From: Alan Greig Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Clustering beginnings - Clair Grant might know. Date: Wed, 06 Mar 2002 09:28:38 +0000 Organization: http://extra.newsguy.com Lines: 93 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> NNTP-Posting-Host: p-971.newsdawg.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!netnews.com!feed2.news.rcn.net!rcn!newsfeed.cwix.com!newscon02.news.prodigy.com!prodigy.com!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews4 Xref: chonsp.franklin.ch alt.sys.pdp10:10631 On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" wrote: >Hmm, clustering in 1980 ? Is that accurate ? By 1982 I had seen a presentation on TOPS-20 clustering (CI, star couplers, hsc-50 controllers). I believe the TOPS-20 clustering shipped before that for VMS. but they could have been done in parallel. Unfortunately Dan Murphy's TOPS-20 clustering paper at http://www.opost.com/dlm/tenex/acmcfs89.txt does not give dates but I note that Clair Grant is credited as part of the TOPS-20 cluster development team and could perhaps provide an answer. Clair can you comment? Here's what it says at the end of Dan's paper: "Many other members of the TOPS-20 group were involved in completing, testing, and polishing the implementation, including Clair Grant, Judy Hall, Ron McLean, Dave Lomatire, Tom Moser, and others". Btw, does KLH-10 emulate any of the CI hardware? Can you cluster TOPS-20 systems under the emulator? >Dweeb >"Barry Treahy, Jr." wrote in message >news:3C83A7E5.7060208@mmaz.com... >> Bob Ceculski wrote: >> >> >isn't this ad 5 years too soon? >> > >> >http://www.theinquirer.net/mm25020202.htm >> > >> >Compaq TV ad to push Itanic >> > >> >But they're Alphas in Pittsburgh... >> >By Mike Magee, 03/03/2002 01:20:54 BST >> > >> >> Compaq seem to like doing these sorts of things. The recent Inform >> (Spring 2002) that came out spoke of how "Compaq invented clustering >> concept 22 years ago with OpenVMS clusters." Humm, does that mean when >> I purchased my home I invented garages? They may own the rights to it >> but it seems a bit disingenuous to take credit for the creation of it... >> >> Barry >> >> > >> >THE WORD ON THE advertising street is that Compaq is preparing a TV ad >> >that will have Alpha aficionados fainting in disbelief. >> >According to sources so close to Compaq that they're not allowed to >> >view the INQUIRER on their PCs, the ad features the Pittsburgh >> >Supercomputing facility. >> > >> >This facility, of course, demonstrates the fearsome power of the Alpha >> >in its high end incarnation. >> > >> >But, we understand, shots of the Pittsburgh site are interleaved with >> >pictures of an unlidded Compaq Itanium, which of course is shipping >> >now after that horrible SNAFU last year. >> > >> >The director of the center - which uses Alpha technology - is made to >> >say something along the lines that the incredible computation power of >> >his installation proves the superiority of the Itanium processor. >> > >> >Which of course it doesn't. And won't for a fair old while. >> > >> >Meanwhile, intrepid girl reporter Eva Glass tells us that the >> >marketing department(s) at Q are very much afraid that if the HP-Q >> >takeover goes ahead on April Fool's day, most will go, go, go. >> > >> >That, she understands, means that little in the way of marketing is >> >going on, the CVs/resumes are flying like confetti, while "cover your >> >ass" is the order of the day. >> > >> >But is having the words Compaq Marketing Specialist on your CV a good >> >recommendation for a future job, Eva mischievously asks. >> > >> > >> >> -- >> >> Barry Treahy, Jr * Midwest Microwave * Vice President & CIO >> >> E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028 >> >> >> > -- Alan ###### From: "Edward C. Bailey" Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: 06 Mar 2002 11:32:44 -0500 Organization: Red Hat, Inc. Lines: 35 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> NNTP-Posting-Host: 172.16.52.200 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: stan.redhat.com 1015432541 7181 172.16.52.200 (6 Mar 2002 16:35:41 GMT) X-Complaints-To: abuse@redhat.com NNTP-Posting-Date: 6 Mar 2002 16:35:41 GMT User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!news.maxwell.syr.edu!logbridge.uoregon.edu!usenet01.sei.cmu.edu!news.cis.ohio-state.edu!news.dfci.harvard.edu!news.ccs.neu.edu!news.redhat.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10625 >>>>> "Alan" == Alan Greig writes: Alan> On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" Alan> wrote: >> Hmm, clustering in 1980 ? Is that accurate ? Alan> By 1982 I had seen a presentation on TOPS-20 clustering (CI, star Alan> couplers, hsc-50 controllers). I believe the TOPS-20 clustering Alan> shipped before that for VMS. but they could have been done in Alan> parallel. Well, in '81 I was working on Galaxy support in MR1, and I recall seeing both "NI" and "CI" cables/hardware in the LCG lab, so it certainly existed at that time. Before and after this time I worked with VMS and I can tell you that "real" clustering under VMS (ie, shared read/write file acess between different cluster members) was quite late -- the functionality that VMS first supported was simply device access over CI. No shared read/write was possible because VMS hadn't moved away from ACP processes towards the distributed lock manager approach they eventually implemented to arbitrate shared device access. My recollection was that TOPS-20 clustering did ship before VMS had "real" clustering; whether it shipped before VMS claimed CI device support, I do not know. My cynical side wonders if VMS' CI device support was rushed out to claim "first support", but I have no data to back that up. It may have been as simple as the fact that the hardware was ready to be sold, and the VAX product line had no way to generate sales on this stuff until VMS provided at least some level of support for the hardware... Ed -- Ed Bailey Red Hat, Inc. http://www.redhat.com/ ###### From: jmfbahciv@aol.com Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 02 08:58:58 GMT Organization: UltraNet Communications, Inc. Lines: 64 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> X-Trace: UmFuZG9tSVZk/bWME6m6FdP124x4EKIDh5VrWFfiP1fJSoaSt3IDXXIedAY2ZHzW X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 7 Mar 2002 11:16:22 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-102 Xref: chonsp.franklin.ch alt.sys.pdp10:10641 In article <3C86B841.FB66242@bellatlantic.net>, bad bob wrote: > > >"Edward C. Bailey" wrote: >> >> >>>>> "Alan" == Alan Greig writes: >> >> Alan> On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" >> Alan> wrote: >> >> >> Hmm, clustering in 1980 ? Is that accurate ? >> >> Alan> By 1982 I had seen a presentation on TOPS-20 clustering (CI, star >> Alan> couplers, hsc-50 controllers). I believe the TOPS-20 clustering >> Alan> shipped before that for VMS. but they could have been done in >> Alan> parallel. >> >> Well, in '81 I was working on Galaxy support in MR1, and I recall seeing >> both "NI" and "CI" cables/hardware in the LCG lab, so it certainly existed >> at that time. >> >> Before and after this time I worked with VMS and I can tell you that "real" >> clustering under VMS (ie, shared read/write file acess between different >> cluster members) was quite late -- the functionality that VMS first >> supported was simply device access over CI. No shared read/write was >> possible because VMS hadn't moved away from ACP processes towards the >> distributed lock manager approach they eventually implemented to arbitrate >> shared device access. >> >> My recollection was that TOPS-20 clustering did ship before VMS had "real" >> clustering; whether it shipped before VMS claimed CI device support, I do >> not know. My cynical side wonders if VMS' CI device support was rushed out >> to claim "first support", but I have no data to back that up. It may have >> been as simple as the fact that the hardware was ready to be sold, and the >> VAX product line had no way to generate sales on this stuff until VMS >> provided at least some level of support for the hardware... >> >That is my recollection too. The HSC stuff was one of the keys for >CI connected storage, and that set included various models that >transitioned to COlorado in the late 70s iirc, I recall Richie >Lary coming in to help with 11/60 microcode, and DMC/KMC microcode >around 76 from ColoSprings. I admit my dates may be off a bit, >maybe a year, but... The clustering project was separate from the CI-HSC implementation. TW did the CI-HSC on both TOPS-10 and TOPS-20. The -10 was first, then he did the -20. Look at the dates on the monitor sources and you'll get an idea of when he did his work. I didn't think that the clustering project was dependent on the CI being done (but I could be very wrong...it was hard enough keeping track of the -10 bit flows). /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 02 12:01:58 GMT Organization: UltraNet Communications, Inc. Lines: 44 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> X-Trace: UmFuZG9tSVZVt4g/7IT1l9GbO+UitWypHMESSGgWXknNrl9CKPt/byOZuCq9vT2R X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 7 Mar 2002 14:19:20 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!netnews.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-97-194 Xref: chonsp.franklin.ch alt.sys.pdp10:10642 In article <3C8760BE.85FA077C@bellatlantic.net>, bad bob wrote: > > >jmfbahciv@aol.com wrote: >> >> In article <3C86B841.FB66242@bellatlantic.net>, >> bad bob wrote: >><> >> >That is my recollection too. The HSC stuff was one of the keys for >> >CI connected storage, and that set included various models that >> >transitioned to COlorado in the late 70s iirc, I recall Richie >> >Lary coming in to help with 11/60 microcode, and DMC/KMC microcode >> >around 76 from ColoSprings. I admit my dates may be off a bit, >> >maybe a year, but... >> >> The clustering project was separate from the CI-HSC implementation. >> TW did the CI-HSC on both TOPS-10 and TOPS-20. The -10 was first, >> then he did the -20. Look at the dates on the monitor sources >> and you'll get an idea of when he did his work. I didn't think >> that the clustering project was dependent on the CI being done >> (but I could be very wrong...it was hard enough keeping track >> of the -10 bit flows). >> >> /BAH >Barb, >You jogged my memory too, I believe you are correct, there was >all the SMP stuff, clustering (Loosely Coupled Systems is the >term that comes to mind again), and all that going on in >what my memory recalls as overlapping activities. Clustering was the -20 group's answer to our SMP. You wouldn't believe the "my thingie's better than your thingie" arguments. > .. It really >was a busy time! Yup. It was one of the good side effects of canceling Jupiter. We got got funding for projects that had nothing to do with hardware. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 02 12:50:52 GMT Organization: UltraNet Communications, Inc. Lines: 41 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> X-Trace: UmFuZG9tSVZEWmtGbBDX0VxpGV9zxckMeU8GMNiwAPj7lD/jCjypv499bob2lulj X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 7 Mar 2002 15:08:14 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-97-194 Xref: chonsp.franklin.ch alt.sys.pdp10:10647 In article <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com>, Alan Greig wrote: >On Thu, 07 Mar 2002 12:43:09 GMT, bad bob >wrote: > > >>Barb, >>You jogged my memory too, I believe you are correct, there was >>all the SMP stuff, clustering (Loosely Coupled Systems is the >>term that comes to mind again), and all that going on in >>what my memory recalls as overlapping activities. It really >>was a busy time! > >And TOPS-10/TOPS-20 customers were getting lots of presentations on >all the upcoming goodies and how this would all scream along with the >new "Jupiter" hardware. Nope. That's marketing. It didn't affect software development... much. We're talking about actual projects we were doing and selling. > >Then it ended. > >So am I right in summing up, consensus is that the Compaq >(Digital/DEC) claim to have invented 'modern' clustering on VMS in >1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS >around that time period? Not TOPS-10. We did SMP instead. TOPS-20 did clustering to achieve their brand of multi-CPU configurations. /BAH > > Subtract a hundred and four for e-mail. ###### Newsgroups: comp.os.vms,alt.sys.pdp10 Distribution: world X-Newsreader: mxrn 6.18-32B References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Reply-To: hoffman@xdelta.zko.dec.nospam Subject: Re: Clustering beginnings - Clair Grant might know. Lines: 23 Message-ID: Date: Wed, 06 Mar 2002 22:38:46 GMT NNTP-Posting-Host: 16.32.80.251 X-Complaints-To: abuse@Compaq.com X-Trace: news.cpqcorp.net 1015454326 16.32.80.251 (Wed, 06 Mar 2002 14:38:46 PST) NNTP-Posting-Date: Wed, 06 Mar 2002 14:38:46 PST Organization: Compaq Computer Corporation Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed.cwix.com!wn2feed!wn3feed!worldnet.att.net!198.6.0.123!uunet!sac.uu.net!news.compaq.com!news.cpqcorp.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10651 .. The following is specific to OpenVMS VAX -- to VAX/VMS, as it was then known -- clustering... CI780 support was first in VAX/VMS V3.1 (1983), HSC50 in V3.3 (1983), full (write-shared) clustering was in V4.0 (1984). Between V3.3 and prior to V4.0, the VAX/VMS releases permitted storage configurations with one-write-accessor-multiple-readers. (At least one regular reader has been using this "ancient" storage configuration for reasons not germaine to this discussion, and the reader encountered problems with the behaviour of the Fibre Channel storage and mount verification. But I digress.) Please follow the "history" links in the FAQ for related details and historical information. ---------------------------- #include ----------------------------- For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com --------------------------- pure personal opinion --------------------------- Hoff (Stephen) Hoffman OpenVMS Engineering hoffman#xdelta.zko.dec.com ###### Message-ID: <3C86B841.FB66242@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 46 Date: Thu, 07 Mar 2002 00:44:23 GMT NNTP-Posting-Host: 138.88.49.225 X-Complaints-To: business-support@verizon.com X-Trace: nwrddc02.gnilink.net 1015461863 138.88.49.225 (Wed, 06 Mar 2002 19:44:23 EST) NNTP-Posting-Date: Wed, 06 Mar 2002 19:44:23 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news.imp.ch!news.stealth.net!news.stealth.net!news-out.visi.com!hermes.visi.com!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!nwrddc02.gnilink.net.POSTED!53ab2750!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10648 "Edward C. Bailey" wrote: > > >>>>> "Alan" == Alan Greig writes: > > Alan> On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" > Alan> wrote: > > >> Hmm, clustering in 1980 ? Is that accurate ? > > Alan> By 1982 I had seen a presentation on TOPS-20 clustering (CI, star > Alan> couplers, hsc-50 controllers). I believe the TOPS-20 clustering > Alan> shipped before that for VMS. but they could have been done in > Alan> parallel. > > Well, in '81 I was working on Galaxy support in MR1, and I recall seeing > both "NI" and "CI" cables/hardware in the LCG lab, so it certainly existed > at that time. > > Before and after this time I worked with VMS and I can tell you that "real" > clustering under VMS (ie, shared read/write file acess between different > cluster members) was quite late -- the functionality that VMS first > supported was simply device access over CI. No shared read/write was > possible because VMS hadn't moved away from ACP processes towards the > distributed lock manager approach they eventually implemented to arbitrate > shared device access. > > My recollection was that TOPS-20 clustering did ship before VMS had "real" > clustering; whether it shipped before VMS claimed CI device support, I do > not know. My cynical side wonders if VMS' CI device support was rushed out > to claim "first support", but I have no data to back that up. It may have > been as simple as the fact that the hardware was ready to be sold, and the > VAX product line had no way to generate sales on this stuff until VMS > provided at least some level of support for the hardware... > That is my recollection too. The HSC stuff was one of the keys for CI connected storage, and that set included various models that transitioned to COlorado in the late 70s iirc, I recall Richie Lary coming in to help with 11/60 microcode, and DMC/KMC microcode around 76 from ColoSprings. I admit my dates may be off a bit, maybe a year, but... bob > Ed > -- > Ed Bailey Red Hat, Inc. http://www.redhat.com/ ###### Message-ID: <3C8760BE.85FA077C@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 33 Date: Thu, 07 Mar 2002 12:43:09 GMT NNTP-Posting-Host: 138.88.49.225 X-Complaints-To: business-support@verizon.com X-Trace: nwrddc02.gnilink.net 1015504989 138.88.49.225 (Thu, 07 Mar 2002 07:43:09 EST) NNTP-Posting-Date: Thu, 07 Mar 2002 07:43:09 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!nwrddc02.gnilink.net.POSTED!53ab2750!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10650 jmfbahciv@aol.com wrote: > > In article <3C86B841.FB66242@bellatlantic.net>, > bad bob wrote: ><> > >That is my recollection too. The HSC stuff was one of the keys for > >CI connected storage, and that set included various models that > >transitioned to COlorado in the late 70s iirc, I recall Richie > >Lary coming in to help with 11/60 microcode, and DMC/KMC microcode > >around 76 from ColoSprings. I admit my dates may be off a bit, > >maybe a year, but... > > The clustering project was separate from the CI-HSC implementation. > TW did the CI-HSC on both TOPS-10 and TOPS-20. The -10 was first, > then he did the -20. Look at the dates on the monitor sources > and you'll get an idea of when he did his work. I didn't think > that the clustering project was dependent on the CI being done > (but I could be very wrong...it was hard enough keeping track > of the -10 bit flows). > > /BAH Barb, You jogged my memory too, I believe you are correct, there was all the SMP stuff, clustering (Loosely Coupled Systems is the term that comes to mind again), and all that going on in what my memory recalls as overlapping activities. It really was a busy time! bob > Subtract a hundred and four for e-mail. ###### From: Alan Greig Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 2002 14:15:45 +0000 Organization: http://extra.newsguy.com Lines: 25 Message-ID: <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> NNTP-Posting-Host: p-989.newsdawg.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews1 Xref: chonsp.franklin.ch alt.sys.pdp10:10656 On Thu, 07 Mar 2002 12:43:09 GMT, bad bob wrote: >Barb, >You jogged my memory too, I believe you are correct, there was >all the SMP stuff, clustering (Loosely Coupled Systems is the >term that comes to mind again), and all that going on in >what my memory recalls as overlapping activities. It really >was a busy time! And TOPS-10/TOPS-20 customers were getting lots of presentations on all the upcoming goodies and how this would all scream along with the new "Jupiter" hardware. Then it ended. So am I right in summing up, consensus is that the Compaq (Digital/DEC) claim to have invented 'modern' clustering on VMS in 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS around that time period? -- Alan ###### Message-ID: <3C877D43.52090610@evms.zko.dec.com> From: Clair Grant X-Mailer: Mozilla 4.51 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 38 Date: Thu, 07 Mar 2002 09:46:27 -0500 NNTP-Posting-Host: 16.32.248.241 X-Complaints-To: abuse@Compaq.com X-Trace: news.cpqcorp.net 1015512568 16.32.248.241 (Thu, 07 Mar 2002 06:49:28 PST) NNTP-Posting-Date: Thu, 07 Mar 2002 06:49:28 PST Organization: Compaq Computer Corporation Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!news-hub.siol.net!zur.uu.net!ash.uu.net!sac.uu.net!news.compaq.com!news.cpqcorp.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10644 I believe that TOPS-20 did demo at DECUS multiple host operating systems sharing files on the CI prior to the first public VMS demo. I think VMS was ready with straight HSC storage first. Although it never became a product we had VMS and TOPS-20 systems on the same CI in the lab. They knew how to keep out of each other's way when it came to files and we could do DECnet over the CI between the operating systems. I always thought this had great potential. Some of the things we used to experiment with in those days on the CI have become reality in today's fibre channel SAN environments. Clair Hoff Hoffman wrote: > .. > > The following is specific to OpenVMS VAX -- to VAX/VMS, as it was > then known -- clustering... > > CI780 support was first in VAX/VMS V3.1 (1983), HSC50 in V3.3 (1983), > full (write-shared) clustering was in V4.0 (1984). Between V3.3 and > prior to V4.0, the VAX/VMS releases permitted storage configurations > with one-write-accessor-multiple-readers. (At least one regular > reader has been using this "ancient" storage configuration for reasons > not germaine to this discussion, and the reader encountered problems > with the behaviour of the Fibre Channel storage and mount verification. > But I digress.) > > Please follow the "history" links in the FAQ for related details > and historical information. > > ---------------------------- #include ----------------------------- > For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com > --------------------------- pure personal opinion --------------------------- > Hoff (Stephen) Hoffman OpenVMS Engineering hoffman#xdelta.zko.dec.com ###### From: Nic Clews Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 2002 15:03:56 +0000 Organization: CSC Computer Sciences Lines: 20 Message-ID: <3C87815C.49A97A73@127.0.0.1> References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> NNTP-Posting-Host: 20.9.166.247 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: lore.csc.com 1015513501 24911 20.9.166.247 (7 Mar 2002 15:05:01 GMT) X-Complaints-To: abuse@news.csc.com NNTP-Posting-Date: Thu, 7 Mar 2002 15:05:01 +0000 (UTC) X-Mailer: Mozilla 4.5 [en] (WinNT; I) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!193.154.160.102.MISMATCH!newsfeed.Austria.EU.net!newsfeed.kpnqwest.at!news-hub.siol.net!zur.uu.net!ash.uu.net!lore.csc.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10655 Alan Greig wrote: > > So am I right in summing up, consensus is that the Compaq > (Digital/DEC) claim to have invented 'modern' clustering on VMS in > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS > around that time period? Just for interest, in the preface of "VAXCLUSTER PRINCIPLES" that I am totally indebted to Steve Reece for the copy, Roy G. Davis writes: "In the early 1980s, Digital Equipment Corporation developed an entirely new technology for multiprocessor computing called clustering. Digital released the first implementation of this technology in late 1983. Known as the VAXcluster configuration, ..." March 25th 1993. -- Regards, Nic Clews CSC Computer Sciences nclews at csc dot com ###### From: young_r@encompasserve.org (Rob Young) Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: 7 Mar 2002 10:20:05 -0600 Organization: Encompasserve Lines: 47 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> <3C87815C.49A97A73@127.0.0.1> NNTP-Posting-Host: eisner.encompasserve.org X-Trace: grandcanyon.binc.net 1015518006 467 192.135.80.34 (7 Mar 2002 16:20:06 GMT) X-Complaints-To: abuse@binc.net NNTP-Posting-Date: Thu, 7 Mar 2002 16:20:06 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!news.binc.net!young_r Xref: chonsp.franklin.ch alt.sys.pdp10:10653 In article <3C87815C.49A97A73@127.0.0.1>, Nic Clews writes: > Alan Greig wrote: >> >> So am I right in summing up, consensus is that the Compaq >> (Digital/DEC) claim to have invented 'modern' clustering on VMS in >> 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS >> around that time period? > > Just for interest, in the preface of "VAXCLUSTER PRINCIPLES" that I am > totally indebted to Steve Reece for the copy, Roy G. Davis writes: > > "In the early 1980s, Digital Equipment Corporation developed an entirely > new technology for multiprocessor computing called clustering. Digital > released the first implementation of this technology in late 1983. Known > as the VAXcluster configuration, ..." > This is very strange... back to that great work: "VAX OpenVMS at 20", edited by Andy Goldstein, on page 60: VMS to OpenVMS: Major Releases VMS V3 April 1982 - 10,000 license (no mention of clustering) VMS V4 September 1984 - 40,000 licenses o Support for new processor - VAX 8600, MV I/II, (v4.1) VAXstation I/II (v4.1) o VAXclusters o Connection manager o Distributed lock manager o Distributed file system (F11BXQP) o Security enhancements o Command line editing and command recall o LAT server o ACL implented o Cluster wide operator control o Variable prompt strings --- So maybe it was internal then, but not out in the wild until September 1984. Assuming Andy got the ship date right and none of this was picked up on prior to publication. Rob ###### Message-ID: <3C8795F4.174ED987@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 29 Date: Thu, 07 Mar 2002 16:30:09 GMT NNTP-Posting-Host: 138.88.49.225 X-Complaints-To: business-support@verizon.com X-Trace: nwrddc01.gnilink.net 1015518609 138.88.49.225 (Thu, 07 Mar 2002 11:30:09 EST) NNTP-Posting-Date: Thu, 07 Mar 2002 11:30:09 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!newsfeed1.cidera.com!Cidera!cyclone1.gnilink.net!spamfinder.gnilink.net!nwrddc01.gnilink.net.POSTED!53ab2750!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10649 Alan Greig wrote: > > On Thu, 07 Mar 2002 12:43:09 GMT, bad bob > wrote: > > >Barb, > >You jogged my memory too, I believe you are correct, there was > >all the SMP stuff, clustering (Loosely Coupled Systems is the > >term that comes to mind again), and all that going on in > >what my memory recalls as overlapping activities. It really > >was a busy time! > > And TOPS-10/TOPS-20 customers were getting lots of presentations on > all the upcoming goodies and how this would all scream along with the > new "Jupiter" hardware. > > Then it ended. > > So am I right in summing up, consensus is that the Compaq > (Digital/DEC) claim to have invented 'modern' clustering on VMS in > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS > around that time period? I believe your statement of summary is factual, accurate and correct. bob > -- > Alan ###### From: Nic Clews Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: Thu, 07 Mar 2002 17:45:03 +0000 Organization: CSC Computer Sciences Lines: 33 Message-ID: <3C87A71F.C5A7EDA@127.0.0.1> References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> <3C87815C.49A97A73@127.0.0.1> NNTP-Posting-Host: 20.9.166.247 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: lore.csc.com 1015523169 312 20.9.166.247 (7 Mar 2002 17:46:09 GMT) X-Complaints-To: abuse@news.csc.com NNTP-Posting-Date: Thu, 7 Mar 2002 17:46:09 +0000 (UTC) X-Mailer: Mozilla 4.5 [en] (WinNT; I) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!uunet!dca.uu.net!ash.uu.net!lore.csc.com!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10654 Rob Young wrote: > > This is very strange... back to that great work: "VAX OpenVMS > at 20", edited by Andy Goldstein, on page 60: Tremendous book. Has the PDF vanished from the website now? > VMS to OpenVMS: Major Releases > VMS V4 September 1984 - 40,000 licenses ... > o VAXclusters > o Connection manager > o Distributed lock manager > o Distributed file system (F11BXQP) ... > o Cluster wide operator control > > So maybe it was internal then, but not out in the wild until September > 1984. Assuming Andy got the ship date right and none of this was > picked up on prior to publication. Putting Clair's response together with some other elements in VAXcluster Principles, device communication (HSC) is vitally important in clustering to Davis. The book goes into fascinating detail about device communication / DSA. On the other hand, it could be a typo! September is late (ish) in the year by my estimation, but there's just a fingers width between 3 and 4. I'm speculating wildly so I'd probably better quit! -- Regards, Nic Clews CSC Computer Sciences nclews at csc dot com ###### From: young_r@encompasserve.org (Rob Young) Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: 7 Mar 2002 13:36:40 -0600 Organization: Encompasserve Lines: 52 Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> <3C87815C.49A97A73@127.0.0.1> <3C87A71F.C5A7EDA@127.0.0.1> NNTP-Posting-Host: eisner.encompasserve.org X-Trace: grandcanyon.binc.net 1015529802 3837 192.135.80.34 (7 Mar 2002 19:36:42 GMT) X-Complaints-To: abuse@binc.net NNTP-Posting-Date: Thu, 7 Mar 2002 19:36:42 +0000 (UTC) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.icl.net!iad-peer.news.verio.net!news.verio.net!newsfeed.cwix.com!news.binc.net!young_r Xref: chonsp.franklin.ch alt.sys.pdp10:10676 In article <3C87A71F.C5A7EDA@127.0.0.1>, Nic Clews writes: > Rob Young wrote: >> >> This is very strange... back to that great work: "VAX OpenVMS >> at 20", edited by Andy Goldstein, on page 60: > > Tremendous book. Has the PDF vanished from the website now? > >> VMS to OpenVMS: Major Releases >> VMS V4 September 1984 - 40,000 licenses > ... >> o VAXclusters >> o Connection manager >> o Distributed lock manager >> o Distributed file system (F11BXQP) > ... >> o Cluster wide operator control >> >> So maybe it was internal then, but not out in the wild until September >> 1984. Assuming Andy got the ship date right and none of this was >> picked up on prior to publication. > > Putting Clair's response together with some other elements in VAXcluster > Principles, device communication (HSC) is vitally important in > clustering to Davis. The book goes into fascinating detail about device > communication / DSA. > > On the other hand, it could be a typo! September is late (ish) in the > year by my estimation, but there's just a fingers width between 3 and 4. > Yeah but... V3 shipped April 1982, "no way" to ship that close on such a major rev level (i.e. September 1983). To clarify (or muddle) a bit more.. if you turn to page 46: "In May 1983, DIGITAL announced VAXClusters" page 47: "However with the advent of the MicroVAX and VAX workstations (concurrent with clusters in 1984)" So it appears to have actually shipped September 1984 (page 60) but was announced 16 months prior! http://www.openvms.compaq.com/openvms/20th/vmsbook.pdf Still active, just tried it. Rob ###### Message-ID: <3C87C806.4B856512@evms.zko.dec.com> From: Clair Grant X-Mailer: Mozilla 4.51 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 35 Date: Thu, 07 Mar 2002 15:05:26 -0500 NNTP-Posting-Host: 16.32.248.241 X-Complaints-To: abuse@Compaq.com X-Trace: news.cpqcorp.net 1015531708 16.32.248.241 (Thu, 07 Mar 2002 12:08:28 PST) NNTP-Posting-Date: Thu, 07 Mar 2002 12:08:28 PST Organization: Compaq Computer Corporation Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-out.visi.com!hermes.visi.com!uunet!ash.uu.net!sac.uu.net!news.compaq.com!news.cpqcorp.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10670 No, I don't think that would be a completely fair statement. What we had on TOPS-20 was not a fully distributed file system and lock manager which is what I think of as the crux of a VMS cluster and what distinguishes it in the market place. To achieve the equivalent of that on TOPS-20 left a whole lot as an exercise for the user to figure out. Clair Alan Greig wrote: > On Thu, 07 Mar 2002 12:43:09 GMT, bad bob > wrote: > > >Barb, > >You jogged my memory too, I believe you are correct, there was > >all the SMP stuff, clustering (Loosely Coupled Systems is the > >term that comes to mind again), and all that going on in > >what my memory recalls as overlapping activities. It really > >was a busy time! > > And TOPS-10/TOPS-20 customers were getting lots of presentations on > all the upcoming goodies and how this would all scream along with the > new "Jupiter" hardware. > > Then it ended. > > So am I right in summing up, consensus is that the Compaq > (Digital/DEC) claim to have invented 'modern' clustering on VMS in > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS > around that time period? > > -- > Alan ###### From: Paul Repacholi Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: 08 Mar 2002 05:08:39 +0800 Organization: iQnet Lines: 17 Sender: prep@k9.prep.synonet.com Message-ID: <87zo1kuluw.fsf@k9.prep.synonet.com> References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> NNTP-Posting-Host: news-01.core.usertools.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: nnrp.waia.asn.au 1015535777 27098 202.154.80.9 (7 Mar 2002 21:16:17 GMT) X-Complaints-To: usenet@nnrp.waia.asn.au NNTP-Posting-Date: Thu, 7 Mar 2002 21:16:17 +0000 (UTC) User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 Cache-Post-Path: angelina!unknown@p132.sw03.auto.usertools.net X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.clara.net!heighliner.fr.clara.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!newsfeed.r-kom.de!fu-berlin.de!newsfeed.iinet.net.au!nntp.waia.asn.au!nnrp.waia.asn.au!!nobody Xref: chonsp.franklin.ch alt.sys.pdp10:10674 Alan Greig writes: > So am I right in summing up, consensus is that the Compaq > (Digital/DEC) claim to have invented 'modern' clustering on VMS in > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by > VMS around that time period? Digital or DEC. Compaq could not find an original idea if you rammed it up their arse and gave them a map. I don't think TOPS-10 ever did clustering. 20s did with CFS, but who remembers them... -- Paul Repacholi 1 Crescent Rd., +61 (08) 9257-1001 Kalamunda. West Australia 6076 Raw, Cooked or Well-done, it's all half baked. EPIC, The Architecture of the future, always has been, always will be. ###### Message-ID: <3C87C6D2.105615AB@trailing-edge.com> Date: Thu, 07 Mar 2002 20:00:18 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> <3C86B841.FB66242@bellatlantic.net> <3C8760BE.85FA077C@bellatlantic.net> <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com> <87zo1kuluw.fsf@k9.prep.synonet.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 35 NNTP-Posting-Host: 63.73.218.130 X-Trace: 1015549219 reader2.ash.ops.us.uu.net 27493 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.clara.net!heighliner.fr.clara.net!news-hub.siol.net!ams.uu.net!dca.uu.net!ash.uu.net!spool0900.news.uu.net!reader0902.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10673 Paul Repacholi wrote: > > Alan Greig writes: > > > So am I right in summing up, consensus is that the Compaq > > (Digital/DEC) claim to have invented 'modern' clustering on VMS in > > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by > > VMS around that time period? > > Digital or DEC. Compaq could not find an original idea if you rammed > it up their arse and gave them a map. I don't think TOPS-10 ever did > clustering. 20s did with CFS, but who remembers them... The archives here do. The GALAXY V5 docs show that CI and CFS support were in the "mainstream" release TOPS-20 kernel in 1984: COPYRIGHT (C), 1984, DIGITAL EQUIPMENT CORPORATION 1.0 SUMMARY GALAXY version 5 is a release primarily aimed at supporting changes in TOPS-20 version 6. The major areas include QUEUE% JSYS support, CI support, CFS support and password encryption support. Due to these needs, a significant amount of work was also done to MOUNTR to improve its internal operations. Finally, GALAXY 5 also contains published changes to GALAXY 4.2. See http://pdp-10.trailing-edge.com/bb-h138e-bm_tops20_v6_1_distr/01/6-1-documentation/galaxy.doc.html Tim. ###### From: Rich Alderson Newsgroups: comp.os.vms,alt.sys.pdp10 Subject: Re: Clustering beginnings - Clair Grant might know. Date: 08 Mar 2002 22:14:58 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 30 Sender: alderson+news@panix2.panix.com Message-ID: References: <3C83A7E5.7060208@mmaz.com> <1j7h8.297$Nw2.2894@news.get2net.dk> NNTP-Posting-Host: panix2.panix.com X-Trace: reader2.panix.com 1015643697 12876 166.84.1.2 (9 Mar 2002 03:14:57 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: Sat, 9 Mar 2002 03:14:57 +0000 (UTC) X-Newsreader: Gnus v5.7/Emacs 20.7 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!out.nntp.be!propagator-SanJose!in.nntp.be!news-out.visi.com!hermes.visi.com!nycmny1-snh1.gtei.net!news.gtei.net!panix!not-for-mail Xref: chonsp.franklin.ch alt.sys.pdp10:10690 Alan Greig writes: > On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" wrote: >> Hmm, clustering in 1980 ? Is that accurate ? > By 1982 I had seen a presentation on TOPS-20 clustering (CI, star couplers, > hsc-50 controllers). I believe the TOPS-20 clustering shipped before that for > VMS. but they could have been done in parallel. In 1982, that presentation would have been related to Jupiter--I saw a similar one at UChicago in 1983. So copyright dates on software will be misleading WRT when clusters *really* became available. We field-tested Tops-20 v6.0 at Stanford LOTS (and found a lot of places where it simply sucked boulders--6.1 came out by the time the field-test ended); this included getting 2 HSC-50's, 6 RA-81's, and a star coupler, and running Blue Peril(TM) under the floor next to the 3Mbit Ethernet (of infamous yellow hue). This included the new NEBULA part of GALAXY, for inter-system communication via the CI. Since I was there for it, it can't have happened any earlier than 1 Oct 1984-- and I was actually in place at LOTS for a month or two before it got under way. I *ran* the field test for v7.0 at Stanford. (Nobody else wanted the job--or the OS.) -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless