From owner-freebsd-stable@FreeBSD.ORG Thu Nov 9 23:09:46 2006 Return-Path: X-Original-To: freebsd-stable@FreeBSD.ORG Delivered-To: freebsd-stable@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F19DB16A40F for ; Thu, 9 Nov 2006 23:09:46 +0000 (UTC) (envelope-from markir@paradise.net.nz) Received: from linda-2.paradise.net.nz (linda-2.paradise.net.nz [203.96.152.181]) by mx1.FreeBSD.org (Postfix) with ESMTP id D46D043D4C for ; Thu, 9 Nov 2006 23:09:44 +0000 (GMT) (envelope-from markir@paradise.net.nz) Received: from smtp-2.paradise.net.nz (tclsnelb1-src-1.paradise.net.nz [203.96.152.172]) by linda-2.paradise.net.nz (Paradise.net.nz) with ESMTP id <0J8H0091KKC7T1@linda-2.paradise.net.nz> for freebsd-stable@FreeBSD.ORG; Fri, 10 Nov 2006 12:09:43 +1300 (NZDT) Received: from [192.168.1.11] (218-101-28-191.dsl.clear.net.nz [218.101.28.191]) by smtp-2.paradise.net.nz (Postfix) with ESMTP id D930325A537; Fri, 10 Nov 2006 12:09:42 +1300 (NZDT) Date: Fri, 10 Nov 2006 12:09:37 +1300 From: Mark Kirkwood In-reply-to: To: Pete French Message-id: <4553B531.1010403@paradise.net.nz> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7bit User-Agent: Thunderbird 1.5.0.7 (X11/20061101) References: Cc: freebsd-stable@FreeBSD.ORG Subject: Re: Dissapointing performance of ciss RAID 0+1 ? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Nov 2006 23:09:47 -0000 Pete French wrote: >> You might be able to speed up the read by playing with the vfs.read_max >> sysctl (try 16 or 32). > > Wow! That makes a huge difference, thanks. Should this not be in 'man tuning' ? > Yeah, I believe I've seen it mentioned *somewhere* with respect to working with RAID (of course, I can't find it now....). The other thing you might find useful for improving sequential performance is using newfs options '-b 32768 -f 4096' - I've found it produces a noticeable performance improvement (can't recall exactly but in the region of 10%). I'm using an older system (3ware 7506 + 4 7200 rpm disks RAID 0 256 K stripe), and with the above newfs settings + vfs.read_max=16 I get about 190Mb/s for 8k reads and 160Mb/s for 8k writes (sustained for quite big files e.g. 16Gb). This is on 6.2 PRERELEASE, and I'm pretty happy with that level of performance (nice job to those of you who have been steadily making 6-STABLE go faster....)! Cheers Mark