re: http://www.garlic.com/~lynn/2014.html#16
ibm press from 17feb1992, "scientific and technical only" http://www.garlic.com/~lynn/2001n.html#6000clusters2 more ibm press 11may1992, "company caught by surprise by national lab interest in cluster computing http://www.garlic.com/~lynn/2001n.html#6000clusters2 however, old email about doing benchmarks for LLNL interested in cluster compute farm of 70 4341s http://www.garlic.com/~lynn/2006y.html#email790220 other old 4341 email http://www.garlic.com/~lynn/lhwemail.html#4341 1980, I get con'ed into doing channel-extender support for IBM STL (now silicon valley lab) that was remoting 300 people from the IMS group to offsite bldg. An effort to release it to customers is blocked by group in POK that have been playing with some serial stuff and they were afraid it might make it more difficult to get what they were doing released. some past posts http://www.garlic.com/~lynn/submisc.html#channel.extender mid-80s, I'm trying to get project started that would have large number of 370 & 801/risc chips in racks. at the same time i'm working with NSF on interconnecting the NSF supercomputer centers (morphs into NSFNET backbone, the precursor to modern internet). old email about schedule conflict between internal meeting and making presentation to director of NSF http://www.garlic.com/~lynn/2011b.html#email850314 .. http://www.garlic.com/~lynn/2007d.html#email850315 .. eventually NSF releases an RFP, but internal politics prevent us from bidding. Director of NSF tries to help by writing company a letter (with support from other gov. agencies), but that just makes the internal politics worse. other old NSF related email http://www.garlic.com/~lynn/lhwemail.html#nsfnet 1988, LLNL has some serial stuff and I'm asked if I can help them get it out as industry standard ... eventually morphs into the fibre-channel standard. Even tho by then it is obsolete, the POK group finally gets their serial stuff released in 1990 with ES/9000 as ESCON. 1992, we are using the LLNL fibre-channel standard stuff as part of the ha/cmp commercial cluster scaleup ... old ref http://www.garlic.com/~lynn/95.html#13 in some of the collection of old cluster scaleup email http://www.garlic.com/~lynn/lhwemail.html#medusa the last one is about cluster computing meeting at LLNL (very end of jan1992). I'm unable to make the meeting so a friend at another vendor fills in for me and then comes by later to fill me in on what happened (only hrs before we are told that cluster scaleup has been transferred and we can't work on anything with more than four processors) http://www.garlic.com/~lynn/2006x.html#email920129 some POK engineers eventually get involved in the fibre-channel stuff, defining a heavyweight protocol layer (that significantly cuts the native thruput) which eventually morphs into FICON ... some past posts http://www.garlic.com/~lynn/submisc.html#ficon -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
