Received: from localhost (unknown [200.46.204.182]) by postgresql.org (Postfix) with ESMTP id CF8EA9F9FCA for ; Mon, 10 Sep 2007 19:03:41 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.182]) (amavisd-maia, port 10024) with ESMTP id 13106-01 for ; Mon, 10 Sep 2007 19:03:34 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from serv01.siteground137.com (ns1.siteground137.com [67.15.250.10]) by postgresql.org (Postfix) with ESMTP id 51F759F9F0B for ; Mon, 10 Sep 2007 19:03:38 -0300 (ADT) Received: from cragnux.laika.com ([209.162.219.253]:63379 helo=dogma.ljc.laika.com) by serv01.siteground137.com with esmtpa (Exim 4.66) (envelope-from ) id 1IUrLw-0002Fc-R1; Mon, 10 Sep 2007 17:03:33 -0500 Subject: Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1 From: Jeff Davis To: Gregory Stark Cc: Luke Lonergan , Carlo Stonebanks , pgsql-performance@postgresql.org In-Reply-To: <87y7fekxga.fsf@oxford.xeocode.com> References: <87y7fekxga.fsf@oxford.xeocode.com> Content-Type: text/plain Date: Mon, 10 Sep 2007 15:03:31 -0700 Message-Id: <1189461811.28581.17.camel@dogma.ljc.laika.com> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 (2.0.4-7) Content-Transfer-Encoding: 7bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - serv01.siteground137.com X-AntiAbuse: Original Domain - postgresql.org X-AntiAbuse: Originator/Caller UID/GID - [0 0] / [47 12] X-AntiAbuse: Sender Address Domain - j-davis.com X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200709/164 X-Sequence-Number: 26709 On Mon, 2007-09-10 at 22:44 +0100, Gregory Stark wrote: > What I don't understand is the bit about "until Postgres gets AIO + the > ability to post multiple concurrent IOs on index probes". Even with AIO your > seek times are not going to be improved by wide raid stripes. And you can't > possibly find the page at level n+1 before you've looked at the page at level > n. Do you mean to be able to probe multiple index keys simultaneously? How > does that work out? > I think he's referring to mirrors, in which there are multiple spindles that can return a requested block. That could mitigate random I/O, if the I/O is asynchronous and something intelligent (OS or controller) can schedule it. Regards, Jeff Davis