From owner-freebsd-hackers Thu May 15 13:19:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA12258 for hackers-outgoing; Thu, 15 May 1997 13:19:26 -0700 (PDT) Received: from superior.mooseriver.com (ppp030-sm0.sirius.com [205.134.229.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA12253 for ; Thu, 15 May 1997 13:19:23 -0700 (PDT) Received: (from jgrosch@localhost) by superior.mooseriver.com (8.8.5/8.8.5) id NAA20638; Thu, 15 May 1997 13:18:17 -0700 (PDT) From: Josef Grosch Message-Id: <199705152018.NAA20638@superior.mooseriver.com> Subject: Re: Cluster Computing in BSD In-Reply-To: <199705151725.KAA15126@phaeton.artisoft.com> from Terry Lambert at "May 15, 97 10:25:37 am" To: terry@lambert.org (Terry Lambert) Date: Thu, 15 May 1997 13:18:16 -0700 (PDT) Cc: hackers@FreeBSD.org Reply-To: jgrosch@sirius.com X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Terry Lambert said: >> > computers, but I can live with it :-). (With six boxes, a common >> > scientific process could take nearly 1/6 of the time on a fast network). >> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >> >> The difference between "could" and "does" is the >> reason for the failure of (nearly) every business unit that sold >> highly parallel/cluster systems. > >Except Goodyear. And Thinking Machines Corp. And Cray Computing. >And Cray Research. And Fujitsu. And... > >I think the list of successes so vastly outnumbers the list of >failures that your parenthetical "nearly" is *way* out of place >here. > > I am going to have to disagree with you here, Terry. While the above mentioned companys produced, in their time, insanely great machines (say the words "Cray XMP" to a fluid dynamics person such as my father and watch them drool) as a profit making company they have not fared as well post-cold war as they did during the Reagan buildup. Thinking Machines is out of business. Cray Computing lost money and was bought by SGI, mostly for their software. Cray Research was spun-off by Seymour Cray to develop the next generation of Crays. The capital to run Cray Research was raised based on Seymour Cray's reputation as one of the giants of this industry. With his untimely death this past Winter I would suspect that Cray Research will be bought up some time this year. Fujitsu is another story. Japanese corporations do not have the same set of fetishes that American corporations have. The business unit that produces the Fujitsu MMP is most likely losses money like it is going out of style. *BUT* Fujitsu the corporation has a license to print money. Japanese corporations do not expect every business unit to turn a profit every quarter. They are content to allow high visibility business units that showcase their technical mastery to loose money while the remainder of the organization does very well. Fujitsu's mainframe division is doing very well, eating Amdahl's lunch. I don't know much about the Goodyear machine but I am willing to wager that the American military and intelligence community helping to underwrite this division of Goodyear, a corporation that is doing very well without it computer division. Just my $0.02 Josef -- Josef Grosch | Another day closer to a | FreeBSD 2.2.1 jgrosch@sirius.com | Micro$oft free world | UNIX for the masses