From oracle-l-bounce@freelists.org Thu Sep 9 09:25:40 2004 Return-Path: Received: from air189.startdedicated.com (root@localhost) by orafaq.com (8.11.6/8.11.6) with ESMTP id i89EPdW09713 for ; Thu, 9 Sep 2004 09:25:39 -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 i89EPbI09703 for ; Thu, 9 Sep 2004 09:25:37 -0500 Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 623AE72C883; Thu, 9 Sep 2004 09:22:59 -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 07661-50; Thu, 9 Sep 2004 09:22:59 -0500 (EST) Received: from turing (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id C148172C877; Thu, 9 Sep 2004 09:22:58 -0500 (EST) In-Reply-To: <004201c48ae9$4c1e8c60$6701a8c0@CVMLAP02> To: oracle-l@freelists.org Cc: cary.millsap@hotsos.com Subject: RE: Raid5 Vs Raid0+1 -- Raw Vs Solaris 9 Concurrent Direct IO UFS MIME-Version: 1.0 Message-ID: From: J.Velikanovs@alise.lv Date: Thu, 9 Sep 2004 17:24:44 +0300 X-MIMETrack: Serialize by Router on ROSS/IT ALISE/LV(Release 5.0.11 |July 24, 2002) at 2004.09.09 17:24:45, Serialize complete at 2004.09.09 17:24:45 Content-type: text/plain Content-Transfer-Encoding: 8bit X-archive-position: 9312 X-ecartis-version: Ecartis v1.0.0 Sender: oracle-l-bounce@freelists.org Errors-To: oracle-l-bounce@freelists.org X-original-sender: J.Velikanovs@alise.lv Precedence: normal Reply-To: J.Velikanovs@alise.lv X-list: oracle-l X-Virus-Scanned: by amavisd-new at freelists.org Hello Caty & List > Second, because a RAID level 5 array has > inherently greater load to manage than a RAID level 1 >array. If an application generates R 1-block reads/sec, and W 1-block >writes/sec, then the two architectures would compare this way: > >- Level 1: would have to process (R + W) I/O requests per second >- Level 5: would have to process (R + 4W) I/O requests per second Can I kindly ask you to clarify few questions? 1. Is 4W figure (in formula above) constant in context of RAID 5 array and not depend on spindles count? I suspect that it can be constant in any RAID5 implementation. In case of 6 spindles block will be distributed as: Disk1 => Block1 Disk2 => Block2 Disk3 => Block3 Disk4 => Checksum 123 Disk5 => Block4 Disk6 => Block5 Disk1 => Block6 Disk2 => Checksum 456 Disk…. Is my assumption correct? . 2. If we need to change one of 3 data blocks belonging to one RAID5 set, then controller/array/we need to make one write to this particular block + read 2 remaining block for check sum calculation + write check sum to 4 block. Do I understand correctly? So for writing one block into RAID5 we need 2W+2R. Or I am wrong? . Thank in advance, Jurijs +371 9268222 (+2 GMT) ============================================ Thank you for teaching me. http://otn.oracle.com/ocm/jvelikanovs.html On 26.08.2004 00:20:00 oracle-l-bounce wrote: >? > >A single-block read from a RAID level 5 array will visit only one block = >on >the array (unless something weird's going on like a block split across >devices, or there's a partial outage going on). > >A single-block read from a RAID level 1 array will also visit only one = >block >(unless there's a block split issue), but the advantage of RAID level 1 = >is >that a good controller can fetch the block from the less busy of two = >disks >storing equally valid copies of the block. > >There is no "read advantage" of level 5 over level 1. In fact, it's = >quite >the contrary. First, because of what I said above. > Second, because a RAID level 5 array has > inherently greater load to manage than a RAID level 1 >array. If an application generates R 1-block reads/sec, and W 1-block >writes/sec, then the two architectures would compare this way: > >- Level 1: would have to process (R + W) I/O requests per second >- Level 5: would have to process (R + 4W) I/O requests per second > >So, for example, if write calls comprise 50% of your I/O call workload = >(that >is, W=3DR), then this is your situation: > >- Level 1: load is 2R I/O requests per second >- Level 5: load is 5R I/O requests per second > >That is, the RAID level 5 system will have to process 2.5x more I/Os per >second than the RAID level 1 system. How could the RAID-5 system keep = >up? >Either with a /lot/ of cache ($$$, and Tim's right; any amount of cache = >can >be overwhelmed by a high enough sustained I/O rate), or by buying a = >/lot/ >more disks. > >...By the time you buy all that stuff, your whole economic motivation = >for >buying RAID level 5 ("it's cheaper, because you don't have to buy as = >many >disks...") is out the window. > >RAID level 5 is /not/ cheaper, because you /do/ have to buy as many = >disks. >And cache. And controller software. ...BAARF. > > >Cary Millsap >Hotsos Enterprises, Ltd. >http://www.hotsos.com >* Nullius in verba * > >Upcoming events: >- Performance Diagnosis 101: 9/14 San Francisco, 10/5 Charlotte, 10/26 >Toronto >- SQL Optimization 101: 8/16 Minneapolis, 9/20 Hartford, 10/18 New = >Orleans >- Hotsos Symposium 2005: March 6-10 Dallas >- Visit www.hotsos.com for schedule details... -- To unsubscribe - mailto:oracle-l-request@freelists.org&subject=unsubscribe To search the archives - http://www.freelists.org/archives/oracle-l/