From owner-freebsd-questions Mon May 7 9:17:41 2001 Delivered-To: freebsd-questions@freebsd.org Received: from schmoo.tclme.org (schmoo.tclme.org [208.24.53.96]) by hub.freebsd.org (Postfix) with SMTP id 13F8637B423 for ; Mon, 7 May 2001 09:17:38 -0700 (PDT) (envelope-from rgreene@tclme.org) Received: (qmail 76048 invoked by uid 1014); 7 May 2001 16:16:06 -0000 Received: from dinky.tclme.org (HELO tclme.org) (rgreene@208.24.53.107) by mail.tclme.org with SMTP; 7 May 2001 16:16:06 -0000 Message-ID: <3AF6CCDB.F1029665@tclme.org> Date: Mon, 07 May 2001 11:27:07 -0500 From: Bob Greene Organization: tclme.org X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.2-2 i686) X-Accept-Language: en MIME-Version: 1.0 To: Greg Lehey Cc: "Andrew C. Hornback" , Steve Blanzy , FreeBSD Questions Subject: Re: Raid References: <000a01c0d57f$2158bb40$0400a8c0@192.168.0.1> <000701c0d581$3dd2da60$0e00000a@tomcat> <20010506101618.B39554@wantadilla.lemis.com> <3AF4BF5A.A03D7278@tclme.org> <20010506130347.F39554@wantadilla.lemis.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Greg Lehey wrote: > > On Saturday, 5 May 2001 at 22:04:58 -0500, Bob Greene wrote: > > Greg Lehey wrote: > >> > >> No, RAID-1 gives you the best performance of any RAID setup. The > >> reason why you need at least 3 disks for RAID-5 is because it is > >> slower, and though it would theoretically work with only two disks, > >> it has no advantages over RAID-1 in this configuration. > >> > > > > Huh? This paragraph makes no sense. > > > > RAID 0 = striped set > > RAID 1 = mirrored set > > RAID 5 = striped set with parity > > > > RAID 1 gives maximum redundancy, at the cost of two writes. The third > > disk in RAID 5 is not a consequence of performance, it's a requirement > > for redundancy. RAID 5 with only 2 disks is a failure condition of a 3 > > disk array. At that point it's effectively just a striped set. > > No, because in degraded 3-disk RAID-5 every third access is intended > for the failed drive. In order to read that data, you need to read > both (all) the other drives and reconstruct it. This can be very slow > if you have a large number of drives in the set. > > Greg > -- Well, you wrote the book; I've only read it. Is this a vinum specific performance penalty? -- Bob Greene rgreene@TclMe.org Pull my finger for my public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message