From owner-freebsd-smp Thu Sep 16 17:45:29 1999 Delivered-To: freebsd-smp@freebsd.org Received: from csvax.cs.caltech.edu (csvax.cs.caltech.edu [131.215.131.131]) by hub.freebsd.org (Postfix) with ESMTP id 4CA4F157EA for ; Thu, 16 Sep 1999 17:45:18 -0700 (PDT) (envelope-from mika@varese.cs.caltech.edu) Received: from varese.cs.caltech.edu (varese.cs.caltech.edu [131.215.78.28]) by csvax.cs.caltech.edu (8.9.1/8.9.1) with ESMTP id RAA23133; Thu, 16 Sep 1999 17:45:16 -0700 (PDT) Received: from localhost.cs.caltech.edu (localhost.cs.caltech.edu [127.0.0.1]) by varese.cs.caltech.edu (8.8.7/8.7.3) with SMTP id RAA08427; Thu, 16 Sep 1999 17:45:15 -0700 (PDT) Message-Id: <199909170045.RAA08427@varese.cs.caltech.edu> X-Authentication-Warning: varese.cs.caltech.edu: localhost.cs.caltech.edu [127.0.0.1] didn't use HELO protocol To: "William R. Somsky" Cc: freebsd-smp@freebsd.org Subject: Re: Examples of FreeBSD SMP success? In-reply-to: Your message of "Thu, 16 Sep 1999 17:17:51 PDT." <19990916171751.A19950@annwn.phys.washington.edu> Date: Thu, 16 Sep 1999 17:45:15 -0700 From: Mika Nystrom Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello there, I can report that FreeBSD's SMP works great for computation-intensive jobs. We routinely run several copies of spice(-like) circuit simulators that use several hundred megabytes of RAM each and monopolize one CPU each. The efficiency, as long as you are not doing I/O, seems to be 95+% (little different from using two machines for this). We have been doing this for years on dual processor 200 MHz Pentium Pro machines. The response time of interactive jobs and other jobs that do not fall into the category of entirely computation-intensive has improved a great deal in that time (as FreeBSD's SMP has come a long way in this regard, even if it hasn't advanced so much "under the hood"). Regards, Mika Nystrom Asynchronous Systems Architecture Project Department of Computer Science California Institute of Technology "William R. Somsky" writes: >We're considering getting several Intel-based machines here in >the UW Physics department for use as desktop work- and computation- >stations, and the question has come up whether we would gain much >by getting dual-CPU (say dual 500 MHZ Pentium-III) as opposed to >single-CPU systems. > >The anticipated use where dual-CPUs could help us would probably >be users either running a computation-intensive job (eg, Mathematica), >while simultaneously doing desktop editing/browsing/mailing/TeXing/etc, >or running two computation- intensive jobs. (We don't expect that >Mathematica or any user job will be multi threaded.) > >What I've been asked to find out what the state of FreeBSD SMP >support is, and if anyone has any real-world examples of using >dual-CPUs under FreeBSD that might be similar to this sort of >situation and what the results have been. > >Being as that I've not tried multi-processing under FreeBSD yet, >does anybody have any input I can give to my users? > >________________________________________________________________________ >Dr. William R. Somsky, Unix Mgr somsky@phys.washington.ed >u >Department of Physics, Box 351560 B432 Physics-Astro Bldg >Univ. of Washington, Seattle WA 98195-1560 206/616-2954 > > >To Unsubscribe: send mail to majordomo@FreeBSD.org >with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message