kjb@zycad.UUCP (Kevin Buchs) (09/03/87)
It was just announced today that Cray Research Cancelled the project Steve Chen was working on and he consequently resigned. Does anyone know what the project he was working on was? Y-MP?, or is that already completed? -- Kevin Buchs 3500 Zycad Dr. Oakdale, MN 55109 (612)779-5548 Zycad Corp. {rutgers,ihnp4,amdahl,umn-cs}!meccts!nis!zycad!kjb
schwartz@gondor.psu.edu (Scott E. Schwartz) (09/04/87)
In the Wall Street Journal yesterday (3-sept-87) there was an article that (as I recall) indicated that Chen left before the Y-MP was done. Apparently the next Cray will be the Cray-3.
perkins@bnrmtv.UUCP (Henry Perkins) (09/04/87)
In article <568@zycad.UUCP>, kjb@zycad.UUCP (Kevin Buchs) writes: > It was just announced today that Cray Research Cancelled the > project Steve Chen was working on and he consequently > resigned. Does anyone know what the project he was working > on was? Y-MP?, or is that already completed? Chen was working on the Cray MP project. This was supposed to be a "Massively Parallel" machine, with 64 or 128 processors as compared to current Cray machines with 2 or 4 processors. The project cost estimates had increased dramatically (over the $100 million mark), so Cray pulled the plug -- they figured they didn't have enough R & D money to support Chen's project for the mid-1990s and still work on short-term enhancements to their existing line of computers. For what it's worth, I think Cray made the right decision. Since the MP project required several fundamental breakthroughs the cost estimates were largely guesswork. It's pretty useless to be 90% done with your R & D and have your company go broke before you can finish. I hope Chen can continue elsewhere, but there are few computer companies with enough money to finance that project. -- {hplabs,amdahl,ames}!bnrmtv!perkins --Henry Perkins It is better never to have been born. But who among us has such luck? One in a million, perhaps.
nelson@ohlone.UUCP (Bron Nelson) (09/05/87)
In article <568@zycad.UUCP>, kjb@zycad.UUCP (Kevin Buchs) writes: > > It was just announced today that Cray Research Cancelled the > > project Steve Chen was working on and he consequently > > resigned. Does anyone know what the project he was working > > on was? Y-MP?, or is that already completed? > > schwartz@gondor.psu.edu (Scott E. Schwartz) > In the Wall Street Journal yesterday (3-sept-87) there was an > article that (as I recall) indicated that Chen left before the > Y-MP was done. Apparently the next Cray will be the Cray-3. This is inaccurate. The project that was cancelled was the MP, *not* the Y-MP. The MP project was the follow-on project to the Y-MP; the Y-MP is the follow-on to the X-MP. The Y-MP is still going on at full speed; it should be introduced long before the Cray-3. Steve Chen was the chief designer for the Y-MP project when it started, but he left the Y-MP project about 2 years ago to devote full time to the MP. The Y-MP project team is intact and working. To make a very long story much too short: As near as I can figure out, the board of directors (and the CEO in particular) felt that the MP project was getting out of control, was getting too expensive, and was too risky. They wanted the project scaled down and cut back. Steve and his group saw things differently and would not compromise the project. Eventually came down to the belief that it should be done right, or not at all, and faced with that, the decision was "not at all." Steve stills believes in the computer he was trying to design, and so is trying to put together his own company and the $100M+ to get it going. For its part, Cray Research has put Lester Davis in charge of picking up the pieces of the MP project. He will have the task of putting together the team of people who will re-define what the Y-MP follow-on project should be like (probably something cheaper and less ambitious). Now, despite the fact that this was very sudden and bad news, and despite the rather dramatic slide in the stock price, I really don't think this is going to have all that much effect on Cray Research as a corporation, and certainly not in the short term. The MP was still many years away from being built (it was not scheduled to be completed until after the Cray-3). The MP project only employed about 180 people out of the 4500 or so in the company; many if not most of those will be reabsorbed into Cray Research. There will be a period of confusion for awhile until the dust settles of course, but Les Davis is a very good guy, and will quite probably come up with a product re-definition in awhile (6 months?), and a re-constituted project (the MP' or something) will re-start. Cray is certainly committed to doing follow-on projects to the X-MP/Y-MP line, it just won't be what Steve Chen envisioned, and it will probably take awhile longer (an extra year?) due to the sever shock mid-stream. The long term effects are of course a lot harder to guage. Steve will no doubt take some of his people with him to his new company; they will be hard to replace. If he manages to put together a company and the $$$ and the machine and make it work, Cray Research will have (yet another) serious competitor in the field. On the other hand, if Steve's new company turns out like Trilogy, this will be hailed as one of the smartest moves Cray Research ever made. In summary as I see it: The MP project was a long-range project, several years from completion. Cancelling the MP project will have only minor repercussions in the short term. The long term effects are unclear. The Y-MP and Cray-3 products are unaffected by this. Unless you happen to be one of the 180 employees directly affected, there seems little cause for alarm. Steven Chen's new company will probably be fun to watch. Let me emphasize IN THE STRONGEST POSSIBLE TERMS that this was only MY opinions and MY speculation, and should NOT be construed to be an official position of Cray Research, nor even to be factually correct. ----------------------- Bron Nelson {ihnp4, lll-lcc!lll-tis}!ohlone!nelson Not the opinions of Cray Research (And if you care: Yes, I am one of the 180)