From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id SAA16913; Tue, 15 Jul 2003 18:02:40 +0200 (MET DST) Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id SAA17150 for ; Tue, 15 Jul 2003 18:02:39 +0200 (MET DST) Received: from nitro.isi.edu (nitro.isi.edu [128.9.208.207]) by nez-perce.inria.fr (8.11.1/8.11.1) with ESMTP id h6FG2bT00720; Tue, 15 Jul 2003 18:02:37 +0200 (MET DST) Received: from moussor.isi.edu (moussor.isi.edu [128.9.208.41]) by nitro.isi.edu (8.11.6p2/8.11.2) with ESMTP id h6FG2Vd21782; Tue, 15 Jul 2003 09:02:31 -0700 (PDT) Received: from localhost (hdaume@localhost) by moussor.isi.edu (8.9.3/8.8.6) with ESMTP id JAA20220; Tue, 15 Jul 2003 09:02:31 -0700 (PDT) X-Authentication-Warning: moussor.isi.edu: hdaume owned process doing -bs Date: Tue, 15 Jul 2003 09:02:31 -0700 (PDT) From: Hal Daume III To: Xavier Leroy cc: Daniel Andor , caml-list@inria.fr Subject: Re: [Caml-list] ocaml on supercomputer? In-Reply-To: <20030715172436.A7268@pauillac.inria.fr> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Loop: caml-list@inria.fr X-Spam: no; 0.00; caml-list:01 threads:01 exploited:01 mpi:01 props:99 bindings:01 ocaml:01 nodes:02 ported:02 i'd:06 basically:06 minor:07 performs:07 runs:07 xeon:09 Sender: owner-caml-list@pauillac.inria.fr Precedence: bulk > OCaml threads do not exploit shared-memory multiprocessing -- > basically, all threads interleave their execution, hence multiple > processors are not exploited. Your best bet is MPI or PVM. Both have > bindings for OCaml. I thought I'd just stick in my own props for the Ocaml MPI bindings. I had an ~11k line ocaml program which I ported to MPI. It only required 500 or so more lines of code to MPI-ify it (with minor restructuring so these parts could be separated out). It now runs on our supercomputer (56th faster computer in the world, according to top500.org) on anywhere from 10 to 170 nodes (IBM Netfinity Cluster P4 Xeon 2 GHz - Myrinet/ 512), and performs incredibly well. I am *very* happing with these bindings and would like to formally thank Xavier for writing them :). Best of luck. - Hal ------------------- To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/ Beginner's list: http://groups.yahoo.com/group/ocaml_beginners