From oracle-l-bounce@freelists.org Wed Jul 7 17:46:46 2004 Return-Path: Received: from air189.startdedicated.com (root@localhost) by orafaq.com (8.11.6/8.11.6) with ESMTP id i67MkJl06785 for ; Wed, 7 Jul 2004 17:46:29 -0500 X-ClientAddr: 206.53.239.180 Received: from turing.freelists.org (freelists-180.iquest.net [206.53.239.180]) by air189.startdedicated.com (8.11.6/8.11.6) with ESMTP id i67Mk9606745 for ; Wed, 7 Jul 2004 17:46:19 -0500 Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id DB5EE72C06D; Wed, 7 Jul 2004 17:27:30 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 22842-93; Wed, 7 Jul 2004 17:27:30 -0500 (EST) Received: from turing (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 2F13372C029; Wed, 7 Jul 2004 17:27:30 -0500 (EST) Received: with ECARTIS (v1.0.0; list oracle-l); Wed, 07 Jul 2004 17:26:04 -0500 (EST) X-Original-To: oracle-l@freelists.org Delivered-To: oracle-l@freelists.org Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 739A172C5CA for ; Wed, 7 Jul 2004 17:26:04 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 23753-07 for ; Wed, 7 Jul 2004 17:26:04 -0500 (EST) Received: from mail.cih.com (moonshine.cih.com [204.69.206.3]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 1DEC472C42F for ; Wed, 7 Jul 2004 17:26:04 -0500 (EST) Received: from hagan.desktop.amazon.com (207-171-180-101.amazon.com [207.171.180.101]) (using SSLv3 with cipher RC4-MD5 (128/128 bits)) (No client certificate requested) by cih.com (Server (Microsoft Exchange Internet Mail Service 5.5.2232.9) ready) with ESMTP id 422EF6D for ; Wed, 7 Jul 2004 15:49:54 -0700 (PDT) From: "Craig I. Hagan" To: oracle-l@freelists.org Subject: Re: Raid 50 Date: Wed, 7 Jul 2004 15:46:49 -0700 User-Agent: KMail/1.6.1 References: <33678E78A2DD4D418396703A750048D401024EA0@RIKER> <2572.64.37.153.21.1089135570.squirrel@www.cosmiccooler.org> In-Reply-To: <2572.64.37.153.21.1089135570.squirrel@www.cosmiccooler.org> MIME-Version: 1.0 Content-Disposition: inline X-UID: 227 Content-type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-Id: <200407071546.49005.hagan@cih.com> X-Virus-Scanned: by amavisd-new at freelists.org X-archive-position: 4611 X-ecartis-version: Ecartis v1.0.0 Sender: oracle-l-bounce@freelists.org Errors-To: oracle-l-bounce@freelists.org X-original-sender: hagan@cih.com Precedence: normal Reply-To: oracle-l@freelists.org X-list: oracle-l X-Virus-Scanned: by amavisd-new at freelists.org On Tuesday 06 July 2004 10:39, David wrote: > I am well aware of the pros and cons of RAID 0, 1, 0+1, 1+0(raid 10), 5, > etc. But, here is a new one on me...RAID 50. > > Thoughts, experience...please elaborate asI just found out it's our RAC IO > backend. Raid50 is a compromise between raid10 and raid5. Basically you create small raid5 sets (usually 4 or 8 disks) and then stripe across the small raid5 sets rather than across mirror pairs. 10,000 foot comparison with raid10: Pro: you get more storage, and somewhat better streaming read performance con: somewhat worse small/random write performance for non full stripe or non stripe aligned operations (you go from a crc+write to a read, modify, crc, write) con: overall write performance can be constrained by the speed of your checksummer con: you lose some level of redundancy. Instead of mirror pairs of two disks, you have raid5 sets of N (usually 4 or 8) disks. A double fault is more likely with a raid5 set. It is still a possibility with raid10, too. con: degraded mode will cause a stripe will require additional compute (usually via the checksummer) to reconstruct the missing data which can have an unpredictable impact upon io performance as well as potentially impairing io to the rest of the array. a raid1 mirror fault will cause degraded performance only to the impacted set, and that usually is represented by a 50% loss of effective read iops if the array load balances (it should) but full write io capacity is retained. (i'm excluding the io cost of folding in a new disk/hot spare for both cases) ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------