From owner-freebsd-questions@FreeBSD.ORG Fri Mar 5 10:17:51 2004 Return-Path: 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 06DA316A4CE for ; Fri, 5 Mar 2004 10:17:51 -0800 (PST) Received: from smtpout.mac.com (A17-250-248-46.apple.com [17.250.248.46]) by mx1.FreeBSD.org (Postfix) with ESMTP id EB5E143D45 for ; Fri, 5 Mar 2004 10:17:50 -0800 (PST) (envelope-from cswiger@mac.com) Received: from mac.com (smtpin07-en2 [10.13.10.152]) by smtpout.mac.com (8.12.6/MantshX 2.0) with ESMTP id i25IHmRu011320; Fri, 5 Mar 2004 10:17:48 -0800 (PST) Received: from [10.1.1.193] ([199.103.21.225]) (authenticated bits=0) by mac.com (Xserve/smtpin07/MantshX 3.0) with ESMTP id i25IHdic005537; Fri, 5 Mar 2004 10:17:46 -0800 (PST) In-Reply-To: References: <20040303222612.W39053@guldivar.globalwire.se> <20040303213641.GA37555@happy-idiot-talk.infracaninophile.co.uk> <200403040050.53556.danny@ricin.com> <40474997.60403@mac.com> Mime-Version: 1.0 (Apple Message framework v612) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <6F5934EE-6ED1-11D8-BB2A-003065ABFD92@mac.com> Content-Transfer-Encoding: 7bit From: Charles Swiger Date: Fri, 5 Mar 2004 13:18:00 -0500 To: Jan Grant X-Mailer: Apple Mail (2.612) cc: freebsd-questions@freebsd.org Subject: Re: 1 processor vs. 2 X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 05 Mar 2004 18:17:51 -0000 On Mar 5, 2004, at 5:57 AM, Jan Grant wrote: > How did you come to this conclusion? For a RAID 5 with a single parity > drive, the reason you zero the disks out completely on initialisation > is > to set up the integrity of the parity check. Then any update to any > RAID5 with single parity requires a read of two drives (the target > sector and the corresponding parity drive), an in-memory exclusive or > against the new data, and two writes. Reads and writes can be in > parallel. You're right, which means I came to my conclusion wrongly, I guess. :-) Part of this was because I was also thinking about how the array behaves after a failure, as you mention next: > The "work" for parity updates only scales linearly with number of disks > if you use a naive parity algorithm. Or, obviously, if a drive fails. Even using a non-naive :-) algorithm, RAID-5 writes still take somewhat more work than RAID-1 writes do in terms of I/O ops. -- -Chuck