From owner-freebsd-questions Mon May 10 6:24:28 1999 Delivered-To: freebsd-questions@freebsd.org Received: from kermit.bertrandt.com (kermit.bertrandt.com [195.30.30.5]) by hub.freebsd.org (Postfix) with ESMTP id B5A0D15130 for ; Mon, 10 May 1999 06:24:23 -0700 (PDT) (envelope-from Michael.Samer@ingolstadt.bertrandt.com) Received: from ho-sv-ex1.bertrandt.com (unverified) by kermit.bertrandt.com (Content Technologies SMTPRS 2.0.15) with ESMTP id for ; Mon, 10 May 1999 15:24:55 +0200 Received: by ho-sv-ex1 with Internet Mail Service (5.5.2232.9) id ; Mon, 10 May 1999 15:31:21 +0200 Message-Id: From: "Samer, Michael, IN" To: "FreeBSD Questions (E-Mail)" Subject: Cluster?! Date: Sun, 9 May 1999 21:04:36 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2232.9) Content-Type: text/plain Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hello BSD Freaks, after a long time of "seduction", I have the possibilty to solve a cute little problem of my company: we are using a porgram called pamcrash on an SGI Octane R12000 which tooks appr. three weeks until it's finished calculating (matrix's!!!). But we're getting more and more work for this program, and the only possibility to solve it is to use more CPU's in this machine. My "inspiration" was to took instead of spending 100.000 US$ in SGI for an 8 CPU version in a Alpha Cluster (beginning with 4*Dual-Alpha 500) with NetBSD or FreeBSD machines. The producer of PamCrash told me that there is an BSD version available. The question is, how and if I can distribute this work to different machines (via PVM (or as far as I remember sth. like CA....) to lower the calculation time to one day by using more and more powerful machines). PamCrash is splitt-able. Has anyone of you worked with PVM or whatever or with distributed computing (apart from rsh)? Does the software sees how many machines are working on the solution (copyright problems, the licence is linked with the CPU ID), and how does it work best together? Is the Alpha SMP Version already running stable (V3.1-FBSD) or is NetBSD better here? Does NBSD or FBSD already supporting the 21264 Alpha , known as EV.6)? Is a F-Ethernet enough to cluster? Does anyone have experience (on universities or research) with Cluster and the efficency? PLEASE MAil to me directely, because I'm just in my country FBSD group/mailer. sam.vanratt@gmx.net or michael.samer@ingolstadt.bertrandt.com Greetings Sam To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message