From owner-freebsd-hackers Tue Aug 18 15:22:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA12814 for freebsd-hackers-outgoing; Tue, 18 Aug 1998 15:22:42 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from vip.consys.com (Comobabi.ConSys.COM [209.141.107.6]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA12787 for ; Tue, 18 Aug 1998 15:22:32 -0700 (PDT) (envelope-from rcarter@pinyon.org) Received: (from pinyon@localhost) by vip.consys.com (8.9.1/8.9.1) id PAA14610; Tue, 18 Aug 1998 15:21:50 -0700 (MST) Date: Tue, 18 Aug 1998 15:21:50 -0700 (MST) From: "Russell L. Carter" Message-Id: <199808182221.PAA14610@vip.consys.com> To: ghelmer@scl.ameslab.gov, Matthew.Alton@anheuser-busch.com Subject: Re: DIY Supercomputers Cc: Hackers@FreeBSD.ORG In-Reply-To: Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG |The biggest objection I've heard to using FreeBSD for parallel processing |clusters was the lack of a FreeBSD version of a certain commercial Fortran |compiler. FreeBSD (with net.inet.tcp.delack_enabled=0 in 3.0-current) |ought to have better network performance over the entire spectrum of |message sizes than Linux 2.0.x, so I want to see whether this has an |observable effect on parallel apps. The NAS Parallel Benchmarks are quite network intensive, there may be numbers for "Beowulf" clusters by now. There weren't when I was running on the DAISy cluster at Sandia (which ran FreeBSD at the time, '94-95). The DAISy numbers I measured for these were the best price/performance of *any* system at the time. Russell To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message