From owner-freebsd-performance@FreeBSD.ORG Fri Dec 16 20:18:54 2005 Return-Path: X-Original-To: freebsd-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 ADB8416A41F for ; Fri, 16 Dec 2005 20:18:54 +0000 (GMT) (envelope-from gbaratto@superb.net) Received: from smail2.superb.net (smail2.superb.net [64.251.80.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4C27D43D4C for ; Fri, 16 Dec 2005 20:18:54 +0000 (GMT) (envelope-from gbaratto@superb.net) Received: from [64.251.80.100] (100.80.251.64.yvr1.superb.net [64.251.80.100]) by smail2.superb.net (Postfix) with ESMTP id 898E91F2CFF; Fri, 16 Dec 2005 12:18:53 -0800 (PST) Message-ID: <43A3219E.7080808@superb.net> Date: Fri, 16 Dec 2005 12:20:46 -0800 From: "Gustavo A. Baratto" User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Gea-Suan Lin References: <20051216064615.GA41775@gslin.org> In-Reply-To: <20051216064615.GA41775@gslin.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-performance@freebsd.org, leo.huang.gd@gmail.com Subject: Re: Benchmark MySQL Performance On FreeBSD And 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: Fri, 16 Dec 2005 20:18:54 -0000 So, is ULE ready for production on 6.0-RELEASE? Can we use it without fear? Cheers Gea-Suan Lin wrote: > Hi, > > In http://blog.gslin.org/archives/2005/12/12/252/ we test more cases, > and summary some important conclusions: > > * SCHED_ULE (kernel options) is faster than SCHED_4BSD. > > * Use kern.timecounter.choice=TSC (sysctl) will be faster than ACPI-fast > or ACPI-safe. (about 10% again) > > And I notice you use kernel-image-2.6.8-2-686-smp in Debian, but not use > "CPUTYPE?= p3" in /etc/make.conf to compile kernel. > > And you should disable these options, it may increase ~10% again: > > -cpu I486_CPU > -cpu I586_CPU > cpu I686_CPU >