From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Original-To: caml-list@yquem.inria.fr Delivered-To: caml-list@yquem.inria.fr Received: from mail1-relais-roc.national.inria.fr (mail1-relais-roc.national.inria.fr [192.134.164.82]) by yquem.inria.fr (Postfix) with ESMTP id D27ECBC57 for ; Tue, 16 Nov 2010 20:07:50 +0100 (CET) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AqICAJpl4kxiizT2kWdsb2JhbACUYI4WAQEBAQkLCgcRAx/AEoVLBIRaiQo X-IronPort-AV: E=Sophos;i="4.59,207,1288566000"; d="scan'208";a="87506212" Received: from nm28-vm0.bullet.mail.ac4.yahoo.com ([98.139.52.246]) by mail1-smtp-roc.national.inria.fr with SMTP; 16 Nov 2010 20:07:50 +0100 Received: from [98.139.52.188] by nm28.bullet.mail.ac4.yahoo.com with NNFMP; 16 Nov 2010 19:07:49 -0000 Received: from [98.139.52.152] by tm1.bullet.mail.ac4.yahoo.com with NNFMP; 16 Nov 2010 19:07:49 -0000 Received: from [127.0.0.1] by omp1035.mail.ac4.yahoo.com with NNFMP; 16 Nov 2010 19:07:49 -0000 X-Yahoo-Newman-Id: 255319.5324.bm@omp1035.mail.ac4.yahoo.com Received: (qmail 87466 invoked from network); 16 Nov 2010 19:07:49 -0000 Received: from [192.168.0.2] (norm@76.21.112.227 with plain) by smtp101.sbc.mail.ac4.yahoo.com with SMTP; 16 Nov 2010 11:07:48 -0800 PST X-Yahoo-SMTP: 6wM9MhqswBCJxy9jDcYebttHhXX021oP4RUg3aXahuvJwcSQLA-- X-YMail-OSG: OmkYB6sVM1mDZXEZuQL4jo2YakTXGsZCGjaIAwTAkm1axp9 FleaaTnb2_rTEsQDJKowm4YEvAJ4bGjKglrDIw4vo50lENhxeV.b2u_3yiqQ XKKLjp9AGshC6HCnLg_jNZlRtOG1juHLgz0UY2oMgIHipn7GNcclt5tCe5UZ FjH_QdkFGh5uwCB3X6sG40S8.Z7N2ZXJ7pTf6RIpxxwvaLGvxDMVRnn5heve aE5jVQ8gkZZruSIolwuUPsmalaPcmMikz2oiea1q6RmZBvyTmVg26A0iyYdO FP4oc4zEoxUH9pS99buYjTDRmxBwusFGyp_xeNe8P5GSeMQGjJA-- X-Yahoo-Newman-Property: ymail-3 References: <20101115182737.42b8dcae@loki.yggdrasil.draxit.de> <4CE228CA.3030503@gmail.com> In-Reply-To: <4CE228CA.3030503@gmail.com> Mime-Version: 1.0 (Apple Message framework v1082) Content-Type: text/plain; charset=us-ascii Message-Id: <89ECD76E-17EF-4DB5-A2C1-4B1EE19D16E9@cap-lore.com> Content-Transfer-Encoding: quoted-printable From: Norman Hardy Subject: Re: [Caml-list] SMP multithreading Date: Tue, 16 Nov 2010 11:07:45 -0800 To: caml-list@yquem.inria.fr X-Mailer: Apple Mail (2.1082) X-Spam: no; 0.00; ocaml:01 ocaml:01 pointers:01 edgar:98 wrote:01 caml-list:01 immutable:01 data:02 optimized:04 depends:04 passing:05 passing:05 computing:05 shared:06 style:93 On 2010 Nov 15, at 22:46 , Edgar Friendly wrote: > It looks like high-performance computing of the near future will be = built out of many machines (message passing), each with many cores = (SMP). One could use message passing for all communication in such a = system, but a hybrid approach might be best for this architecture, with = use of shared memory within each box and message passing between. Of = course the best choice depends strongly on the particular task. >=20 > In the long run, it'll likely be a combination of a few large, = powerful cores (Intel-CPU style w/ the capability to run a single thread = as fast as possible) with many many smaller compute engines (GPGPUs or = the like, optimized for power and area, closely coupled with memory) = that provides the highest performance density. OCaml code should be able to share immutable OCaml data with other = processes just as it shares libraries. See http://cap-lore.com/Software/pch.html . Some of the ideas there might be improved with hardware support. Admission: If I had read all of the interesting pointers given on this = thread I would never finish sending this e-mail.=