From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 24 Aug 2001 10:33:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 24 Aug 2001 10:33:45 -0400 Received: from [208.48.139.185] ([208.48.139.185]:51148 "HELO forty.greenhydrant.com") by vger.kernel.org with SMTP id ; Fri, 24 Aug 2001 10:33:29 -0400 Date: Fri, 24 Aug 2001 07:33:38 -0700 From: David Rees To: linux-kernel@vger.kernel.org Subject: Re: 3ware RAID1 sequential read speed slower than write speed (2.4.8-ac10) Message-ID: <20010824073338.A16254@greenhydrant.com> Mail-Followup-To: David Rees , linux-kernel@vger.kernel.org In-Reply-To: <20010824090655.29285.qmail@yusufg.portal2.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010824090655.29285.qmail@yusufg.portal2.com>; from yusufg@outblaze.com on Fri, Aug 24, 2001 at 09:06:55AM -0000 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Aug 24, 2001 at 09:06:55AM -0000, Yusuf Goolamabbas wrote: > P3-450 with 256MB RAM with a 3ware 6200 attached to 2 20GB Western > Digitial 7200RPM Caviar Drive WD200BB > > Running bonnie++ on both an ext3 > and ext2 partition, I get the following results > > Version 1.01d ------Sequential Output------ --Sequential Input- --Random- > -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- > Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP > ext3 512M 6020 98 41022 57 6673 8 4619 77 21834 14 237.1 2 > > > Version 1.01d ------Sequential Output------ --Sequential Input- --Random- > -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- > Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP > ext2 512M 6385 99 39969 32 9371 10 5540 91 27864 14 320.7 2 > > > Whilst, I think the write performance is quite good. The read > performance seems to be quite bad. I expected read performance to be > better than write performance for RAID-1 configuration > > Any ideas,patches to try ? Yes, for a mirrored setup you would expect to see almost double the random seeks and read rates equaling write rates. I think the block size that bonnie uses is small, so it doesn't give the RAID a chance to alternate disks for block reads. At least this is what seemed to happen when I was testing software RAID-1 using bonnie (not bonnie++). What's the speed of one disk? Can you try 2.4.8-ac9, -ac10 includes a 3ware driver update. I'll post some numbers from the software raid testing I was doing when I get to work for comparison. If anyone wants to send me a 3ware card (hint, hint ;-), I'll gladly run some tests on that as well to compare. Heck, I'll even write a full review comparing 3ware to software raid! -Dave