transaction every 5 seconds or so? You start running out of disk >bandwidth in a hurry. Yes, real users - experienced ones doing the same >small number of tasks repetitively - can generate that kind of transaction >rate. That's 100 TPS, if I understand you correctly. http://www.zdnet.com/eweek/stories/general/0,11011,2436153,00.html benchmarked a couple of databases on a 450MHz Pentium II with 512MB of RAM and a RAID on Linux; Interbase got about 25 "TPS" with 100 simulated users; PostgreSQL got about 70. The code to do the benchmark is on the web site, although I doubt it's any standard benchmark; information about the RAID configuration, clearly the most important factor, is not. I think Oracle and Informix are higher-performance databases than PostgreSQL or Interbase. http://www.cab.u-szeged.hu/local/linux/text/linux-oracle.txt is an amateur benchmark of SCO Oracle running under emulation on Linux on a 486DX2-66 with 16MB of RAM on a Conner SCSI CFP-1060S disk. It reported 25-60 "TPS" on this machine using homebrew tests. (This was in 1995.) Various other web pages claim thousands and even tens of thousands of "TPS" on four and six-way machines running Linux. Searching on tpc.org shows no TPC-C benchmarks for Linux. My guess, then, is that it is possible to handle 100 TPS on a fairly cheap PC; I don't know about "without even noticing". >>"output far in excess of one visible page" is the only one of these >>that seems like a significant advantage for server performance to me. > >Maybe, but that doesn't make the browser any less intelligent. Right --- what I said was completely wrong. I was thinking only about how things look from the server's end. (Actually, even from that POV, the browser is a bit smarter. It has a Back button.) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 26 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Mon, 03 Jul 2000 18:28:46 EDT Date: Mon, 03 Jul 2000 22:28:46 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-lei1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.onemain.com!feed1.onemain.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59130 In article , Anne & Lynn Wheeler wrote: >At the time, if MVS was ipl'ed on one of the mainframes while test >cells were connected and doing anything ... mean-time-between mvs >crash was about 15 minutes. As a result, the guys on the test cells >needed dedicated machine time ... running a small, dedicated real-time >monitor. > >The objective of making the I/O supervisor absolutely bullet proof >... so no matter what I/O things happened ... the I/O supervisor >wouldn't contribute to any sort of system failure. This would allow >the various mainframes to all be run with operating systems >... performing a number of useful services while concurrently >supporting testing of multiple test cells. Were the I/O devices capable of DMA? It seems that if the disk was DMAing to the wrong physical memory, it would be very difficult to avoid system failure. (The I/O supervisor wouldn't even need to contribute. :) I guess I should go read about mainframe channels. -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### Sender: lynn@LYNNPC Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 64 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 03 Jul 2000 23:11:19 GMT NNTP-Posting-Host: 209.63.28.138 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962665879 209.63.28.138 (Mon, 03 Jul 2000 17:11:19 MDT) NNTP-Posting-Date: Mon, 03 Jul 2000 17:11:19 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59094 kragen@dnaco.net (Kragen Sitaker) writes: > Were the I/O devices capable of DMA? It seems that if the disk was > DMAing to the wrong physical memory, it would be very difficult to > avoid system failure. (The I/O supervisor wouldn't even need to > contribute. :) devices weren't capable of DMA'ing directly into memory ... all I/O went thru the channel hardware interface where the channel hardware controlled the address & length of data transfer. the channel, controller & device architecture was tightly controlled and including specifications about things like short busy, interrupts in specific sequences, error indicators stored in specific ways. flaky devices could present "short busy" ... where the proscribed processor was to immediately retry the operation ... however the short busy might never clear ... and the processor might go into a tight loop retrying the operation. a flacky device also might forget to provide an interrupt at all. a standard device was proscribed at only presenting an error interrupt in conjunction with on ongoing operation. a flaky device might present an error interrupt out of the blue ... not in conjunction with anything else. operating system could hang because of tight loop or missing interrupts (not real crash ... but something requiring manual reboot to clear the conditionq). Another situation was hot interrupts which could exhaust system resources and lead to system failure because of running out of storage (similar but different to some of the internet denial of service attacks which exhausted system resources and resulted in system crash). Hot interrupts might need both device and channel fenching to get the system back under control. Other situations were simple software logic bugs in the kernel where the device specifications had the software design operating in a specific way because the device alwas worked in a specific way. When the device operated in a very anomolous manner ... it would expose a software bug in the kernel. There could be hundreds of these conditions every day that had never been seen in a production data processing operation (or possibly less than once or twice per year across the whole machine install base of thousands & thousands of mainframes). the situation was slightly aggrevated by the fact that the mainframe operating system had extensive error retry and error recovery routines ... in some cases 10-100 times the amount of error retry & recovery code compared to typical non-mainframe operating systems. As a result there was a lot more code that could have possible logic failures when exposed to extremely anomolous & unanticipated modes of operation. These weren't production devices ... they were engineering hardware in the process of development and test. There were hundreds of anomolous conditions that never had been experienced in normal production operation. Rather than a controlled, data processing environment operating according to some standard ... the disk engineering development and test environment was an extremely hostile data processing opreation where severe error and anomolous operation was the rule instead of the exception. -- Anne & Lynn Wheeler | lynn@adcomsys.net, lynn@garlic.com http://www.garlic.com/~lynn/ http://www.adcomsys.net/lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 60 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Mon, 03 Jul 2000 20:18:59 EDT Date: Tue, 04 Jul 2000 00:18:59 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc01.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59123 In article , Mike Meredith at home wrote: >In article , > kragen@dnaco.net (Kragen Sitaker) writes: >> At these prices, a PC starts to look pretty attractive as a terminal >> replacement :) > >Rubbish. A PC does more than a terminal (but not an awful lot >more), but has a number of disadvantages :- > >* It's noisy. Any PC with a fan in is noisy enough to irritate > anyone who is used to a silent office --- we had a number of > complaints from people who hated their new PC's because of the > noise. And in a number of cases we had to replace their PC's > with old terminals. True. I'd really like to get a NetWinder or some other fanless PC for my bedroom. >* It's less reliable. A terminal's only moving parts is the > keyboard and they just don't fail very often. We've got old > Volker Craigs serving duty as Sun consoles that must be 15 > years old, and apart from looking a bit rough are still working > fine. If you get rid of the fan, the only remaining moving parts are disk drives. Burn a $20 Etherboot ROM for each PC, and you can get rid of those, too. And, while there are surely terminals that have been in constant use for 15 years, there are many, many more that crapped out years ago. Moving parts fail, but non-moving parts do, too. $5 PC keyboards fail frequently, by the way --- they have a lifetime of perhaps three years. >* They're harder to maintain. You can train a money to fix the > common problems with terminals --- freezing the screen with > Control-S, user entering Setup and fiddling with the settings, > etc. Fixing a broken PC setup is much harder. Yes you can swap > a PC out (and explain to the user that they can't use their > Zip/Jazz/etc drives because the spares don't have them) and fix > it back at base, but it still takes time. > >I think we're paying somewhere in the region of $1,000 for our >standard PC (some apps won't work with a minimum spec), so $200 >for a PC replacement looks quite good (actually we'd need >Etherterminals or similar which are more like $500). It sounds like you're not using your PCs as terminals; instead, you are using them as computers. I was talking about using them as terminals. It's interesting that Etherterminals cost more than computers set up as netbooting X-terminals. What, exactly, is an Etherterminal? A terminal that can telnet or LAT, or something else? :) -- Kragen Sitaker The Internet stock bubble didn't burst on 1999-11-08. Hurrah! The power didn't go out on 2000-01-01 either. :) ###### From: hmv@port.ac.uk (Mike Meredith at home) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 3 Jul 2000 19:02:58 -0700 Organization: Customer of Planet Online Lines: 83 Sender: mike@port.ac.uk Message-ID: References: <8joosu$p0h@netaxs.com> <8jq0ga$7pf$3@bob.news.rcn.net> NNTP-Posting-Host: modem-14.baranduin.dialup.pol.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: newsg3.svr.pol.co.uk 962652666 25100 62.136.135.14 (3 Jul 2000 19:31:06 GMT) NNTP-Posting-Date: 3 Jul 2000 19:31:06 GMT X-Complaints-To: abuse@theplanet.net X-Newsreader: knews 1.0b.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!news.theplanet.net!newspost.theplanet.net!lucifer!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59062 In article <8jq0ga$7pf$3@bob.news.rcn.net>, jmfbahciv@aol.com writes: > In article <8joosu$p0h@netaxs.com>, lwinson@bbs.cpcn.com (lwin) wrote: >>Remember, too, the terminals were all "dumb", all formatting >>and the like was handled by the mainframe. > > Only when the dumb terminals were the only devices available. > There were "smart" terminals, too. And, as these devices > were developed mainframe OS development began to support > them. In fact there were places experimenting with very intelligent terminals in the late 1970's --- at the University of Portsmouth there were terminals that could run BASIC programs, edit files locally, and use the central ICL 19xx as a file server (over a serial line). And I'm sure we won't the only ones. >>I'm not sure what a "small PC server" would represent, but would >>it have adequate I/O to handle that, esp with dumb cheap >>terminals? I think you could probably spec up a PC server to do it, but it wouldn't be small, and it would probably cost about the same as say a loaded Sun E450 (and I'd rather go for the Sun). > That's not the key, IMO. Memory protection would be much more > important. In order to do mainframe work, the computer would > have to be able to prevent a user from trashing another > users (or process) address space. As long as the PC is running a sensible operating system (like Unix), there would be no problem with memory protection. A more serious concern would be reliability. I've run PC server systems running Unix since 1990, and whilst sometimes you get rock solid systems that run 24x7 for a year on end, sometimes you get complete lemons. And you can't predict what kind of box you will get --- I've had lemons from big name manufacturers, and great boxes from no-name box shifters (and the reverse). In the same length of time, I've seen roughly twice as many VAXen, Alpha's, Sun's, and HP boxen go past. They're a lot more reliable, and when you do have a problem, the support is better. >>Today those are run on PCs. The mainframe >>wasn't purchased for them, but was used none the less. > > And now the management of those systems is a nightmare. > For instance, nstead of one full system backup, the admins > now have 1000s of full system backups if they really want > to take care of their stable of users. From what I can gather (I work in the same office as the poor sods that run the PC network, but don't have any official involvement), the big problem is maintaining the PC's (installing new o/s releases, patches, upgrading applications, removing unauthorised applications, etc) rather than backing up. Of course we do backups properly --- stuff on the fileservers is backed up, and we regard the PC hard drives as temporary storage. > What has been happening is that the individual user has > now become responsible for operations, data integrity, > hardware maintenance, software maintenance, etc, etc, etc,. Some do it, and do it properly (or at least it works afterwards). Some do it, and make a complete mess of it. And blame the support staff afterwards. Some ask support staff to do it. And it costs *far* more to provide proper support to 1,000 PC users than to 1,000 terminal users (even X-Terminals). Of course most places don't do proper support. > One of the advantages of a mainframe environment is that > all of these functions were invisible to the user, allowing > him/her to concentrate on their work rather than on the > longevity of their work. Apart from the downtime :) > Have you done your backups today? :-)) Yup, including the two trips to the off-site firesafe. Mind you I'm only filling in until the operators return :) ###### From: hmv@port.ac.uk (Mike Meredith at home) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 3 Jul 2000 19:25:33 -0700 Organization: Customer of Planet Online Lines: 40 Sender: mike@port.ac.uk Message-ID: References: <8joosu$p0h@netaxs.com> NNTP-Posting-Host: modem-14.baranduin.dialup.pol.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: newsg3.svr.pol.co.uk 962652668 25100 62.136.135.14 (3 Jul 2000 19:31:08 GMT) NNTP-Posting-Date: 3 Jul 2000 19:31:08 GMT X-Complaints-To: abuse@theplanet.net X-Newsreader: knews 1.0b.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!news.theplanet.net!newspost.theplanet.net!lucifer!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59065 In article , kragen@dnaco.net (Kragen Sitaker) writes: > When you say "dumb cheap terminals", what do you mean by "dumb"? A > 3270, which is roughly as smart as a Web browser without JavaScript? > And what do you mean by "cheap"? http://www.mticom.com/wyse.htm lists > used WYSE 30 terminals each with a 90-day warranty at $130, WYSE 50s > with a one-year warranty at $190 each, and color WYSE 370s with a > one-year warranty for $435. Used. Used 3270s apparently cover a > slightly wider range on both ends. > > At these prices, a PC starts to look pretty attractive as a terminal > replacement :) Rubbish. A PC does more than a terminal (but not an awful lot more), but has a number of disadvantages :- * It's noisy. Any PC with a fan in is noisy enough to irritate anyone who is used to a silent office --- we had a number of complaints from people who hated their new PC's because of the noise. And in a number of cases we had to replace their PC's with old terminals. * It's less reliable. A terminal's only moving parts is the keyboard and they just don't fail very often. We've got old Volker Craigs serving duty as Sun consoles that must be 15 years old, and apart from looking a bit rough are still working fine. * They're harder to maintain. You can train a money to fix the common problems with terminals --- freezing the screen with Control-S, user entering Setup and fiddling with the settings, etc. Fixing a broken PC setup is much harder. Yes you can swap a PC out (and explain to the user that they can't use their Zip/Jazz/etc drives because the spares don't have them) and fix it back at base, but it still takes time. I think we're paying somewhere in the region of $1,000 for our standard PC (some apps won't work with a minimum spec), so $200 for a PC replacement looks quite good (actually we'd need Etherterminals or similar which are more like $500). ###### From: genew@shuswap.net (Gene Wirchenko) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 04 Jul 2000 05:12:58 GMT Organization: Okanagan Internet Junction Lines: 28 Message-ID: <39617206.34902445@news.shuswap.net> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> Reply-To: genew@shuswap.net NNTP-Posting-Host: salmonarm3-04.shuswap.net X-Trace: news.junction.net 962687478 5058 139.142.177.134 (4 Jul 2000 05:11:18 GMT) X-Complaints-To: abuse@junction.net NNTP-Posting-Date: 4 Jul 2000 05:11:18 GMT X-Newsreader: Forte Free Agent 1.1/32.230 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cyclone.bc.net!news.junction.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59082 Anne & Lynn Wheeler wrote: [snip] >operating system could hang because of tight loop or missing >interrupts (not real crash ... but something requiring manual reboot >to clear the conditionq). Another situation was hot interrupts which ^^^^^^^^^^ What is this, please? >could exhaust system resources and lead to system failure because of >running out of storage (similar but different to some of the internet >denial of service attacks which exhausted system resources and >resulted in system crash). Hot interrupts might need both device and >channel fenching to get the system back under control. ^^^^^^^^ What is this, please? [snip] Sincerely, Gene Wirchenko Computerese Irregular Verb Conjugation: I have preferences. You have biases. He/She has prejudices. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 04 Jul 2000 00:46:08 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: <9nsvls45r9g4c9makfr8u34bf6ho2m5qmp@4ax.com> References: <395D6E84.DB233C22@cmc.com> <395D79C0.4B72BBE8@cmc.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.137.17 X-Original-NNTP-Posting-Host: 207.148.137.17 X-Trace: 4 Jul 2000 00:46:09 -0700, 207.148.137.17 Lines: 56 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!nntp.flash.net!nntp.cadvision.com!news.cadvision.com!207.148.137.17 Xref: chonsp.franklin.ch alt.folklore.computers:59117 On Sat, 01 Jul 2000 05:13:01 GMT, kragen@dnaco.net (Kragen Sitaker) wrote: >In article <395D79C0.4B72BBE8@cmc.com>, Lars Poulsen wrote: >>Kragen Sitaker wrote: >>> . . . but now "ASPs" --- application service providers, the new >>> name for service bureaus --- are a hot buzzword. Companies like >>> Interliant are growing quite rapidly, despite bleeding megabucks. >> >>The statement was that the service bureaus that provided jobs for >>MVS system programmers are going away. ASPs don't propose to run >>MVS ... > >Don't they? I thought some of them did. . . > >>Also, it will take a while before we find out if "there is a pony" >>in the ASP business. > >True enough. > >>> Even in the real world, outside of businesses, many people use >>> huge "service bureaus" like Google, AltaVista, Yahoo, and Hotmail >>> to provide many of their computing services. >> >>And none of those charge the users for services rendered ... > >Most don't. I use pobox.com to handle my email, though, which does. I >suspect that most of the free services run by for-profit companies will >either have to resort to outright villainy (as AltaVista, for example, >has --- selling their users' search terms to DoubleClick! For shame!) >or fail. > >Google is an interesting case; they make money by selling their search >service to Web sites for a bit less than a penny (down to about 0.6 >cents) per query, plus a bit of prostitution (selling ads), which is >probably doomed in the long run. Their free service is a way to build >a good brand so people (like, e.g., Yahoo) will want to buy their >service. > >>But indeed, some of those are good candidates for mainframes. >>One particular kind of profitable ASP is the web hosting facility. >>An S/390 with 40,000 instances of Linux makes one heck of a hosting >>facility. Still no MVS, though. > >I know an S/390 has run 40,000 instances of Linux, but I think none of >them were running anything. My experience with the ratio of RAM >required to boot Linux and to do something useful with it on PCs >suggests that 1,000 to 5,000 might be a more reasonable number. IIRC from the announcement (at work), they were each running an Apache web server. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 4 Jul 2000 12:56:21 GMT Organization: Neosoft (using Airnews.net!) Lines: 17 Message-ID: X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Tue Jul 4 07:56:21 2000 NNTP-Posting-Host: !`5mB1k-W-1_tlE (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!europa.netcrusader.net!209.113.65.250!korova.insync.net!solomon.io.com!nntp.giganews.com!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:59066 On Tue, 04 Jul 2000 05:12:58 GMT, Gene Wirchenko wrote: >>to clear the conditionq). Another situation was hot interrupts which > ^^^^^^^^^^ > What is this, please? Tsk, Gene...spelling flames are lame. >>channel fenching to get the system back under control. > ^^^^^^^^ > What is this, please? This is another typo, but this time for a term that you may not have heard of: "fencing". Fencing a channel, or a device, on a 370/390 means to disable it as much as possible, and then to ignore any further I/O interrupts from it; I don't know how this is done at the device level, but at the channel level, the I/O interrupt mask is set to disable further interrupts from that channel. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 04 Jul 2000 16:52:39 GMT Lines: 18 Message-ID: <39621657$0$8303$3c090ad1@news.plethora.net> NNTP-Posting-Host: b6f17a9f.news.plethora.net X-Trace: 962729559 gemini.plethora.net 8303 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!feeder.via.net!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59102 In article <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net>, Jay Maynard wrote: >A web browser has, in addition to the ability to act like a dumb terminal, >the ability to use many different fonts, versatile special formatting, >handle output far in excess of one visible page and navigate through it, and >the ability to figure out what's being sent and deal with it appropriately >in the current environment, just off the top of my head. All of these are >far beyond the 3270's capabilities. Well, I don't know about the fonts; I've seen lots of web browsers that don't display "fonts", don't "navigate" through pages, and that otherwise fail to comply with this description. -s -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### From: wiss@eelwing.arda (Jonas Wissting) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 4 Jul 2000 19:29:30 +0200 Organization: Utfors AB Lines: 29 Message-ID: References: <8joosu$p0h@netaxs.com> NNTP-Posting-Host: md4690540.utfors.se Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: yggdrasil.utfors.se 962732651 15440 212.105.5.64 (4 Jul 2000 17:44:11 GMT) X-Complaints-To: abuse@utfors.se NNTP-Posting-Date: 4 Jul 2000 17:44:11 GMT X-Newsreader: knews 1.0b.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!serra.unipi.it!hammer.uoregon.edu!news.algonet.se!algonet!nntp.se.dataphone.net!newsfeed.sunet.se!news01.sunet.se!news.utfors.se!luthien.arda!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59070 In article , hmv@port.ac.uk (Mike Meredith at home) writes: ... > * They're harder to maintain. You can train a money to fix the > common problems with terminals --- freezing the screen with > Control-S, Pleace can someone explain this C-s? How do I reset a xterm after hitting C-s? it is to close to C-d. Jonas -- http://wiss.unx.nu http://linux.unx.nu Another Glitch in the Call We don't need no indirection We don't need no flow control No data typing or declarations Did you leave the lists alone? Hey! Hacker! Leave those lists alone! Chorus: All in all, it's just a pure-LISP function call. All in all, it's just a pure-LISP function call. ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 04 Jul 2000 21:31:53 +0200 Organization: My own Private Self Lines: 24 Message-ID: <6uituladzq.fsf@chonsp.franklin.ch> References: <8joosu$p0h@netaxs.com> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 962739114 647 10.0.3.2 (4 Jul 2000 19:31:54 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 4 Jul 2000 19:31:54 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:59132 wiss@eelwing.arda (Jonas Wissting) writes: > In article , > hmv@port.ac.uk (Mike Meredith at home) writes: > ... > > * They're harder to maintain. You can train a money to fix the > > common problems with terminals --- freezing the screen with > > Control-S, > > Pleace can someone explain this C-s? Ctrl-S, ASCII DC* (Device Control 3), make computer pause sending data to the terminal. > How do I reset a xterm after > hitting C-s? it is to close to C-d. No need to reset, just unblock. For this use Ctrl-Q, ASCII DC1. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 4 Jul 2000 18:25:33 GMT Organization: Neosoft (using Airnews.net!) Lines: 8 Message-ID: <3839E0A647FD4215.62BAC440B58C07E6.FDAEA7FEEB5401C6@lp.airnews.net> X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <39621657$0$8303$3c090ad1@news.plethora.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Tue Jul 4 13:25:33 2000 NNTP-Posting-Host: !aIN,1k-XA30/uE (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!serra.unipi.it!hammer.uoregon.edu!newsfeed.direct.ca!look.ca!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:59142 On 04 Jul 2000 16:52:39 GMT, Peter Seebach wrote: >Well, I don't know about the fonts; I've seen lots of web browsers that don't >display "fonts", don't "navigate" through pages, and that otherwise fail >to comply with this description. Lynx doesn't do fonts, true...what others are you thinking of that don't, say, navigate through pages? They would seem to be pretty useless as web browsers... ###### Message-ID: <39623E02.78E9C854@thinkage.ca> From: "Alan T. Bowler" Organization: Thinkage Ltd. X-Mailer: Mozilla 4.5 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 23 Date: Tue, 04 Jul 2000 15:41:54 -0400 NNTP-Posting-Host: 192.102.11.4 X-Trace: nnrp1.uunet.ca 962739827 192.102.11.4 (Tue, 04 Jul 2000 15:43:47 EDT) NNTP-Posting-Date: Tue, 04 Jul 2000 15:43:47 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.cwix.com!prairie.attcanada.net!newsfeed.attcanada.net!142.77.1.188!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59134 Kragen Sitaker wrote: > > Were the I/O devices capable of DMA? It seems that if the disk was > DMAing to the wrong physical memory, it would be very difficult to > avoid system failure. (The I/O supervisor wouldn't even need to > contribute. :) Devices did not do DMA. A channel is essentially a piece of hardware that does shared DMA for many devices. That way you have a common interface, and since it it shared you can can afford to have extra hardware that understands paging, bounds and other forms memory protection. Having a separate IO/DMA processor like this can offload a lot of work from a CPU, and avoids quirks that crop up if every device/controller team tries to rengineer DMA processing. (The terminology varies a bit for different mainframes. In some a "channel" is one of these shared IO/DMA processors, in other mainframes you have a one or more of these IO processors and each supports many "channels"). Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line was something of an exception. However it eventually evolved something equivalent.) ###### From: als@thangorodrim.de (Alexander Schreiber) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 4 Jul 2000 19:59:57 GMT Organization: Home of the BOFH Lines: 24 Approved: May you live in interesting times. Message-ID: References: <8joosu$p0h@netaxs.com> Reply-To: als@usenet.thangorodrim.de NNTP-Posting-Host: bofh.csn.tu-chemnitz.de X-Trace: narses.hrz.tu-chemnitz.de 962740797 6292 134.109.108.7 (4 Jul 2000 19:59:57 GMT) X-Complaints-To: usenet@news.tu-chemnitz.de NNTP-Posting-Date: 4 Jul 2000 19:59:57 GMT X-Newsreader: slrn (0.9.5.2 UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeeds.belnet.be!news.belnet.be!uni-erlangen.de!news.tu-chemnitz.de!als Xref: chonsp.franklin.ch alt.folklore.computers:59169 On Tue, 4 Jul 2000 19:29:30 +0200, Jonas Wissting wrote about Re: Does the word "mainframe" still have a meaning?: >In article , > hmv@port.ac.uk (Mike Meredith at home) writes: >... >> * They're harder to maintain. You can train a money to fix the >> common problems with terminals --- freezing the screen with >> Control-S, > >Pleace can someone explain this C-s? How do I reset a xterm after >hitting C-s? it is to close to C-d. Ctrl-S and Ctrl-Q are terminal control handshake. Ctrl-S ''stops'' the terminal (for instance for reading text that scrolls by too fast) and Ctrl-Q ''unstops'' it again (for instance if you are done reading whats on the screen). Regards, Alex. -- ------------------------------------------------------------------------------ EMail : als@thangorodrim.de | WWW : http://www.thangorodrim.de/ If privacy is outlawed, only outlaws will have | Ceterum censeo Parva Mollia privacy. (Philip Zimmerman, author of PGP) | esse delendam. ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 4 Jul 2000 20:15:58 GMT Organization: The National Capital FreeNet Lines: 10 Message-ID: <8jtglu$ft6$1@freenet9.carleton.ca> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 962741758 16294 134.117.136.30 (4 Jul 2000 20:15:58 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 4 Jul 2000 20:15:58 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.stanford.edu!news.kjsl.com!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59159 "Alan T. Bowler" (atbowler@thinkage.ca) writes: > ... > Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line > was something of an exception. However it eventually evolved > something equivalent.) RD (Read Direct), WD (Write Direct) IBM System 360 Principles of Operations and follow-ons. Supervisor mode only. ###### From: hmv@port.ac.uk (Mike Meredith at home) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 4 Jul 2000 20:02:42 -0700 Organization: Customer of Planet Online Lines: 48 Sender: mike@port.ac.uk Message-ID: References: <8joosu$p0h@netaxs.com> NNTP-Posting-Host: modem-102.aros.dialup.pol.co.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: news7.svr.pol.co.uk 962740863 26218 62.136.126.102 (4 Jul 2000 20:01:03 GMT) NNTP-Posting-Date: 4 Jul 2000 20:01:03 GMT X-Complaints-To: abuse@theplanet.net X-Newsreader: knews 1.0b.1 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeeds.belnet.be!news.belnet.be!newsfeed00.sul.t-online.de!newsfeed01.sul.t-online.de!t-online.de!news.netcologne.de!skynet.be!diablo.theplanet.net!news.theplanet.net!newspost.theplanet.net!lucifer!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59174 In article , kragen@dnaco.net (Kragen Sitaker) writes: > In article , > Mike Meredith at home wrote: >>* It's noisy. Any PC with a fan in is noisy enough to irritate > > True. I'd really like to get a NetWinder or some other fanless PC for > my bedroom. I've got an iMac in my bedroom. It's *almost* quiet enough not to disturb me whilst sleeping, and probably would be if it wasn't right by my bed. I'd suggest something without a hard disk. >>* It's less reliable. A terminal's only moving parts is the > > If you get rid of the fan, the only remaining moving parts are disk > drives. Burn a $20 Etherboot ROM for each PC, and you can get rid of > those, too. It'll be much more reliable than an ordinary PC, but a bit less reliable than a simple terminal --- it's still got more to go wrong. > And, while there are surely terminals that have been in constant use > for 15 years, there are many, many more that crapped out years ago. > Moving parts fail, but non-moving parts do, too. Hmm. It's been a while since we last used large numbers of plain terminals, but I can't remember more than a very small number going 'bang'. Of course they weren't 15 years old then. > It sounds like you're not using your PCs as terminals; instead, you are > using them as computers. I was talking about using them as terminals. Of course we're using them as computers --- using PC's solely as terminals strikes me as a pretty daft thing to do (unless a migration to a client-server application is imminent). I was talking about how terminals are better than PC's if all you want to use them for is as terminals. > It's interesting that Etherterminals cost more than computers set up as > netbooting X-terminals. What, exactly, is an Etherterminal? A > terminal that can telnet or LAT, or something else? :) Don't take my word for the price of an Etherterminal --- I haven't checked a price on them for more than a year. An Etherterminal is as you've guessed a terminal attached via an ethernet rather than serial line and capable of telnet. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <39621657$0$8303$3c090ad1@news.plethora.net> <3839E0A647FD4215.62BAC440B58C07E6.FDAEA7FEEB5401C6@lp.airnews.net> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 05 Jul 2000 15:04:20 GMT Lines: 23 Message-ID: <39634e74$0$8320$3c090ad1@news.plethora.net> NNTP-Posting-Host: 16419dcf.news.plethora.net X-Trace: 962809460 gemini.plethora.net 8320 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59163 In article <3839E0A647FD4215.62BAC440B58C07E6.FDAEA7FEEB5401C6@lp.airnews.net>, Jay Maynard wrote: >On 04 Jul 2000 16:52:39 GMT, Peter Seebach wrote: >>Well, I don't know about the fonts; I've seen lots of web browsers that don't >>display "fonts", don't "navigate" through pages, and that otherwise fail >>to comply with this description. >Lynx doesn't do fonts, true...what others are you thinking of that don't, >say, navigate through pages? They would seem to be pretty useless as web >browsers... "www", the command-line web browser. Prints the entire page, in-order, once, and presents you with a menu of links. No ability to scroll back and forth. It may seem useless, but we installed it by customer request, and our blind user spent many ours cheerfully browsing a medium that some people think is "purely visual". -s -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### From: genew@shuswap.net (Gene Wirchenko) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 05 Jul 2000 15:18:28 GMT Organization: Okanagan Internet Junction Lines: 36 Message-ID: <39635195.157080@news.shuswap.net> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> Reply-To: genew@shuswap.net NNTP-Posting-Host: salmonarm3-11.shuswap.net X-Trace: news.junction.net 962810197 15892 139.142.177.141 (5 Jul 2000 15:16:37 GMT) X-Complaints-To: abuse@junction.net NNTP-Posting-Date: 5 Jul 2000 15:16:37 GMT X-Newsreader: Forte Free Agent 1.1/32.230 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!cyclone.bc.net!news.junction.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59144 jmaynard@thebrain.conmicro.cx (Jay Maynard) wrote: >On Tue, 04 Jul 2000 05:12:58 GMT, Gene Wirchenko wrote: >>>to clear the conditionq). Another situation was hot interrupts which >> ^^^^^^^^^^ >> What is this, please? > >Tsk, Gene...spelling flames are lame. It's not a flame. I'm trying to follow the discussion in general, but since I lack specific knowledge, how would I know it's a typo? For all I knew, "q" could have been a suffix with a meaning. I mean, this is the computer field and we have all sorts of in[|s]ane abbrevs. >>>channel fenching to get the system back under control. >> ^^^^^^^^ >> What is this, please? > >This is another typo, but this time for a term that you may not have heard >of: "fencing". Fencing a channel, or a device, on a 370/390 means to disable >it as much as possible, and then to ignore any further I/O interrupts from >it; I don't know how this is done at the device level, but at the channel >level, the I/O interrupt mask is set to disable further interrupts from that >channel. Thanks. Sincerely, Gene Wirchenko Computerese Irregular Verb Conjugation: I have preferences. You have biases. He/She has prejudices. ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 50 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 05 Jul 2000 16:16:01 GMT NNTP-Posting-Host: 209.63.29.108 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962813761 209.63.29.108 (Wed, 05 Jul 2000 10:16:01 MDT) NNTP-Posting-Date: Wed, 05 Jul 2000 10:16:01 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!feeder.via.net!news.he.net!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59160 jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: > > This is another typo, but this time for a term that you may not have heard > of: "fencing". Fencing a channel, or a device, on a 370/390 means to disable > it as much as possible, and then to ignore any further I/O interrupts from > it; I don't know how this is done at the device level, but at the channel > level, the I/O interrupt mask is set to disable further interrupts from that > channel. oops, sorry about the finger-fumble/brain-check a lot of isolation was at the controller level ... first hit every address on the controller with hdv/clrio ... then hit every address on the channel with hdv/clrio and then clrch. the development & test labs had all the test cells on channel switches (I don't have reference at the moment so can't double check the model number) some number of the older 2919(?) with rotary switches, and a couple 3919(?). it was possible to dedicate a channel per test cell (still a significant improvement over having dedicated mainframe per machine). the development & engineering was on the 2nd floor in bldg. 14 ... before moving to bldg. 86. The product test lab was in bldg. 15. Product test lab was a separate organization from development and engineering and had authority to not release a product for first customer ship if it failed produuct test. Product test lab had some rather interesting boxes ... including a rather large environmental test chamber (humidity, air pressure, temperature, etc) and various kinds of error injection boxes. one of the things about mainframe ... both hardware & software ... after development, alpha tests, beta tests, etc ... was all done ... then there was a six month (elapsed) product test & Q/A cycle before first customer ship (FCS). there seem to be a lot of stuff these days that ship alpha or beta test as product (as soon as development finishes). random urls: http://www.garlic.com/~lynn/94.html#15 http://www.garlic.com/~lynn/95.html#3 http://www.garlic.com/~lynn/96.html#18 http://www.garlic.com/~lynn/99.html#31 http://www.garlic.com/~lynn/99.html#54 -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 14 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 05 Jul 2000 16:21:31 GMT NNTP-Posting-Host: 209.63.29.108 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962814091 209.63.29.108 (Wed, 05 Jul 2000 10:21:31 MDT) NNTP-Posting-Date: Wed, 05 Jul 2000 10:21:31 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!skynet.be!newsfeed.skycache.com!Cidera!63.211.125.72!cyclone2.usenetserver.com!news-out.usenetserver.com!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59161 Anne & Lynn Wheeler writes: > bldg. 15. Product test lab was a separate organization from > development and engineering and had authority to not release a product > for first customer ship if it failed produuct test. Product test lab there was even separation policy that people/badges with access to development & enginnering could not also have access to product test (& product test people couldn't have access to development & engineering). there were only a couple badges that I was aware of that were authorized for both environments. -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### Message-ID: <39636F5D.1643E7EC@thinkage.ca> From: "Alan T. Bowler" Organization: Thinkage Ltd. X-Mailer: Mozilla 4.5 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 25 Date: Wed, 05 Jul 2000 13:24:45 -0400 NNTP-Posting-Host: 192.102.11.4 X-Trace: nnrp1.uunet.ca 962817998 192.102.11.4 (Wed, 05 Jul 2000 13:26:38 EDT) NNTP-Posting-Date: Wed, 05 Jul 2000 13:26:38 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!cpk-news-hub1.bbnplanet.com!news.gtei.net!news3.bellglobal.com!torn!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59135 "Heinz W. Wiggeshoff" wrote: > > "Alan T. Bowler" (atbowler@thinkage.ca) writes: > > > ... > > Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line > > was something of an exception. However it eventually evolved > > something equivalent.) > > RD (Read Direct), WD (Write Direct) IBM System 360 Principles of > Operations and follow-ons. Supervisor mode only. Yes those existed, but almost nothing ever used them. Almost all systems had nothing wired up to the direct I/O connections. I know of only one that that a homebrew real time (calendar) clock, that was used at boot time instead of asking the operator for the date and time. I suspect that a few /360-44 systems that were actually used for real time control might have had some low bandwidth homebrew devices that connected by the direct I/O stuff. Some other mainframes also had similiar opcodes but these ended up being used not for actual I/O to external devices, but internal control functions. (Access to some information fields in control store etc.) ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 5 Jul 2000 17:45:31 GMT Organization: Neosoft (using Airnews.net!) Lines: 13 Message-ID: <5DDD142628B75745.F18A9C431CD41C02.78EFB1A83F3F84B3@lp.airnews.net> X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Wed Jul 5 12:45:32 2000 NNTP-Posting-Host: !cYt$1k-WH'Rcme (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!serra.unipi.it!hammer.uoregon.edu!newsfeed.direct.ca!look.ca!news.he.net!sn-xit-03!supernews.com!cabal10.airnews.net!news.airnews.net!cabal12.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:59140 On Wed, 05 Jul 2000 13:24:45 -0400, Alan T. Bowler wrote: >Some other mainframes also had similiar opcodes but these ended >up being used not for actual I/O to external devices, but internal >control functions. (Access to some information fields in control >store etc.) The 360/67 used the direct I/O facility for interprocessor communications in MP configurations. The SIGP (signal processor) facility came in with the 370. Since the direct I/O facility had the potential to hang the machine indefinitely, I doubt it saw very much use at all, anywhere. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 08 Jul 00 08:01:23 GMT Organization: UltraNet Communications, Inc. Lines: 33 Message-ID: <8k71dc$qdd$2@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> X-Trace: QCEJhnp5gmgIaD7KLXSLuXa/6ediDyAznbLs61PDnjk= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 8 Jul 2000 10:56:44 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-242 Xref: chonsp.franklin.ch alt.folklore.computers:59237 In article <3965dee3$0$8312$3c090ad1@news.plethora.net>, seebs@plethora.net (Peter Seebach) wrote: >In article <39635195.157080@news.shuswap.net>, >Gene Wirchenko wrote: >> It's not a flame. I'm trying to follow the discussion in >>general, but since I lack specific knowledge, how would I know it's a >>typo? For all I knew, "q" could have been a suffix with a meaning. I >>mean, this is the computer field and we have all sorts of in[|s]ane >>abbrevs. > >conditionq abbrvp? > >:) > >Anyway, "conditionq" looks like a shorthand for "condition >queue" which sounds >like, if it's not a technical term, it ought to be. That's how I read the original post. And it kinda made some sense in the context although I started thinking about a queue of conditions like a line printer listing queue. Having many definitions for the same word seems to be unique to the computer biz. One also has to figure out the _year_ the poster is commenting about. I really would like to be a fly on the wall of a historian 100 years from now :-). I might even learn some new swears. /BAH /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 08 Jul 00 10:01:35 GMT Organization: UltraNet Communications, Inc. Lines: 19 Message-ID: <8k78eo$daf$1@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> X-Trace: c/ZbcIxpkmHTMKmC15y5lKiUuEpS2UJBDeV+1CUDLY4= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 8 Jul 2000 12:56:56 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!feeder.via.net!netnews.com!feed1.news.rcn.net!rcn!216-164-248-213 Xref: chonsp.franklin.ch alt.folklore.computers:59238 In article , Anne & Lynn Wheeler wrote: >Anne & Lynn Wheeler writes: >> bldg. 15. Product test lab was a separate organization from >> development and engineering and had authority to not release a product >> for first customer ship if it failed produuct test. Product test lab > >there was even separation policy that people/badges with access to >development & enginnering could not also have access to product test >(& product test people couldn't have access to development & >engineering). there were only a couple badges that I was aware of that >were authorized for both environments. What was the reasoning behind that enforced separation? It doesn't sound terribly efficient to me. /BAH Subtract a hundred and four for e-mail. ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 105 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 05 Jul 2000 22:43:05 GMT NNTP-Posting-Host: 209.63.29.106 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962836985 209.63.29.106 (Wed, 05 Jul 2000 16:43:05 MDT) NNTP-Posting-Date: Wed, 05 Jul 2000 16:43:05 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeedZ.netscum.dQ!netscum.int!news.he.net!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59328 .... 3880/3830 reference http://www.garlic.com/~lynn/99.html#54 above ref. was situation where i got initial blame for very poor performance in the product test lab (bldg. 15). They had a 3033 with operating system providing a lot of online services as well as product testing services on a dozen channels. at the time of the mentioned incident they had a 3830 disk controller with 16 3330-II disks drives supporting online services. the 3880 disk controller was starting product test and they appeared to have one that was reasonably far along & passed all the functional & error handling tests. One weekend they decided to swap the 3880 for their production 3830 to see what happened. that monday their online performance went all to pieces and I started getting calls. Turns out that product test had a rather simple performance acceptance test ... it was basically a single stream VS1 jobstream that ran on 3330-ii and compared elapsed time with 3830 controller and 3880 controller. the 3880 had already passed this performance test. the problem was rather more complex. the 3830 was a fast horizontal microcoded machine that handled all operations. The 3880 was a JIB-prime (vertical) microprocessor with hardware assist to handle actual data-movement. While 3830 only need to handle 1.5mbytes/sec data movement and the 3880 needed to handle 3mbyte/sec data streaming mode ... the old 1.5mbytes/sec was a byte at a time channel transfer ... while data streaming was 8bytes at a time channel transfer (i.e. channel handshaking was occuring 1/8th as often). In any case, while 3880 supported higher peak data transfer ... the rest of the 3880 processing was much slower than the 3830. One of these areas was in controller "house-keeping" after a disk operation completes. In the 3830 case, this was effectively instantaneously but it the 3880 case, it was taking 1mills to 1.5mills longer. In order to have the 3880 pass the VS1 performance acceptance test, they modified the way the controller operates so that end-of-operation interrupt was presented to the channel/processor before the controller had completed housekeeping. In the VS1 case, this resulted in posted an interrupt to the operating system, waking up some application/subsystem, the application/subsystem doing some processing and then initiating the next I/O operation. Effectively the controller house-keeping overhead was being overlapped with VS1 processing. However, in the online service case with lots of concurrent tasks going on, thee were almost alwas requests waiting for the controller. When the controller signaled end-of-operation to the channnel/processor ... the operating system took the interrupt and then checked the queue of pending requests and would immediately "redrive" the controller &/or device with a new request. This would typically happen within tens of microseconds ... and find the controller still busy with house-keeping duties from the previous operation. The processor then was presented with, CC1, csw stored, control unit busy (i.e. status modifier plus busy; SM+BUSY in the channel status word). The operating system then had to mark the controller busy and wait until it freed itself up. Any time a controller presents SM+BUSY, it then has to present a later interrupt to indicate that it was now free. While the VS1 workload almost never experienced the problem, the online service was doing double START I/O operations, seeing double I/O interrupts, and having a large percentage of operations being delayed by 1.5mills (for the 3880 compared to the 3830). This wasn't happening in the VS1 performance test case because it was effectively single-thread, not doing multiple concurrent operations, and effectively overlapping application/subsystem processing with the control unit housekeeping overhead. Fortunately this incident occured nearly 6 months before scheduled first customer ship of the 3880 ... discovering the problem at this point allowed it to be addressed before the boxes being deployed in customer installations. In general, the switch from stand-alone engineering, development and test to working in an operating system environment allowed for a lot of problems to be flushed out before first customer shipment (that otherwise wouldn't have been encountered until after the boxes were at customers). Another problem that the post-interrupt house-keeping resulted in was finding problems during the house-keeping (not directly related to the actual data transfer) that needed to be reported to the operating system. The standard process for reporting problem is adding the unit check flag to the status stored when the interrupt occurs. Since the final interrupt on the process has already occured ... there was no process for presenting the unit check error flag as part of the post-interrupt house-keeping. The engineers decided to solve this problem was by presenting an unsoliccted unit check. This turns out to be a violation of the mainframe channel architecture (it is possible to do it ... but the specification that the mainframe operating system is written to ... effectively makes unsolicited unit check interrupt an invalid operation). I spent three months playing middle man between the san jose controller engineers and the POK channel engineers. random refs: http://www.garlic.com/~lynn/2000.html#9 -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers From: Terry Kennedy Subject: Re: Does the word "mainframe" still have a meaning? X-Complaints-To: Email abuse@spc.edu if this posting is inappropriate User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (BSD/OS/4.1 (i386)) NNTP-Posting-Host: gate.tmk.com Organization: St. Peter's College, US Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> X-Trace: spcuna.spc.edu 962841654 13953 terry [204.141.35.61] Date: Thu, 6 Jul 2000 00:00:55 GMT Lines: 29 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!nuq-feed.news.verio.net!dfw-peer.news.verio.net!ord-feed.news.verio.net!news.verio.net!news.spc.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59197 Anne & Lynn Wheeler writes: > Fortunately this incident occured nearly 6 months before scheduled > first customer ship of the 3880 ... discovering the problem at this > point allowed it to be addressed before the boxes being deployed in > customer installations. In general, the switch from stand-alone > engineering, development and test to working in an operating system > environment allowed for a lot of problems to be flushed out before > first customer shipment (that otherwise wouldn't have been encountered > until after the boxes were at customers). Indeed. One of the things that IBM did very well was turn "good ideas" into "products" by involving a whole bunch of reliability/maintainability folks. Some years ago, I was working with a disk drive manufacturer (30MB ST- 506 as a "leading edge" product, which should give you an idea of the time- frame). Some of their lead designers had come from IBM, from the 3380 drive project. They developed a wonderfully innovative drive, which unfortunately tended to fail within a few weeks, mostly due to not having the type of folks that IBM did to do the polishing of the product. There were a whole variety of issues with the drive, some of which got fixed and some didn't (and some as midnight skunk works projects). The man- ufacturer stopped making drives before they shipped production quantities, and they hung on for a few years as a media manufacturer for other drive makers. Terry Kennedy http://www.tmk.com terry@tmk.com Jersey City, NJ USA ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 7 Jul 2000 13:18:30 GMT Organization: The MITRE Corporation Lines: 24 Message-ID: <8k4lb6$9t3$1@top.mitre.org> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 962975910 10147 128.29.251.13 (7 Jul 2000 13:18:30 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 7 Jul 2000 13:18:30 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:59195 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: >"Alan T. Bowler" (atbowler@thinkage.ca) writes: >> >> Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line >> was something of an exception. However it eventually evolved >> something equivalent.) > RD (Read Direct), WD (Write Direct) IBM System 360 Principles of > Operations and follow-ons. Supervisor mode only. RD and WD were not available on all S/360 models, and AFAIK they were never standard features on models where they were available. As has been mentioned in other threads, the architecture documented in the "Principles of Operation" (PoO) manual were for the entire S/360 product line, including instructions that might be optional or unavailable on specific S/360 models. RD and WD were very limited in function: essentially they would present a byte on the direct output interface (WD) or read what was presented to the interface (RD); there was no channel programming for the interface because there was no channel. Joe Morris ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 07 Jul 2000 13:45:07 GMT Lines: 20 Message-ID: <3965dee3$0$8312$3c090ad1@news.plethora.net> NNTP-Posting-Host: bdca172b.news.plethora.net X-Trace: 962977507 gemini.plethora.net 8312 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59325 In article <39635195.157080@news.shuswap.net>, Gene Wirchenko wrote: > It's not a flame. I'm trying to follow the discussion in >general, but since I lack specific knowledge, how would I know it's a >typo? For all I knew, "q" could have been a suffix with a meaning. I >mean, this is the computer field and we have all sorts of in[|s]ane >abbrevs. conditionq abbrvp? :) Anyway, "conditionq" looks like a shorthand for "condition queue" which sounds like, if it's not a technical term, it ought to be. -s -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 9 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 07 Jul 2000 16:02:15 GMT NNTP-Posting-Host: 209.63.29.113 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 962985735 209.63.29.113 (Fri, 07 Jul 2000 10:02:15 MDT) NNTP-Posting-Date: Fri, 07 Jul 2000 10:02:15 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-hog.berkeley.edu!ucberkeley!news.he.net!easynews!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59308 seebs@plethora.net (Peter Seebach) writes: > Anyway, "conditionq" looks like a shorthand for "condition queue" which sounds > like, if it's not a technical term, it ought to be. but it really was figner fumble -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### From: dowe@krikkit.localdomain (Dowe Keller) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> Message-ID: X-Newsreader: slrn (0.9.4.3 UNIX) NNTP-Posting-Host: 209.234.196.99 X-Original-NNTP-Posting-Host: 209.234.196.99 Date: 7 Jul 2000 15:40:22 -0700 X-Trace: 7 Jul 2000 15:40:22 -0700, 209.234.196.99 Organization: news.sierratel.com Lines: 20 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!209.155.26.10!news.sierratel.com!dowe Xref: chonsp.franklin.ch alt.folklore.computers:59179 On 07 Jul 2000 13:45:07 GMT, Peter Seebach wrote: >In article <39635195.157080@news.shuswap.net>, >Gene Wirchenko wrote: >> It's not a flame. I'm trying to follow the discussion in >>general, but since I lack specific knowledge, how would I know it's a >>typo? For all I knew, "q" could have been a suffix with a meaning. I >>mean, this is the computer field and we have all sorts of in[|s]ane >>abbrevs. > >conditionq abbrvp? I'm not trying to pick nicks, but shouldn't it be (abbrvp conditionq) -- dowe@sierratel.com --- This is the theory that Jack built. This is the flaw that lay in the theory that Jack built. This is the palpable verbal haze that hid the flaw that lay in... ###### From: Juergen Nickelsen Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 08 Jul 2000 00:48:47 +0200 Organization: [Posted via] Interactive Networx Lines: 9 Sender: ni@goting.jn.berlin.snafu.de Message-ID: References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> NNTP-Posting-Host: ip133.berlin64.pub-ip.de.psi.net X-Newsreader: Gnus v5.6.45/XEmacs 21.1 - "Canyonlands" Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeed00.sul.t-online.de!newsfeed01.sul.t-online.de!t-online.de!unlisys!news.snafu.de!goting.jn.berlin.snafu.de!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59351 dowe@krikkit.localdomain (Dowe Keller) writes: > I'm not trying to pick nicks, but shouldn't it be > (abbrvp conditionq) t -- Juergen Nickelsen ###### From: genew@shuswap.net (Gene Wirchenko) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 08 Jul 2000 15:48:09 GMT Organization: Okanagan Internet Junction Lines: 20 Message-ID: <3966be3d.7055837@news.shuswap.net> References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> Reply-To: genew@shuswap.net NNTP-Posting-Host: salmonarm3-27.shuswap.net X-Trace: news.junction.net 963071248 24911 139.142.177.157 (8 Jul 2000 15:47:28 GMT) X-Complaints-To: abuse@junction.net NNTP-Posting-Date: 8 Jul 2000 15:47:28 GMT X-Newsreader: Forte Free Agent 1.1/32.230 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsfeed.direct.ca!look.ca!cyclone.bc.net!news.junction.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59277 Anne & Lynn Wheeler wrote: >seebs@plethora.net (Peter Seebach) writes: >> Anyway, "conditionq" looks like a shorthand for "condition queue" which sounds >> like, if it's not a technical term, it ought to be. > >but it really was figner fumble But it doesn't look like it! If it were finger fumble, I'd expect the odd character to be close to the one before, say "conditionm". That's why I didn't figure "conditionq" as a typo. Sincerely, Gene Wirchenko Computerese Irregular Verb Conjugation: I have preferences. You have biases. He/She has prejudices. ###### From: genew@shuswap.net (Gene Wirchenko) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sat, 08 Jul 2000 15:48:10 GMT Organization: Okanagan Internet Junction Lines: 18 Message-ID: <3966bebd.7183070@news.shuswap.net> References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> Reply-To: genew@shuswap.net NNTP-Posting-Host: salmonarm3-27.shuswap.net X-Trace: news.junction.net 963071249 24911 139.142.177.157 (8 Jul 2000 15:47:29 GMT) X-Complaints-To: abuse@junction.net NNTP-Posting-Date: 8 Jul 2000 15:47:29 GMT X-Newsreader: Forte Free Agent 1.1/32.230 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!cyclone.bc.net!news.junction.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59282 dowe@krikkit.localdomain (Dowe Keller) wrote: [snip] >This is the theory that Jack built. >This is the flaw that lay in the theory that Jack built. >This is the palpable verbal haze that hid the flaw that lay in... Is there more? If so, please post it! Sincerely, Gene Wirchenko Computerese Irregular Verb Conjugation: I have preferences. You have biases. He/She has prejudices. ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 8 Jul 2000 16:05:10 GMT Organization: The MITRE Corporation Lines: 30 Message-ID: <8k7jfm$565$1@top.mitre.org> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> <8k78eo$daf$1@bob.news.rcn.net> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 963072310 5317 128.29.251.13 (8 Jul 2000 16:05:10 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 8 Jul 2000 16:05:10 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!paloalto-snf1.gtei.net!news.gtei.net!newsfeed.stanford.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:59194 jmfbahciv@aol.com writes: > Anne & Lynn Wheeler wrote: >>there was even separation policy that people/badges with access to >>development & enginnering could not also have access to product test >>(& product test people couldn't have access to development & >>engineering). there were only a couple badges that I was aware of that >>were authorized for both environments. >What was the reasoning behind that enforced separation? >It doesn't sound terribly efficient to me. I'm not speaking from experience (I've never worked for IBM, although there were a couple of offers -- which some people said were to shut me up at user group meetings...) but the separation sounds quite reasonable if you're trying to make sure that the product (a) works as documented, (b) doesn't produce unexpected results ("the principle of least astonishment"), and (c) can be installed and used by mere mortals. Once you allow the designers to become informal resources during product test ("Oh, as long as you're here, how am I supposed to load a tape when we've got pressure columns intead of vacuum columns?") you're no longer testing in a real-world customer environment. A product test should live or die in an environment close to what will be found after the customer has bought it. Of course, today's software products (such as Windows) *always* pass such a rigid set of tests before they are shipped to the end users. NOT. Joe Morris ###### Sender: lynn@LYNNLT Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> <8k78eo$daf$1@bob.news.rcn.net> Reply-To: Anne & Lynn Wheeler From: Anne & Lynn Wheeler Message-ID: Organization: Wheeler&Wheeler Lines: 18 User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 08 Jul 2000 16:16:27 GMT NNTP-Posting-Host: 209.63.29.114 X-Complaints-To: support@adcomsys.net X-Trace: news-west.eli.net 963072987 209.63.29.114 (Sat, 08 Jul 2000 10:16:27 MDT) NNTP-Posting-Date: Sat, 08 Jul 2000 10:16:27 MDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!sat.cyberhighway.net!news-west.eli.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59319 jmfbahciv@aol.com writes: > What was the reasoning behind that enforced separation? > It doesn't sound terribly efficient to me. I've observed (& other people have also) that if product and product test (quality control )report to the same executive ... that when product test says that shipping of boxes has to be delayed for six months because of engineering &/or quality reasons ... an executive might over-rule the product test group and ship the box anyway. having the responsibilities in completely different organizations help separate possibilities of internal "conflict of interest" with regard to executive decisions about shipping a product vis-a-vis the quality control of the product. -- Anne & Lynn Wheeler | lynn@garlic.com, finger for pgp key http://www.garlic.com/~lynn/ ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 08 Jul 2000 16:33:56 GMT Lines: 16 Message-ID: <396757f4$0$8310$3c090ad1@news.plethora.net> NNTP-Posting-Host: 39db8369.news.plethora.net X-Trace: 963074036 gemini.plethora.net 8310 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59310 In article , Anne & Lynn Wheeler wrote: >seebs@plethora.net (Peter Seebach) writes: >> Anyway, "conditionq" looks like a shorthand for "condition queue" which sounds >> like, if it's not a technical term, it ought to be. >but it really was figner fumble Which, if it's not a technical term, oughdt to be. -s [typo left in on purpose] -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Sun, 09 Jul 00 08:38:27 GMT Organization: UltraNet Communications, Inc. Lines: 22 Message-ID: <8k9nv5$pre$5@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> <396757f4$0$8310$3c090ad1@news.plethora.net> X-Trace: MhsxcyPtD0+fFgDWlq08sLM2NOX0pCmptrlRY2FJQ40= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 9 Jul 2000 11:33:57 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-255-30 Xref: chonsp.franklin.ch alt.folklore.computers:59426 In article <396757f4$0$8310$3c090ad1@news.plethora.net>, seebs@plethora.net (Peter Seebach) wrote: >In article , >Anne & Lynn Wheeler wrote: >>seebs@plethora.net (Peter Seebach) writes: >>> Anyway, "conditionq" looks like a shorthand for "condition >>>queue" which sounds >>> like, if it's not a technical term, it ought to be. > >>but it really was figner fumble > >Which, if it's not a technical term, oughdt to be. > >-s >[typo left in on purpose] Did you guys ever have to change the name of a variable to avoid figner fumbles? /BAH Subtract a hundred and four for e-mail. ###### Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> <3966bebd.7183070@news.shuswap.net> From: Dowe Keller X-Newsreader: Gnus v5.5/Emacs 20.3 Date: 08 Jul 2000 12:50:34 -0700 Message-ID: Lines: 8 NNTP-Posting-Host: 209.234.196.98 X-Original-NNTP-Posting-Host: 209.234.196.98 X-Trace: 8 Jul 2000 13:10:28 -0700, 209.234.196.98 Organization: news.sierratel.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!209.155.26.10!news.sierratel.com!209.234.196.98 Xref: chonsp.franklin.ch alt.folklore.computers:59477 I don't really know, Its from my fortune cookie file. When I saw it, I knew I must make it my .signature :-) dowe@sierratel.com --- This is the theory that Jack built. This is the flaw that lay in the theory that Jack built. This is the palpable verbal haze that hid the flaw that lay in... ###### Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> From: jata@aepiax.net (Julian Thomas) Subject: Re: Does the word "mainframe" still have a meaning? Message-ID: <3967d17d$1$wg$mr2ice@news.epix.net> X-Newsreader: MR/2 Internet Cruiser Edition for OS/2 v2.19zf/19zf Lines: 21 Date: Sun, 09 Jul 2000 01:12:32 GMT NNTP-Posting-Host: 199.224.125.80 X-Complaints-To: abuse@epix.net X-Trace: news1.epix.net 963105152 199.224.125.80 (Sat, 08 Jul 2000 21:12:32 EDT) NNTP-Posting-Date: Sat, 08 Jul 2000 21:12:32 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!128.32.206.60.MISMATCH!news-hog.berkeley.edu!ucberkeley!news-xfer.epix.net!news1.epix.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59456 In <39636F5D.1643E7EC@thinkage.ca>, on 07/05/00 at 01:24 PM, "Alan T. Bowler" said: >> RD (Read Direct), WD (Write Direct) IBM System 360 Principles of >> Operations and follow-ons. Supervisor mode only. >Yes those existed, but almost nothing ever used them. Except for an early MICR device (1419 ??) that used the interface. -- Julian Thomas: jt . epix @ net http://home.epix.net/~jt remove letter a for email (or switch . and @) In the beautiful Finger Lakes Wine Country of New York State! Boardmember of POSSI.org - Phoenix OS/2 Society, Inc http://www.possi.org -- -- Error: Keyboard not attached. Press F1 to continue. ###### Newsgroups: alt.folklore.computers Followup-To: Anne & Lynn Wheeler References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39617206.34902445@news.shuswap.net> From: jata@aepiax.net (Julian Thomas) Subject: Re: Does the word "mainframe" still have a meaning? Message-ID: <3967d4c9$2$wg$mr2ice@news.epix.net> X-Newsreader: MR/2 Internet Cruiser Edition for OS/2 v2.19zf/19zf Lines: 32 Date: Sun, 09 Jul 2000 01:26:36 GMT NNTP-Posting-Host: 199.224.125.80 X-Complaints-To: abuse@epix.net X-Trace: news1.epix.net 963105996 199.224.125.80 (Sat, 08 Jul 2000 21:26:36 EDT) NNTP-Posting-Date: Sat, 08 Jul 2000 21:26:36 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-hog.berkeley.edu!ucberkeley!news-xfer.epix.net!news1.epix.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59458 In , on 07/05/00 at 10:43 PM, Anne & Lynn Wheeler said: >The engineers decided to solve this problem was by presenting an >unsoliccted unit check. This turns out to be a violation of the mainframe >channel architecture (it is possible to do it ... but the specification >that the mainframe operating system is written to .... effectively makes >unsolicited unit check interrupt an invalid operation). I spent three >months playing middle man between the san jose controller engineers and >the POK channel engineers. Probably right after I got out of the streaming channel business. I was coordinating the debug of that monstrosity (we had boards with more EC wires on them than anyone thought was possible), working first with CTVs and then the earliest Cutter-Whitney units that were shipped to POK from SJ (I still have one of the wooden shipping crates which is now doing duty as a woodbox). But where did the OS folks come into this discussion? -- Julian Thomas: jt . epix @ net http://home.epix.net/~jt remove letter a for email (or switch . and @) In the beautiful Finger Lakes Wine Country of New York State! Boardmember of POSSI.org - Phoenix OS/2 Society, Inc http://www.possi.org -- -- There are 3 kinds of people: those who can count & those who can't. ###### Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <35A5E226011C8A1D.508638B850D4B114.D8FC648800C0B958@lp.airnews.net> <884219A0B1529AD2.B470CFBDFB5D9EB1.3A0869F842B9242B@lp.airnews.net> <8jq544$9l0$1@top.mitre.org> <3960aae8$1$wg$mr2ice@news.epix.net> From: jata@aepiax.net (Julian Thomas) Subject: Re: Does the word "mainframe" still have a meaning? Message-ID: <3967d636$4$wg$mr2ice@news.epix.net> X-Newsreader: MR/2 Internet Cruiser Edition for OS/2 v2.19zf/19zf Lines: 25 Date: Sun, 09 Jul 2000 01:32:41 GMT NNTP-Posting-Host: 199.224.125.80 X-Complaints-To: abuse@epix.net X-Trace: news1.epix.net 963106361 199.224.125.80 (Sat, 08 Jul 2000 21:32:41 EDT) NNTP-Posting-Date: Sat, 08 Jul 2000 21:32:41 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-hog.berkeley.edu!ucberkeley!news-xfer.epix.net!news1.epix.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59455 In , on 07/03/00 at 03:58 PM, Anne & Lynn Wheeler said: >This configuration was replicated in Boulder when they moved the IMS >field support team to a bldg. on the other side of I70. Because of >regulations weren't able to use microwave between the roofs of the two >buildings and so instead installed T1 infrared modems on polls on the two >buildings. There was concern about signal quality with the infrared >modems because of rain fade. I remember reading about the IR T1 link option as a clever way to bypass the regulatory agencies. Interesting that building expansion could muck up the alignment. -- Julian Thomas: jt . epix @ net http://home.epix.net/~jt remove letter a for email (or switch . and @) In the beautiful Finger Lakes Wine Country of New York State! Boardmember of POSSI.org - Phoenix OS/2 Society, Inc http://www.possi.org -- -- Welcome to the Microsoft tour. Please don't feed the lawyers. ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 9 Jul 2000 20:33:02 GMT Organization: The National Capital FreeNet Lines: 46 Message-ID: <8kanhu$3pp$1@freenet9.carleton.ca> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 963174782 3897 134.117.136.30 (9 Jul 2000 20:33:02 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 9 Jul 2000 20:33:02 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!easynet-tele!easynet.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.frontiernet.net!nntp.gctr.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59568 Joe Morris (jcmorris@jmorris-pc.MITRE.ORG) writes: >ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > >>"Alan T. Bowler" (atbowler@thinkage.ca) writes: >>> >>> Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line >>> was something of an exception. However it eventually evolved >>> something equivalent.) > >> RD (Read Direct), WD (Write Direct) IBM System 360 Principles of >> Operations and follow-ons. Supervisor mode only. After I posted this, I cracked one of my PoOs to find that the actual opcodes are RDD and WRD. I'm surprised that I got away with the above - power of suggestion, or bit-rot amongst us geezergeeks? > >RD and WD were not available on all S/360 models, and AFAIK they were >never standard features on models where they were available. As has >been mentioned in other threads, the architecture documented in the >"Principles of Operation" (PoO) manual were for the entire S/360 >product line, including instructions that might be optional or >unavailable on specific S/360 models. > Which brings back memories of MVCIN - move character inverse. At a PPOE, there was a lot of PL/I text processing for typesetting. A huge amount of CPU time was spent trimming off trailing blanks. Even with VARYING length strings, it was possible to have blanks padding to the maximum declared length. So when I spied MVCIN in the /370 PoO, I figured that an assembler routine could flip the argument, then use the PL/I VERIFY and SUBSTR builtin function to strip off leading blanks, followed by another call to the MVCIN routine. Sadly, the 4381 didn't have that instruction, and management wouldn't let me frig with the micro- code floppy. I wound up writing an assembler routine named TRIMCV, for TRIM (as in SNOBOL 4) Character Varying. The basic idea was to loop from the end of the string EXecuting a CLCL (Compare Logical Character Long). Each time through the loop, the compare address was decremented by the length of the previous comparision, and the next compare length was doubled. When a non-blank was detected, the compare length was reset to one, so that one could sneak up to the last such character. Obviously, the routine had to test for null and all-blank strings. The tremendous cost savings didn't rescue the company. B-) ###### From: bbreynolds@aol.comskipthis (Bruce B. Reynolds) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Lines: 13 NNTP-Posting-Host: ladder05.news.aol.com X-Admin: news@aol.com Date: 10 Jul 2000 01:54:02 GMT References: <396757f4$0$8310$3c090ad1@news.plethora.net> Organization: AOL http://www.aol.com X-Newsreader: Session Scheduler Message-ID: <20000709215402.02775.00000933@nso-bd.aol.com> Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed00.sul.t-online.de!t-online.de!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!europa.netcrusader.net!152.163.239.129!portc01.blue.aol.com!audrey04.news.aol.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59560 In article <396757f4$0$8310$3c090ad1@news.plethora.net>, seebs@plethora.net (Peter Seebach) writes: >>but it really was figner fumble > >Which, if it's not a technical term, oughdt to be. > It already is: lapsus digitali. -- Bruce B. Reynolds, Independent/Legacy Systems Consultant: Trailing Edge Technologies, Glenside PA---Sweeping Up Behind Data Processing Dinosaurs ###### Message-ID: <3969F84E.8AB52B6B@thinkage.ca> From: "Alan T. Bowler" Organization: Thinkage Ltd. X-Mailer: Mozilla 4.5 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> <8kanhu$3pp$1@freenet9.carleton.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 28 Date: Mon, 10 Jul 2000 12:22:38 -0400 NNTP-Posting-Host: 192.102.11.4 X-Trace: nnrp1.uunet.ca 963246259 192.102.11.4 (Mon, 10 Jul 2000 12:24:19 EDT) NNTP-Posting-Date: Mon, 10 Jul 2000 12:24:19 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!news.maxwell.syr.edu!cyclone.bc.net!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59497 "Heinz W. Wiggeshoff" wrote: > Which brings back memories of MVCIN - move character inverse. At a > PPOE, there was a lot of PL/I text processing for typesetting. A huge > amount of CPU time was spent trimming off trailing blanks. Even with > VARYING length strings, it was possible to have blanks padding to the > maximum declared length. So when I spied MVCIN in the /370 PoO, I > figured that an assembler routine could flip the argument, then use > the PL/I VERIFY and SUBSTR builtin function to strip off leading blanks, > followed by another call to the MVCIN routine. Sadly, the 4381 didn't > have that instruction, and management wouldn't let me frig with the micro- > code floppy. > > I wound up writing an assembler routine named TRIMCV, for TRIM (as in > SNOBOL 4) Character Varying. The basic idea was to loop from the end > of the string EXecuting a CLCL (Compare Logical Character Long). Each > time through the loop, the compare address was decremented by the length > of the previous comparision, and the next compare length was doubled. > When a non-blank was detected, the compare length was reset to one, so > that one could sneak up to the last such character. Obviously, the > routine had to test for null and all-blank strings. The tremendous cost > savings didn't rescue the company. I thought the usual idiom for this was to use a translate to flip the string left to right, and then use a translate and test to scan for a non-blank. I never coded this I just heard it described. The Honeywell-6000 family is where I do most of my assembler coding and it has translate and test opcodes to work in either direction. ###### Newsgroups: alt.folklore.computers From: jsaum@world.std.com (Jim Saum) Subject: Re: Does the word "mainframe" still have a meaning? Sender: news@world.std.com (Mr Usenet Himself) Message-ID: Date: Mon, 10 Jul 2000 17:52:36 GMT References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> <3967d17d$1$wg$mr2ice@news.epix.net> NNTP-Posting-Host: ppp0c018.std.com Organization: Software Tool & Die X-Newsreader: MT-NewsWatcher 2.4.4 Lines: 42 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!news.nikoma.de!tiscalinetde!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!news.umass.edu!world!jsaum Xref: chonsp.franklin.ch alt.folklore.computers:59489 In article <3967d17d$1$wg$mr2ice@news.epix.net>, jata@aepiax.net (Julian Thomas) wrote: >In <39636F5D.1643E7EC@thinkage.ca>, on 07/05/00 > at 01:24 PM, "Alan T. Bowler" said: > >>> RD (Read Direct), WD (Write Direct) IBM System 360 Principles of >>> Operations and follow-ons. Supervisor mode only. > >>Yes those existed, but almost nothing ever used them. > >Except for an early MICR device (1419 ??) that used the interface. MICR (magnetic ink character recognition) check-processing equipment was (and still is) used at banks and check-clearing centers. Aside from reading the magnetically-encoded information along the bottom edge of the checks, the devices have a row of stacker pockets like a card sorter. The 360-era MICR hardware (IBM 1419 sounds right, but I'm not sure) presented the 360 with a real-time constraint: after reading a check, the host program had a small number of milliseconds (I don't remember the precise number, but it was something like 3 or 5 ms) to decide which stacker pocket to drop the check into, or a default decision would be made by the hardware. A midrange 360 model was capable of processing only something like 200-1000 instructions before the deadline. So the MICR device was connected to one of the 360's six external-interrupt signal-in lines (another part of the Direct Control feature that included the RDD and WRD instructions), and there was standard support in DOS/360 for a user-supplied external-interrupt handler to make the stacker decision and send a response back to the device in the required amount of time. I don't know if any of the MICR shops drove the devices with OS/360 instead of DOS; if so, presumably they had to write OS mods, since IIRC there was no standard support like this in OS. This always seemed to me more like something one would have expected to see on a dedicated minicomputer than on a 360. - Jim Saum ###### From: Eric Sosman Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 10 Jul 2000 13:52:58 -0400 Organization: Sun Microsystems Lines: 35 Message-ID: <396A0D7A.D589C509@east.sun.com> References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> <3966bebd.7183070@news.shuswap.net> NNTP-Posting-Host: tardis.east.sun.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.icl.net!colt.net!news0.de.colt.net!blackbush.xlink.net!newsfeed.germany.net!newsfeed.easynews.net!easynews.net!easynet-melon!easynet.net!btnet-feed2!btnet!carbon.eu.sun.com!new-usenet.uk.sun.com!eastnews1.east.sun.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59507 Gene Wirchenko wrote: > > dowe@krikkit.localdomain (Dowe Keller) wrote: > > [snip] > > >This is the theory that Jack built. > >This is the flaw that lay in the theory that Jack built. > >This is the palpable verbal haze that hid the flaw that lay in... > > Is there more? If so, please post it! It's (slightly misquoted, I think) from "A Space Child's Mother Goose" (or maybe "The S.C.M.G.") by Winsor and Parry. I don't know whether it's still in print, but it's a delightful little read if you can get hold of it. Even the footnotes are funny! From (possibly faulty) memory: The colloid and the crystalloid Were joshing just in jest. The colloid called the crystalloid A pseudo-anapest. [*] Some called them physical and some called them chemic, And some thought the whole affair was slightly academic. [*] The pseudo-anapest Moves awkwardly at best. Its feet are long, uneven, and retractile. Who hunts this beast in rhythm Must take his meter withm, And still may only bag a ptero-dactyl. -- Eric.Sosman@east.sun.com ###### From: robert@bonomi.invalid Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 10 Jul 2000 21:26:38 GMT Organization: Not Much Lines: 37 Sender: robert@bonomi.invalid Message-ID: <8kdf2e0p4u@enews3.newsguy.com> References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> NNTP-Posting-Host: p-003.newsdawg.com Originator: robert@bonomi.invalid X-Newsreader: trn 4.0-test69 (20 September 1998) Originator: bonomi@diskless. (Robert Bonomi) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news.idt.net!howland.erols.net!newsfeed.lotsanews.com.MISMATCH!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59552 In article <8k9nv5$pre$5@bob.news.rcn.net>, wrote: > [.. munch ..] > >Did you guys ever have to change the name of a variable to >avoid figner fumbles? Does 'gave up, and had the progam _accept_ the mis-spelling' count? This was a wonderful, _weird_, extremely effective, suite of applications. Totally batch oriented (actual *punch-card* submissions, no less). and included using the system _assembler_ as a "word processor". "Configuring" a customer consisted of a set of specifications, and some macroes. This was run through the assembler. The -output- of that pass was fed back into the assembler as *input*. *That* output was -again- fed to the assembler as input. The output from _that_ pass, was (yet again!) fed to the assembler (in _pieces, as it was source for some -seventeen- *different* programs) as input. At which time the actual 'object code' for each of those programs was produced. Trivia: If you wanted a literal quote-mark in the final program, it took (I think I've got this right) *TWENTY*SEVEN* consecutive occurances in the 'source' file. Anyway, part of the 'specifications' was what reports the customer wanted, and what the layout of them should look like. This involved a "FORM" directive, oddly enough. Several of the people preparing the specification sets for running had an absolutely *terrible* time getting that word right. Finally, the progammer 'gave up', and added another macro to the library. It did precisely _two_ things: 1) output a 'warning' message. The exact text of which was: "LOOK TOAD!! IT'S FORM, NOT FROM." 2) Invoked the 'FORM' macro with the same arguments that it, the "FROM" macro, had been called with. ###### From: robert@bonomi.invalid Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 10 Jul 2000 21:30:32 GMT Organization: Not Much Lines: 38 Sender: robert@bonomi.invalid Message-ID: <8kdf9o0pcg@enews3.newsguy.com> References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> NNTP-Posting-Host: p-051.newsdawg.com Originator: robert@bonomi.invalid X-Newsreader: trn 4.0-test69 (20 September 1998) Originator: bonomi@diskless. (Robert Bonomi) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!feed2.onemain.com!feed1.onemain.com!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!152.163.239.129!portc01.blue.aol.com!howland.erols.net!newsfeed.lotsanews.com.MISMATCH!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59555 In article <8k9nv5$pre$5@bob.news.rcn.net>, wrote: > [.. munch ..] > >Did you guys ever have to change the name of a variable to >avoid figner fumbles? Does 'gave up, and changed the software to _accept_ the finger-fublem" count? If so.... This was a wonderful, _weird_, extremely effective, suite of applications. Totally batch oriented (actual *punch-card* submissions, no less). and included using the system _assembler_ as a "word processor". "Configuring" a customer consisted of a set of specifications, and some macroes. This was run through the assembler. The -output- of that pass was fed back into the assembler as *input*. *That* output was -again- fed to the assembler as input. The output from _that_ pass, was (yet again!) fed to the assembler (in _pieces, as it was source for some -seventeen- *different* programs) as input. At which time the actual 'object code' for each of those programs was produced. Trivia: If you wanted a literal quote-mark in the final program, it took (I think I've got this right) *TWENTY*SEVEN* consecutive occurances in the 'source' file. Anyway, part of the 'specifications' was what reports the customer wanted, and what the layout of them should look like. This involved a "FORM" directive, oddly enough. Several of the people preparing the specification sets for running had an absolutely *terrible* time getting that word right. Finally, the progammer 'gave up', and added another macro to the library. It did precisely _two_ things: 1) output a 'warning' message. The exact text of which was: "LOOK TOAD!! IT'S FORM, NOT FROM." 2) Invoked the 'FORM' macro with the same arguments that it, the "FROM" macro, had been called with. ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 10 Jul 2000 13:01:05 -1000 Organization: Canada France Hawai`i Telescope Lines: 13 Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> <3967d17d$1$wg$mr2ice@news.epix.net> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 963270065 14505 128.171.80.135 (10 Jul 2000 23:01:05 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 10 Jul 2000 23:01:05 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.direct.ca!look.ca!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59602 >>>>> "Jim" == Jim Saum writes: Jim> The 360-era MICR hardware (IBM 1419 sounds right, but I'm not sure) Jim> presented the 360 with a real-time constraint: after reading a check, Jim> the host program had a small number of milliseconds (I don't remember Jim> the precise number, but it was something like 3 or 5 ms) to decide Jim> which stacker pocket to drop the check into, or a default decision Jim> would be made by the hardware. A midrange 360 model was capable of Jim> processing only something like 200-1000 instructions before the Jim> deadline. Harumph. :-) It was called a 1419 when it was hooked to a 1401. And you didn't think the 1401 did real time work :-) ###### From: jcmorris@jmorris-pc.MITRE.ORG (Joe Morris) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 11 Jul 2000 12:44:55 GMT Organization: The MITRE Corporation Lines: 30 Message-ID: <8kf4s7$926$1@top.mitre.org> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> <3967d17d$1$wg$mr2ice@news.epix.net> NNTP-Posting-Host: jmorris-pc.mitre.org X-Trace: top.mitre.org 963319495 9286 128.29.251.13 (11 Jul 2000 12:44:55 GMT) X-Complaints-To: usenet@news.mitre.org NNTP-Posting-Date: 11 Jul 2000 12:44:55 GMT X-Newsreader: NN version 6.5.0 (NOV) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!newsfeed.stanford.edu!arclight.uoregon.edu!news.tufts.edu!blanket.mitre.org!news.mitre.org!jmorris-pc.MITRE.ORG!jcmorris Xref: chonsp.franklin.ch alt.folklore.computers:59496 Jim Thomas writes: >>>>>> "Jim" == Jim Saum writes: > Jim> The 360-era MICR hardware (IBM 1419 sounds right, but I'm not sure) > Jim> presented the 360 with a real-time constraint: after reading a check, > Jim> the host program had a small number of milliseconds (I don't remember > Jim> the precise number, but it was something like 3 or 5 ms) to decide > Jim> which stacker pocket to drop the check into, or a default decision > Jim> would be made by the hardware. A midrange 360 model was capable of > Jim> processing only something like 200-1000 instructions before the > Jim> deadline. >Harumph. :-) It was called a 1419 when it was hooked to a 1401. And you >didn't think the 1401 did real time work :-) The "1419" number strongly suggests that the machine was developed for use with a 14xx system (cf. the venerable 1403 printer) but the 1419 was definitely an S/360 device. And yes, it was at least *defined* under OS/360: a quick look in my archives turned up the 1971-vintage "OS/360 Field Engineering Handbook" where it shows that the last byte of the UCBTYP field has defined values of X'1D' for a 1419 "primary control unit", and X'1E' for a 1419 or 1275 "secondary control unit". Both are members of the device class "unit record" (UCBTYP byte 3 = X'08'). Never having worked in a banking shop I don't know how its timing restrictions were handled in an OS/360 environment. Joe Morris ###### From: "GerardS" Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> <3967d17d$1$wg$mr2ice@news.epix.net> <8kf4s7$926$1@top.mitre.org> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 53 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Message-ID: <7aKa5.97$qx4.588@newsfeed.slurp.net> Date: Tue, 11 Jul 2000 14:10:30 -0500 NNTP-Posting-Host: 208.149.16.236 X-Trace: newsfeed.slurp.net 963342083 208.149.16.236 (Tue, 11 Jul 2000 14:01:23 CDT) NNTP-Posting-Date: Tue, 11 Jul 2000 14:01:23 CDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.slurp.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59664 |Joe Morris wrote: | Jim Thomas writes: | >>>>>> "Jim" == Jim Saum writes: | > Jim> The 360-era MICR hardware (IBM 1419 sounds right, but I'm not sure) | > Jim> presented the 360 with a real-time constraint: after reading a check, | > Jim> the host program had a small number of milliseconds (I don't remember | > Jim> the precise number, but it was something like 3 or 5 ms) to decide | > Jim> which stacker pocket to drop the check into, or a default decision | > Jim> would be made by the hardware. A midrange 360 model was capable of | > Jim> processing only something like 200-1000 instructions before the | > Jim> deadline. | | >Harumph. :-) It was called a 1419 when it was hooked to a 1401. And you | >didn't think the 1401 did real time work :-) | | The "1419" number strongly suggests that the machine was developed for | use with a 14xx system (cf. the venerable 1403 printer) but the 1419 | was definitely an S/360 device. | | And yes, it was at least *defined* under OS/360: a quick look in my | archives turned up the 1971-vintage "OS/360 Field Engineering Handbook" | where it shows that the last byte of the UCBTYP field has defined | values of X'1D' for a 1419 "primary control unit", and X'1E' for a | 1419 or 1275 "secondary control unit". Both are members of the device | class "unit record" (UCBTYP byte 3 = X'08'). Never having worked in | a banking shop I don't know how its timing restrictions were handled | in an OS/360 environment. | Joe Morris _________________________ The 1419 was originally hooked up to the 14xx series of computers and later, the 360, and still later, the 370 series of computers. Under MFT/MVT, it had a special SVC (? or something from the SVCLIB) that had to be made resident to provide the necessary response (from the program in the CPU) to the 1419 as to which stacker pocket the check had to be sent to. I don't remember the timing involved, but it was tight. Frequently, the program running the check readind/sorting had to run in the top priority to get the response back to the 1419 in time, or else the checks went to the bit bucket (one of the default stacker pockets), and the "missed" check had to be re-read (re-sorted). Anyway, under VS/1. SVS, and MVS, that particular SVC "program" had to be made fixed resident. But because of longer path lengths and page faults, the CPU program running the 1419 had to run in a V=R region or partition, and also, of course, at a very high priority. By the way, at the bank, we called the 1419s the "trucks", both because of their operation and size. Gerard S. ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 11 Jul 2000 19:34:49 GMT Organization: The National Capital FreeNet Lines: 10 Message-ID: <8kfssp$rrm$1@freenet9.carleton.ca> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> <8kanhu$3pp$1@freenet9.carleton.ca> <3969F84E.8AB52B6B@thinkage.ca> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 963344089 28534 134.117.136.30 (11 Jul 2000 19:34:49 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 11 Jul 2000 19:34:49 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.datacomm.ch!newsmaster-01.atnet.at!atnet.at!newsrouter.chello.at!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newspeer1.nac.net!nntp.frontiernet.net!nntp.gctr.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59667 "Alan T. Bowler" (atbowler@thinkage.ca) writes: > > I thought the usual idiom for this was to use a translate to flip > the string left to right, and then use a translate and test to scan > for a non-blank. I never coded this I just heard it described. This is suggested in the PoO manual, BUT, there's a snag: the TRanslate only handles up to 256 bytes, making the method a PITA for longer strings. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Tue, 11 Jul 2000 23:42:11 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> <8kanhu$3pp$1@freenet9.carleton.ca> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.131.123 X-Original-NNTP-Posting-Host: 207.148.131.123 X-Trace: 11 Jul 2000 23:42:12 -0700, 207.148.131.123 Lines: 66 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeed.direct.ca!look.ca!nntp.cadvision.com!news.cadvision.com!207.148.131.123 Xref: chonsp.franklin.ch alt.folklore.computers:59692 On 9 Jul 2000 20:33:02 GMT, ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) wrote: >Joe Morris (jcmorris@jmorris-pc.MITRE.ORG) writes: > >>ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: >> >>>"Alan T. Bowler" (atbowler@thinkage.ca) writes: >>>> >>>> Non-DMA I/O is very rare in mainframe systems. (The PDP-10 line >>>> was something of an exception. However it eventually evolved >>>> something equivalent.) >> >>> RD (Read Direct), WD (Write Direct) IBM System 360 Principles of >>> Operations and follow-ons. Supervisor mode only. > > After I posted this, I cracked one of my PoOs to find that the actual > opcodes are RDD and WRD. I'm surprised that I got away with the above - > power of suggestion, or bit-rot amongst us geezergeeks? >> >>RD and WD were not available on all S/360 models, and AFAIK they were >>never standard features on models where they were available. As has >>been mentioned in other threads, the architecture documented in the >>"Principles of Operation" (PoO) manual were for the entire S/360 >>product line, including instructions that might be optional or >>unavailable on specific S/360 models. >> > Which brings back memories of MVCIN - move character inverse. At a > PPOE, there was a lot of PL/I text processing for typesetting. A huge > amount of CPU time was spent trimming off trailing blanks. Even with > VARYING length strings, it was possible to have blanks padding to the > maximum declared length. So when I spied MVCIN in the /370 PoO, I > figured that an assembler routine could flip the argument, then use > the PL/I VERIFY and SUBSTR builtin function to strip off leading blanks, > followed by another call to the MVCIN routine. Sadly, the 4381 didn't > have that instruction, and management wouldn't let me frig with the micro- > code floppy. > > I wound up writing an assembler routine named TRIMCV, for TRIM (as in > SNOBOL 4) Character Varying. The basic idea was to loop from the end > of the string EXecuting a CLCL (Compare Logical Character Long). Each > time through the loop, the compare address was decremented by the length > of the previous comparision, and the next compare length was doubled. > When a non-blank was detected, the compare length was reset to one, so > that one could sneak up to the last such character. Obviously, the > routine had to test for null and all-blank strings. The tremendous cost > savings didn't rescue the company. > > B-) IIRC can't you only EXecute an SS format instruction like CLC, but *execute* an RR format instruction like CLCL: doesn't the former modifies the length/count, but the latter (and here my memory may be fuzzy) increments two pointers and a limit count (in R0?), normally used only when you have to exceed 255 bytes by a chunk. For text processing, EXecuting a CLC would normally be the way to go. But you may have had to use CLCL to be able to scan backwards. Don't have a PoP handy offline right now, and it's been years since I did this stuff daily. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 12 Jul 2000 08:37:48 GMT Organization: Neosoft (using Airnews.net!) Lines: 19 Message-ID: <151C6219B2D908A2.A8321BA411AE8575.131358ED7C71228E@lp.airnews.net> X-Orig-Message-ID: References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> <8kanhu$3pp$1@freenet9.carleton.ca> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Wed Jul 12 03:37:48 2000 NNTP-Posting-Host: !b^441k-W="(nd, (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.datacomm.ch!newsmaster-01.atnet.at!atnet.at!newsrouter.chello.at!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:59632 On Tue, 11 Jul 2000 23:42:11 -0600, Brian Inglis wrote: >IIRC can't you only EXecute an SS format instruction like CLC, >but *execute* an RR format instruction like CLCL: doesn't the >former modifies the length/count, but the latter (and here my >memory may be fuzzy) increments two pointers and a limit count >(in R0?), normally used only when you have to exceed 255 bytes by >a chunk. No...EX modifies the second byte of any instruction (except for EX itself; if you try to EX an EX, you get a program check) by ORing it with the low order byte of the first operand register. EXing an SS instruction, in general, changes the length byte, and is quite common. EXing an RR instruction changes the registers used by the instruction. This is true for CLCL as well as things like LR, so you wind up selecting which of the various registers holds the address and length of the data to be compared. Only a fre instructions have implied register operands, and CLCL isn't one. ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 12 Jul 2000 11:10:54 GMT Organization: The National Capital FreeNet Lines: 36 Message-ID: <8khjnu$fue$1@freenet9.carleton.ca> References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <8k4lb6$9t3$1@top.mitre.org> <8kanhu$3pp$1@freenet9.carleton.ca> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 963400254 16334 134.117.136.30 (12 Jul 2000 11:10:54 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 12 Jul 2000 11:10:54 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.frontiernet.net!nntp.gctr.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59666 Brian Inglis (Brian.dot.Inglis@SystematicSw.ab.ca) writes: > > IIRC can't you only EXecute an SS format instruction like CLC, > but *execute* an RR format instruction like CLCL: doesn't the > former modifies the length/count, but the latter (and here my > memory may be fuzzy) increments two pointers and a limit count > (in R0?), normally used only when you have to exceed 255 bytes by > a chunk. Fuzzy indeed. B-) With my /370 PoO open beside me: MVC, CLC, TR, TRT handle 256 bytes max. EX can handle a 2, 4, 6 byte instruction on a halfword boundary EXcept EX. The assembler format EX R1,D2(X2,B2) specifies two operands: a target instruction ( ...2 ) and a modifier register, R1. "When the R1 field _is not zero_, [emph. mine] bits 8 to 15 [of the target] are ORed with bits 24 to 31 of [R1]" So it's the programmer's choice whether to modify the target or not. A further caution imposed by the transition from /360 to /370: some instructions, esp. MVCL and CLCL, are subject to refetch when they are interrupted. Therefore, R1 shouldn't designate an address or length register for those two, because the registers are updated at interrupt time. This would be a bitch to debug! P.S. The statements above are incomprehensible to those raised on the trash cranked out by Intel etc. Most of us start counting bits at 0, left to right. And sane architectures don't swap bytes when storing 2 or 4 bytes to memory. ###### From: "Hank Murphy" Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <8jqqv6$etb$1@top.mitre.org> <39623E02.78E9C854@thinkage.ca> <8jtglu$ft6$1@freenet9.carleton.ca> <39636F5D.1643E7EC@thinkage.ca> <3967d17d$1$wg$mr2ice@news.epix.net> <8kf4s7$926$1@top.mitre.org> <7aKa5.97$qx4.588@newsfeed.slurp.net> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 28 X-Newsreader: Microsoft Outlook Express 4.72.3110.1 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Message-ID: Date: Mon, 17 Jul 2000 03:17:45 GMT NNTP-Posting-Host: 209.178.148.89 X-Complaints-To: abuse@earthlink.net X-Trace: newsread2.prod.itd.earthlink.net 963803865 209.178.148.89 (Sun, 16 Jul 2000 20:17:45 PDT) NNTP-Posting-Date: Sun, 16 Jul 2000 20:17:45 PDT Organization: EarthLink Inc. -- http://www.EarthLink.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!colt.net!news.maxwell.syr.edu!newsfeed1.earthlink.net!newsmaster1.prod.itd.earthlink.net!newsread2.prod.itd.earthlink.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59869 GerardS wrote in message <7aKa5.97$qx4.588@newsfeed.slurp.net>... >|Joe Morris wrote: >| Jim Thomas writes: >| >>>>>> "Jim" == Jim Saum writes: >Under MFT/MVT, it had a special SVC (? or something from the SVCLIB) that >had to be made resident to provide the necessary response (from the program >in the CPU) to the 1419 as to which stacker pocket the check had to be sent >to. I don't remember the timing involved, but it was tight. BSAM - "stacker selection routine" or sometimes "pocket selection routine". These ran as PCI appendages, which had to be in SVCLIB. (PCI = Program Controlled Interrupt = an interrupt generated in the middle of a chain of I/O channel commands.) The successor to the 1419, the 3890, has the stacker selection code downloaded into the control unit, reducing the timing requirements. The host still has to read the buffer periodically, so it can still stop the sorter if it's too slow. ###### From: Charles Richmond Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Mon, 17 Jul 2000 00:34:57 -0700 Organization: Cannine Computer Center Lines: 23 Message-ID: <3972B721.317ED655@ev1.net> References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> Reply-To: richmond@ev1.net X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!news.maxwell.syr.edu!sjc-peer.news.verio.net!news.verio.net!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59916 robert@bonomi.invalid wrote: > > In article <8k9nv5$pre$5@bob.news.rcn.net>, wrote: > > > [.. munch ..] > > > > >Did you guys ever have to change the name of a variable to > >avoid figner fumbles? > > Does 'gave up, and had the progam _accept_ the mis-spelling' count? > Once I and a cow-orker were conspiring^h^h^hworking on a program together on a Unix workstation. There was some command (I do *not* remember which) that I just kept typing wrong. So I created an alias for the misspelling so it would work as well as the correct spelling...it really bothered the cow-orker to see me type it wrong, and have it work anyway. :-) -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 19 Jul 00 08:55:54 GMT Organization: UltraNet Communications, Inc. Lines: 26 Message-ID: <8l44qn$kdc$1@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> <87vgy3bqki.fsf@cartman.azz.net> X-Trace: dG1Y5DVqeXYZKqJ/UO/uZ54uCoqlnxzkSdMIdV/YTtw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 19 Jul 2000 11:52:55 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-32 Xref: chonsp.franklin.ch alt.folklore.computers:60036 In article <87vgy3bqki.fsf@cartman.azz.net>, Adam Sampson wrote: >Charles Richmond writes: > >> Once I and a cow-orker were conspiring^h^h^hworking on a program >> together on a Unix workstation. There was some command (I do *not* >> remember which) that I just kept typing wrong. So I created an >> alias for the misspelling so it would work as well as the correct >> spelling...it really bothered the cow-orker to see me type it wrong, >> and have it work anyway. :-) > >At my last job, I inherited a .profile with the following in it: > >alias amke=make >alias mkae=make >alias maek=make >alias meak=make > >and so on. > It's these damn keyboards; they can't keep up with fast typing. /BAH Subtract a hundred and four for e-mail. ###### Sender: azz@cartman.azz.net Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> From: Adam Sampson Message-ID: <87vgy3bqki.fsf@cartman.azz.net> Lines: 22 X-Newsreader: Gnus v5.6.45/XEmacs 21.1 - "20 Minutes to Nikko" Date: 18 Jul 2000 12:55:41 +0100 NNTP-Posting-Host: 212.159.21.249 X-Complaints-To: abuse@plus.net.uk X-Trace: stones 963947125 212.159.21.249 (Tue, 18 Jul 2000 20:05:25 BST) NNTP-Posting-Date: Tue, 18 Jul 2000 20:05:25 BST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news1.ebone.net!news.ebone.net!diablo.netcom.net.uk!netcom.net.uk!peer.news.th.u-net.net!u-net!btnet-peer0!btnet-feed1!btnet!landlord!stones.POSTED!cartman.azz.net!nobody Xref: chonsp.franklin.ch alt.folklore.computers:60054 Charles Richmond writes: > Once I and a cow-orker were conspiring^h^h^hworking on a program > together on a Unix workstation. There was some command (I do *not* > remember which) that I just kept typing wrong. So I created an > alias for the misspelling so it would work as well as the correct > spelling...it really bothered the cow-orker to see me type it wrong, > and have it work anyway. :-) At my last job, I inherited a .profile with the following in it: alias amke=make alias mkae=make alias maek=make alias meak=make and so on. -- Adam Sampson azz@gnu.org ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> <87vgy3bqki.fsf@cartman.azz.net> <8l44qn$kdc$1@bob.news.rcn.net> Subject: Re: Does the word "mainframe" still have a meaning? Lines: 30 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <964033561.976160@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-89.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Wed, 19 Jul 2000 15:02:09 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 964033290 204.92.64.17 (Wed, 19 Jul 2000 15:01:30 EDT) NNTP-Posting-Date: Wed, 19 Jul 2000 15:01:30 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!blackbush.xlink.net!newsfeed01.sul.t-online.de!t-online.de!newsfeed.online.be!sunqbc.risq.qc.ca!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60040 wrote in message news:8l44qn$kdc$1@bob.news.rcn.net... > > < snip > > It's these damn keyboards; they can't keep up with > fast typing. > My problem is the reverse . . . my digital dyslexia increases disproportionally to the size of the system it's attached to (and of course fat men wear gold watches). > /BAH > > Subtract a hundred and four for e-mail. Hmm . . . what keyboard is your favorite ? My personal preference is the action on the 3270 terminal and s/34 (solid, a bit clunky, Selectricish - no RSI worries) and . . . does any body know if there's a company extant that makes that sort of action for PCs ? (no hard kludges please) Rick ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Thu, 20 Jul 00 08:37:04 GMT Organization: UltraNet Communications, Inc. Lines: 39 Message-ID: <8l6o3l$279$2@bob.news.rcn.net> References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> <87vgy3bqki.fsf@cartman.azz.net> <8l44qn$kdc$1@bob.news.rcn.net> <964033561.976160@server16.cable.com> X-Trace: fOLxhNkyOavwv6Mw//IlufaLC3NwUo6rPPR4bz/+xFs= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 20 Jul 2000 11:34:13 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news.cs.utwente.nl!newspump.monmouth.com!newspeer.monmouth.com!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-245-231 Xref: chonsp.franklin.ch alt.folklore.computers:60104 In article <964033561.976160@server16.cable.com>, "P Linnane" wrote: > > wrote in message news:8l44qn$kdc$1@bob.news.rcn.net... >> > > >< snip > > >> It's these damn keyboards; they can't keep up with >> fast typing. >> > >My problem is the reverse . . . my digital dyslexia increases >disproportionally to the size of the system it's attached >to (and of course fat men wear gold watches). ROTFLMAO! Try ROTting your keyboard. >Hmm . . . what keyboard is your favorite ? >My personal preference is the action on the 3270 terminal >and s/34 (solid, a bit clunky, Selectricish - no RSI worries) My most favorite "keyboard" was really a terminal (why they've separated the two is beyond my old fartiness). DEC called it a VT06 (I don't remember the real manufacturer.) And, one of the best things about that terminal, besides the usable-ness of the keyboard) is that the video screen was very easy on the eyes...green rather than that damnedable blue that VT05s and VT50s had... I know, I know...VT05s weren't blue but they were to my eyes. In addition, I didn't see the screen flicker no matter how many flourescent(sp?) bulbs were above my head. /BAH Subtract a hundred and four for e-mail. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: Wed, 19 Jul 2000 21:00:51 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <8joosu$p0h@netaxs.com> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.137.228 X-Original-NNTP-Posting-Host: 207.148.137.228 X-Trace: 19 Jul 2000 21:00:52 -0700, 207.148.137.228 Lines: 28 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-was.dfn.de!nntp-out.monmouth.com!newspeer.monmouth.com!nntp.cadvision.com!news.cadvision.com!207.148.137.228 Xref: chonsp.franklin.ch alt.folklore.computers:60130 On Mon, 17 Jul 2000 00:34:57 -0700, Charles Richmond wrote: >robert@bonomi.invalid wrote: >> >> In article <8k9nv5$pre$5@bob.news.rcn.net>, wrote: >> > >> [.. munch ..] >> >> > >> >Did you guys ever have to change the name of a variable to >> >avoid figner fumbles? >> >> Does 'gave up, and had the progam _accept_ the mis-spelling' count? >> >Once I and a cow-orker were conspiring^h^h^hworking on a program >together on a Unix workstation. There was some command (I do *not* >remember which) that I just kept typing wrong. So I created an >alias for the misspelling so it would work as well as the correct >spelling...it really bothered the cow-orker to see me type it wrong, >and have it work anyway. :-) You mean like fgpre and mroe? Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Eric Chomko Newsgroups: alt.folklore.computers Subject: Re: Does the word "mainframe" still have a meaning? Date: 21 Jul 2000 18:33:13 GMT Organization: IDT Internet Services Lines: 25 Message-ID: <8la519$h1q@nnrp2.farm.idt.net> References: <8joosu$p0h@netaxs.com> <39617206.34902445@news.shuswap.net> <39635195.157080@news.shuswap.net> <3965dee3$0$8312$3c090ad1@news.plethora.net> <396757f4$0$8310$3c090ad1@news.plethora.net> <8k9nv5$pre$5@bob.news.rcn.net> <8kdf2e0p4u@enews3.newsguy.com> <3972B721.317ED655@ev1.net> <87vgy3bqki.fsf@cartman.azz.net> NNTP-Posting-Host: u3.farm.idt.net X-Newsreader: TIN [UNIX 1.3 unoff BETA release 961025] Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news.augsburg.net!news.idt.net!nntp.farm.idt.net!u3.farm.idt.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60198 Adam Sampson wrote: : Charles Richmond writes: : > Once I and a cow-orker were conspiring^h^h^hworking on a program : > together on a Unix workstation. There was some command (I do *not* : > remember which) that I just kept typing wrong. So I created an : > alias for the misspelling so it would work as well as the correct : > spelling...it really bothered the cow-orker to see me type it wrong, : > and have it work anyway. :-) : At my last job, I inherited a .profile with the following in it: : alias amke=make : alias mkae=make : alias maek=make : alias meak=make : and so on. WOW! The one I had was simply: alias mroe=more Eric