From owner-freebsd-current@FreeBSD.ORG Fri Mar 31 08:43:52 2006 Return-Path: X-Original-To: freebsd-current@freebsd.org Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1457916A41F for ; Fri, 31 Mar 2006 08:43:52 +0000 (UTC) (envelope-from ganbold@micom.mng.net) Received: from publicd.ub.mng.net (publicd.ub.mng.net [202.179.0.88]) by mx1.FreeBSD.org (Postfix) with ESMTP id 160AC43D49 for ; Fri, 31 Mar 2006 08:43:51 +0000 (GMT) (envelope-from ganbold@micom.mng.net) Received: from [202.179.0.164] (helo=[192.168.0.18]) by publicd.ub.mng.net with esmtpa (Exim 4.60 (FreeBSD)) (envelope-from ) id 1FPFJg-000Da1-Kr; Fri, 31 Mar 2006 17:49:12 +0900 Message-ID: <442CEBDD.4060902@micom.mng.net> Date: Fri, 31 Mar 2006 17:44:13 +0900 From: Ganbold User-Agent: Thunderbird 1.5 (X11/20060202) MIME-Version: 1.0 To: Kris Kennaway References: <07018E951F2FBF42B40B3D61086788344C1D9D@ednex511.dsto.defence.gov.au> <442B81B9.30206@micom.mng.net> <20060330150610.GA13098@xor.obsecurity.org> In-Reply-To: <20060330150610.GA13098@xor.obsecurity.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-current@freebsd.org Subject: Re: mysql performance test results under FreeBSD-7.0-CURRENT X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 31 Mar 2006 08:43:52 -0000 Kris Kennaway wrote: > On Thu, Mar 30, 2006 at 03:59:05PM +0900, Ganbold wrote: > >> Thyer, Matthew wrote: >> >>> A couple of things I have noticed in the kernel messages: >>> >>> WARNING: WITNESS option enabled, expect reduced performance. >>> >>> >> Yes, I left this option in case if there comes something wrong. >> > > Then your performance numbers are useless, sorry. > I removed WITNESS and INVARIANTS options from kernel and tested again. http://www.mnbsd.org/ftp/mysql_test_results.txt I hope it will be useful to someone now. Ganbold > >>> You need to use the 80 core conductor cables instead of the old standard >>> 40 core conductor IDE cables to get the best performance from your hard >>> drives. Just use another cable like you already have connected to the >>> 40GB drive. >>> >>> >> That is true, I don't have any UDMA100 cables right now:( >> > > Ditto. > > Sorry, but benchmarking is hard. > > Kris