From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Original-To: caml-list@sympa.inria.fr Delivered-To: caml-list@sympa.inria.fr Received: from mail2-relais-roc.national.inria.fr (mail2-relais-roc.national.inria.fr [192.134.164.83]) by sympa.inria.fr (Postfix) with ESMTPS id 6F0BF7EE4B for ; Sat, 5 Oct 2013 03:25:14 +0200 (CEST) Received-SPF: None (mail2-smtp-roc.national.inria.fr: no sender authenticity information available from domain of philippe.wang.lists@gmail.com) identity=pra; client-ip=74.125.82.42; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="philippe.wang.lists@gmail.com"; x-sender="philippe.wang.lists@gmail.com"; x-conformance=sidf_compatible Received-SPF: Pass (mail2-smtp-roc.national.inria.fr: domain of philippe.wang.lists@gmail.com designates 74.125.82.42 as permitted sender) identity=mailfrom; client-ip=74.125.82.42; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="philippe.wang.lists@gmail.com"; x-sender="philippe.wang.lists@gmail.com"; x-conformance=sidf_compatible; x-record-type="v=spf1" Received-SPF: None (mail2-smtp-roc.national.inria.fr: no sender authenticity information available from domain of postmaster@mail-wg0-f42.google.com) identity=helo; client-ip=74.125.82.42; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="philippe.wang.lists@gmail.com"; x-sender="postmaster@mail-wg0-f42.google.com"; x-conformance=sidf_compatible X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AtYBANdpT1JKfVIqm2dsb2JhbABZgz9Swi8IFg4BAQEBAQYLCwkUKIIlAQEFQAE4AQMMAQUFCwMCCC4iEgEFARwGE4d0Aw+eHI9chC8nDYllAQUMjgyFYwOOaYkYjEyBZ4FhGCmBY4JsO4E1 X-IPAS-Result: AtYBANdpT1JKfVIqm2dsb2JhbABZgz9Swi8IFg4BAQEBAQYLCwkUKIIlAQEFQAE4AQMMAQUFCwMCCC4iEgEFARwGE4d0Aw+eHI9chC8nDYllAQUMjgyFYwOOaYkYjEyBZ4FhGCmBY4JsO4E1 X-IronPort-AV: E=Sophos;i="4.90,1037,1371074400"; d="scan'208";a="35653469" Received: from mail-wg0-f42.google.com ([74.125.82.42]) by mail2-smtp-roc.national.inria.fr with ESMTP/TLS/RC4-SHA; 05 Oct 2013 03:25:12 +0200 Received: by mail-wg0-f42.google.com with SMTP id m15so2830952wgh.5 for ; Fri, 04 Oct 2013 18:25:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date:message-id:subject :from:to:cc:content-type; bh=Fh+GW8fuTJMb0JHsYgVvRAdsAGAnOin/j6B84kd/KwA=; b=IyAOUnruJEoPzbOKLAsJg2Q0jed5foBPHFBPYw5rF+BpJcpiGPEUnlxTbsrCJkh8SV zGrtbKjz7nXy1aV7Da9s9rZEXgEjzTHDeOxWdTSb/1r/RkqLmFDd0yPsFlsxgbKG2EsO 14VuupzAddV7loYMWKk51H6HUD0evUuN1zHqz5ScekjihB/dhxUua51lvS8R7TWxRp8u hFsl11ApoUmeN+FB8EqYC8GdxywQnOgXNPsie5WWltx/uC4SQC8cca3AIjoiU4pDXRHG NzoFdZDdcW4iJqeYul6dPA1XFB+0hLT66z0X64ubCIJSatbBPu2g7zt5nV/ZX6+S4jqE Vabw== MIME-Version: 1.0 X-Received: by 10.180.72.226 with SMTP id g2mr9651615wiv.52.1380936313617; Fri, 04 Oct 2013 18:25:13 -0700 (PDT) Sender: philippe.wang.lists@gmail.com Received: by 10.216.42.134 with HTTP; Fri, 4 Oct 2013 18:25:13 -0700 (PDT) In-Reply-To: <20131005022537.20edf93e@narfi.yggdrasil.draxit.de> References: <20131004014649.2b4e7750@narfi.yggdrasil.draxit.de> <201310041544.r94FiR1h085910@givry.fdupont.fr> <20131005022537.20edf93e@narfi.yggdrasil.draxit.de> Date: Sat, 5 Oct 2013 02:25:13 +0100 X-Google-Sender-Auth: 9CBcDM_KniJQ4x4KSMmd_tiinvI Message-ID: From: Philippe Wang To: Wolfgang Draxinger Cc: OCaml Mailing List , Francis Dupont , Mathias Bourgoin Content-Type: text/plain; charset=ISO-8859-1 Subject: Re: [Caml-list] Current state of threading parallelism? On Sat, Oct 5, 2013 at 1:25 AM, Wolfgang Draxinger wrote: > On Fri, 04 Oct 2013 17:44:27 +0200 > Francis Dupont wrote: > >> => yes, this is the official way (no shared memory, no fine grain >> locking, etc). Note this was the way followed by Python too for >> the same reasons (to make the intepreter multi-threaded was considered >> as too hard for the expected best result, i.e., ~4 time maximum speed >> up). > > Well, there are certain problems that can be largely parallelized in a > way that performance scales well with the number of cores. Those > problems usually also perform in constant memory (of course this is all > mutable state then, so not really pure). I'm thinking of problems that > could be put as well on GPUs here, like signal decomposition, error > diffusion models, stuff like that. You might want to get a look at SPOC: http://www.algo-prog.info/spoc/ N.B. It's even in OPAM! -- Philippe Wang mail@philippewang.info