From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: 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 81BE1BC57 for ; Tue, 16 Nov 2010 21:35:44 +0100 (CET) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AsYEADt64kzRVdW0k2dsb2JhbACGPY10hicBhzZNCBUBAQEBCQkKCREDH6YNiWKCGIUOLohZAQEDBYVGBIpY X-IronPort-AV: E=Sophos;i="4.59,207,1288566000"; d="scan'208";a="78644264" Received: from mail-yx0-f180.google.com ([209.85.213.180]) by mail4-smtp-sop.national.inria.fr with ESMTP; 16 Nov 2010 21:35:32 +0100 Received: by yxj4 with SMTP id 4so673695yxj.39 for ; Tue, 16 Nov 2010 12:35:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type; bh=H8Bv89B1GoNbzEkp/JUW1SPzmvz751O+SRrqydKkrI4=; b=DHDxH+/0AlwMhwMFewrbX8+CCEFGFfQDr076NSLFSopZtYQJU45v8Qyqt3SC12F6aN dk5+s5+pH73XJETXC+Ycpro4zwMSn9cA6YiNp8PRJvR3RDPGyfhRcsYR/caG6s1WhH1d cfou3KQXoNhUwKVJ1zY5uLZ6/Q5ACBQ/8af70= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=gYJhmzL9FdZ7VNgG+6Wg0cfeyOix99hZQNWmr3ZlzfOmsBBD8cbybOCzHeEFVeefEQ s/jHcggf0qyAZPJxiYGChoYQ4iOQ3gLnUdO5BEqu0Msd4/s66GHyp5Yj8yGCQhh4gKvz DsVIv9kSYFEa6ir18FGT2XhF/lbVW6tjpfZFw= MIME-Version: 1.0 Received: by 10.91.7.15 with SMTP id k15mr10199151agi.167.1289939729953; Tue, 16 Nov 2010 12:35:29 -0800 (PST) Received: by 10.91.154.3 with HTTP; Tue, 16 Nov 2010 12:35:29 -0800 (PST) In-Reply-To: <1289927042.16005.176.camel@thinkpad> References: <20101115182737.42b8dcae@loki.yggdrasil.draxit.de> <4CE228CA.3030503@gmail.com> <1289927042.16005.176.camel@thinkpad> Date: Tue, 16 Nov 2010 22:35:29 +0200 Message-ID: Subject: Re: [Caml-list] SMP multithreading From: Eray Ozkural To: Gerd Stolpmann Cc: Edgar Friendly , caml-list@yquem.inria.fr Content-Type: multipart/alternative; boundary=00163628415662b3180495317f52 X-Spam: no; 0.00; eray:01 ozkural:01 gerd:01 stolpmann:01 compiler:01 runtime:01 runtime:01 compiler:01 eray:01 ozkural:01 bilkent:01 gerd:01 stolpmann:01 bilkent:01 edgar:98 --00163628415662b3180495317f52 Content-Type: text/plain; charset=ISO-8859-1 On Tue, Nov 16, 2010 at 7:04 PM, Gerd Stolpmann wrote: > Am Montag, den 15.11.2010, 22:46 -0800 schrieb Edgar Friendly: > * As somebody mentioned "implicit parallelization": Don't expect > anything from this. Even if a good compiler finds ways to > parallelize 20% of the code (which would be a lot), the runtime > effect would be marginal. 80% of the code is run at normal speed > (hopefully) and dominates the runtime behavior. The point is > that such compiler-driven code improvements are only local > optimizations. For getting good parallelization results you need > to restructure the design of the program - well, maybe > compiler2.0 can do this at some time, but this is not in sight. > I think you are underestimating parallelizing compilers. -- Eray Ozkural, PhD candidate. Comp. Sci. Dept., Bilkent University, Ankara http://groups.yahoo.com/group/ai-philosophy http://myspace.com/arizanesil http://myspace.com/malfunct --00163628415662b3180495317f52 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

On Tue, Nov 16, 2010 at 7:04 PM, Gerd St= olpmann <inf= o@gerd-stolpmann.de> wrote:
Am Montag, den 15.11.2010, 22:46 -0800 schrieb Edgar Friendly:
=A0 =A0 =A0* As somebody mentioned "implicit parallelization": D= on't expect
=A0 =A0 =A0 =A0anything from this. Even if a good compiler finds ways to =A0 =A0 =A0 =A0parallelize 20% of the code (which would be a lot), the run= time
=A0 =A0 =A0 =A0effect would be marginal. 80% of the code is run at normal = speed
=A0 =A0 =A0 =A0(hopefully) and dominates the runtime behavior. The point i= s
=A0 =A0 =A0 =A0that such compiler-driven code improvements are only local<= br> =A0 =A0 =A0 =A0optimizations. For getting good parallelization results you= need
=A0 =A0 =A0 =A0to restructure the design of the program - well, maybe
=A0 =A0 =A0 =A0compiler2.0 can do this at some time, but this is not in si= ght.

I think you are underestimating parallelizin= g compilers.


--
Eray Ozkural, PhD= candidate.=A0 Comp. Sci. Dept., Bilkent University, Ankara
http://groups.yahoo= .com/group/ai-philosophy
h= ttp://myspace.com/arizanesil ht= tp://myspace.com/malfunct

--00163628415662b3180495317f52--