From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on yquem.inria.fr X-Spam-Level: X-Spam-Status: No, score=0.0 required=5.0 tests=AWL autolearn=disabled version=3.1.3 Received: from mail2-relais-roc.national.inria.fr (mail2-relais-roc.national.inria.fr [192.134.164.83]) by yquem.inria.fr (Postfix) with ESMTP id 871EFBBCA for ; Tue, 13 May 2008 03:24:18 +0200 (CEST) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AjwCAMqKKEjUnw7XcGdsb2JhbACCMY9gAQwFAgQHE5lk X-IronPort-AV: E=Sophos;i="4.27,476,1204498800"; d="scan'208";a="10642058" Received: from fhw-relay07.plus.net ([212.159.14.215]) by mail2-smtp-roc.national.inria.fr with ESMTP; 13 May 2008 03:24:18 +0200 Received: from [80.229.56.224] (helo=beast.local) by fhw-relay07.plus.net with esmtp (Exim) id 1JvjFZ-0003S2-KH for caml-list@yquem.inria.fr; Tue, 13 May 2008 02:24:17 +0100 From: Jon Harrop Organization: Flying Frog Consultancy Ltd. To: caml-list@yquem.inria.fr Subject: Re: [Caml-list] Re: Why OCaml rocks Date: Tue, 13 May 2008 02:19:23 +0100 User-Agent: KMail/1.9.9 References: <200805090139.54870.jon@ffconsultancy.com> <20080512132223.GA24858@annexia.org> <1210639362.478.6.camel@flake.lan.gerd-stolpmann.de> In-Reply-To: <1210639362.478.6.camel@flake.lan.gerd-stolpmann.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200805130219.23224.jon@ffconsultancy.com> X-Plusnet-Relay: daae3963474512d869a950d71c32df40 X-Spam: no; 0.00; ocaml:01 gerd:01 stolpmann:01 ocaml:01 unoptimized:01 runtime:01 speedup:01 12%:98 blog:98 blog:98 frog:98 wrote:01 caml-list:01 seems:03 overhead:04 On Tuesday 13 May 2008 01:42:42 Gerd Stolpmann wrote: > Am Montag, den 12.05.2008, 14:22 +0100 schrieb Richard Jones: > > This is just barely faster than Jon's OCaml version using message > > passing (12% faster on my test machine[0]). Which just seems to show > > that the overhead of message passing _isn't_ the problem here[1]. > > I've just written my own distributed version. You find my comments and > timings here: > > http://blog.camlcity.org/blog/parallelmm.html > > The code is here: > > https://godirepo.camlcity.org/svn/lib-ocamlnet2/trunk/code/examples/rpc/mat >rixmult/ > > In this (very unoptimized) multiplier message passing accounts for ~25% > of the runtime. Even for 2 cores there is already a speedup. 10 cores > (over a network) are about 4 times faster than a single core without > message passing. For what values of "n"? -- Dr Jon D Harrop, Flying Frog Consultancy Ltd. http://www.ffconsultancy.com/products/?e