From owner-freebsd-questions@FreeBSD.ORG Tue Jul 29 10:26:53 2003 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 C797037B401 for ; Tue, 29 Jul 2003 10:26:53 -0700 (PDT) Received: from mailhub.yumyumyum.org (dsl092-171-091.wdc1.dsl.speakeasy.net [66.92.171.91]) by mx1.FreeBSD.org (Postfix) with ESMTP id 303E343F75 for ; Tue, 29 Jul 2003 10:26:52 -0700 (PDT) (envelope-from culverk@yumyumyum.org) Received: by mailhub.yumyumyum.org (Postfix, from userid 1001) id 9ACC13B2; Tue, 29 Jul 2003 13:25:32 -0400 (EDT) Received: from localhost (localhost [127.0.0.1]) by mailhub.yumyumyum.org (Postfix) with ESMTP id 96B592A1; Tue, 29 Jul 2003 13:25:32 -0400 (EDT) Date: Tue, 29 Jul 2003 13:25:32 -0400 (EDT) From: Kenneth Culver To: Darryl Hoar In-Reply-To: <006901c355eb$0b914090$0701a8c0@darryl> Message-ID: <20030729132203.V53251@alpha.yumyumyum.org> References: <006901c355eb$0b914090$0701a8c0@darryl> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-questions@freebsd.org Subject: Re: Calling all raid experts 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: Tue, 29 Jul 2003 17:26:54 -0000 > Greetings, > I need to build a file server for our marketing departments documents > and images. I want to use Freebsd. Since the data is large, and > backups would be difficult I was wondering if RAID would be a solution. > > I thought that RAID 5 would be the ticket, but after reading up on it, > maybe not. > > Isn't RAID 5 the one where if a disk fails, you plug a new one it and it > regenerates the lost data ? > It does that through the use of some checksum data that's spread across the disks. Reads on raid 5 can be significantly faster than single disk because data is striped across multiple disks, but writes don't get as much of a boost because for every write, there are several operations, including the recalculation of a checksum, and the writing of both the checksum and data, then a re-reading of them. Writes are usually faster than single disk writes, but the gain is not nearly as much as with reads. Ken