From owner-freebsd-performance@FreeBSD.ORG Wed Jun 29 13:58:34 2005 Return-Path: X-Original-To: performance@freebsd.org Delivered-To: freebsd-performance@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D1CB016A41C; Wed, 29 Jun 2005 13:58:34 +0000 (GMT) (envelope-from dsze@alumni.uwaterloo.ca) Received: from mail.distrust.net (mail.distrust.net [69.93.230.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9FB2343D4C; Wed, 29 Jun 2005 13:58:34 +0000 (GMT) (envelope-from dsze@alumni.uwaterloo.ca) Received: from eeyore.distrust.net (CPE00a0c978120d-CM00122570472e.cpe.net.cable.rogers.com [70.28.248.121]) (authenticated bits=0) by mail.distrust.net (8.13.3/8.13.3) with ESMTP id j5TDwUHk083869 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 29 Jun 2005 08:58:33 -0500 (CDT) (envelope-from dsze@alumni.uwaterloo.ca) Message-Id: <6.2.1.2.2.20050629095600.05a54268@mail.distrust.net> X-Mailer: QUALCOMM Windows Eudora Version 6.2.1.2 Date: Wed, 29 Jun 2005 09:58:39 -0400 To: Robert Watson From: David Sze In-Reply-To: <20050629133209.Q74117@fledge.watson.org> References: <6.2.1.2.2.20050620105116.06376688@wheresmymailserver.com> <20050623144611.GA64879@bibipentium.lonres.com> <20050629133209.Q74117@fledge.watson.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Virus-Scanned: ClamAV 0.85.1/960/Tue Jun 28 23:31:06 2005 on mail.distrust.net X-Virus-Status: Clean X-Mailman-Approved-At: Wed, 29 Jun 2005 14:06:39 +0000 Cc: Steve Roome , performance@freebsd.org Subject: Re: FreeBSD MySQL still WAY slower than Linux X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 29 Jun 2005 13:58:34 -0000 At 01:33 PM 29/06/2005 +0100, Robert Watson wrote this to All: >On Thu, 23 Jun 2005, Steve Roome wrote: > >>>The different threading libraries are more for completeness. In my last >>>test I saw <10% difference between them on amd64. >> >>Well, I finally got some tests out for FreeBSD/i386 with -current, Here >>we go with a bunch of results of FreeBSD 6 with mysql and different >>threading libraries, as for you, there's sadly not enough in it so far. >>Maybe I need to tune a whole bunch of stuff I've missed ? > >What version of MySQL are you using? I saw much better FreeBSD MySQL >performance with 4.0.x than 4.1.x in terms of transactions per second, and >am wondering what changed in MySQL -- did it start using different IPC >primitives, threading primitives, etc? I realize it complicates testing, >but if you have a chance to compare 4.0.x and 4.1.x across linux and >freebsd, that would be quite helpful, so we can see if there's a change in >relative performance. Most of my work to optimize MySQL performance with >SMP has been using 4.0.x, FYI, so 4.1.x might require different >optimization strategies. I haven't got around to finishing all the benchmarks yet, but on 5.4-RELEASE-p2 amd64, I see about 5-7% better performance with 4.0.x versus 4.1.x for the select-key benchmark, and no improvement for update-select (both with InnoDB tables, using the "my-innodb-heavy-4G.cnf" file).