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 mail4-relais-sop.national.inria.fr (mail4-relais-sop.national.inria.fr [192.134.164.105]) by yquem.inria.fr (Postfix) with ESMTP id 76CC5BBAF for ; Mon, 21 Dec 2009 17:18:47 +0100 (CET) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AkgAAA8uL0uYToAQkWdsb2JhbACbTgEBAQEJCwoHEwW6KIQuBA X-IronPort-AV: E=Sophos;i="4.47,431,1257116400"; d="scan'208";a="52605630" Received: from mailgate7.iss.soton.ac.uk ([152.78.128.16]) by mail4-smtp-sop.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-SHA; 21 Dec 2009 17:18:47 +0100 Received: from mailgate7.iss.soton.ac.uk (localhost.localdomain [127.0.0.1]) by mailgate7.iss.soton.ac.uk (8.13.8/8.13.4) with ESMTP id nBLGIH6J019930; Mon, 21 Dec 2009 16:18:17 GMT Received: from UOS-MSG00010-SI.soton.ac.uk (uos-msg00010-si.soton.ac.uk [152.78.119.33]) by mailgate7.iss.soton.ac.uk (mailgate7.iss.soton.ac.uk [152.78.128.16]) envelope-from with ESMTP id lBKGIH0968625938aU ret-id none; Mon, 21 Dec 2009 16:18:17 +0000 Received: from UOS-CL-EX7-L2.soton.ac.uk ([152.78.128.97]) by UOS-MSG00010-SI.soton.ac.uk ([152.78.119.33]) with mapi; Mon, 21 Dec 2009 16:18:17 +0000 From: "Fischbacher T." To: Mihamina Rakotomandimby , "caml-list@yquem.inria.fr" , "ocsigen@sympa.mancoosi.univ-paris-diderot.fr" Date: Mon, 21 Dec 2009 16:15:20 +0000 Subject: RE: [Caml-list] Re: multicore wish Thread-Topic: [Caml-list] Re: multicore wish Thread-Index: AcqCWDFk2zJrIBfDRYmwJ/2hQTE3GQAAJjsy Message-ID: <263BA63009B97B4CB8A65660DAD3D7D0651B38EAC6@UOS-CL-EX7-L2.soton.ac.uk> References: <4B2D2BC1.6020204@msu.edu> <200912200443.57698.jon@ffconsultancy.com> <874onko3mp.fsf_-_@frosties.localdomain>,<20091221171936.75d1186b@pbmiha.malagasy.com> In-Reply-To: <20091221171936.75d1186b@pbmiha.malagasy.com> Accept-Language: en-US, en-GB Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US, en-GB Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-smtpf-Report: client=relay,white,ipv6; mail=; rcpt= X-UOS-MailScanner-Information: Please contact Serviceline@soton.ac.uk for more information X-UOS-MailScanner-ID: nBLGIH6J019930 X-UOS-MailScanner: Found to be clean X-UOS-MailScanner-From: t.fischbacher@soton.ac.uk X-Spam: no; 0.00; beginners:01 caml-list:01 depends:04 trivially:05 anyway:05 useless:07 quite:08 memory:09 memory:09 actually:10 example:10 beginner:12 however:13 calculation:13 difference:14 Mihamina, > Ok, so for the beginner I am (must I ask on the beginners ML?): is > multicore support just useless or not? That *entirely* depends on what you want to do. If, for example, you=20 have to do a large calculation that is limited by memory and not by CPU, or, if you have an application that is trivially parallelized anyway,=20 multicore support won't make much of a difference. There are=20 (many) other applications, however, where it does matter quite a lot. Actually, the biggest effect of multicore architectures I see is to shift the emphasis from raw CPU power to memory bandwidth. --=20 Thomas