From owner-freebsd-questions@FreeBSD.ORG Sun Sep 24 21:33:43 2006 Return-Path: X-Original-To: freebsd-questions@freebsd.org Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DE58916A415 for ; Sun, 24 Sep 2006 21:33:43 +0000 (UTC) (envelope-from ohartman@uni-mainz.de) Received: from mailgate02.zdv.uni-mainz.de (mailgate02.zdv.Uni-Mainz.DE [134.93.178.132]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3B87B43D53 for ; Sun, 24 Sep 2006 21:33:42 +0000 (GMT) (envelope-from ohartman@uni-mainz.de) Received: from exfront01.zdv.uni-mainz.de ([134.93.176.49]) by mailgate02.zdv.uni-mainz.de with ESMTP; 24 Sep 2006 23:33:41 +0200 Received: from mail.uni-mainz.de ([134.93.176.49]) by exfront01.zdv.Uni-Mainz.DE with Microsoft SMTPSVC(6.0.3790.1830); Sun, 24 Sep 2006 23:33:18 +0200 Received: from [192.168.1.128] ([85.178.27.112] RDNS failed) by mail.uni-mainz.de over TLS secured channel with Microsoft SMTPSVC(6.0.3790.1830); Sun, 24 Sep 2006 23:33:18 +0200 Message-ID: <4516F999.2080309@mail.uni-mainz.de> Date: Sun, 24 Sep 2006 23:33:13 +0200 From: "O. Hartmann" User-Agent: Thunderbird 1.5.0.7 (X11/20060917) MIME-Version: 1.0 To: ajmm@portugalmail.pt References: <1159126289.4516dd118b4f6@webmail3.portugalmail.pt> In-Reply-To: <1159126289.4516dd118b4f6@webmail3.portugalmail.pt> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable X-OriginalArrivalTime: 24 Sep 2006 21:33:18.0491 (UTC) FILETIME=[0D18CEB0:01C6E021] Cc: ohartman@uni-mainz.de, freebsd-questions@freebsd.org Subject: Re: Hitachi disk performance X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Sep 2006 21:33:43 -0000 ajmm@portugalmail.pt wrote: > =20 > =20 > Hi! =20 > =20 > I saw your posts with title "Performance issue amd64". I don't know if = this =20 > could help, but I'm having the same problem with ICH5 SATA controller a= nd the =20 > same Hitachi drives on linux. I cannot activate DMA, and bandwidth is l= esser=20 > than 10MB/s. I could not solve the problem, until now. BTW, with "any" = other=20 > drives, the problem does not exist. =20 > =20 > Antonio Marques =20 > __________________________________________________________ > Continua a preferir gastar mais? > Compare o pre=E7o da sua liga=E7=E3o =E0 Internet > http://acesso.portugalmail.pt/compare > =20 I did not dig deeper into that problem - due to the lack of suitable=20 hardware. Bonnie shows about 55 MB/s sustained performance, sometimes=20 higher, burst reaches 117 MB/s. With the old disks, both values were=20 about 25 MB/s higher. I doubt this is only a partition layout related=20 issue. The average performance droped remarkable, especially while=20 compiling world. At the Hitachi website I did not find any hint about issues with several = controllers or especially with SATA PseudoRAID like nForce4 offers, hope = this can be fixed anyway and it's a simple firmware issue. Checking the disks via the Hitachi brewn drive fitness tool gives no=20 suspicious data, mode is SATA II/300, no accoustic or energy management=20 and so on ... I would appreciate any solvage, if you have one! Well, in your case, I would check whether your drive is in SATA 150 mode = (factory setting) or SATA 300, as I know the ICH5 is only SATA 150 and=20 drives with problems switching to SATA 150 while in SATA 300 could cause = problems. Regards, Oliver