X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id CD6DC3A4C16 for ; Fri, 19 Nov 2004 22:46:50 +0000 (GMT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 32929-10 for ; Fri, 19 Nov 2004 22:46:42 +0000 (GMT) Received: from frank.wiles.org (frank.wiles.org [24.124.39.75]) by svr1.postgresql.org (Postfix) with ESMTP id 687143A4F36 for ; Fri, 19 Nov 2004 21:13:20 +0000 (GMT) Received: from kungfu (frank.wiles.org [24.124.39.75]) by frank.wiles.org (8.12.11/8.12.11) with SMTP id iAJLFp2p008077; Fri, 19 Nov 2004 15:15:52 -0600 Date: Fri, 19 Nov 2004 15:13:37 -0600 From: Frank Wiles To: pg@fastcrypt.com Cc: ajanian@scottrade.com, pgsql-performance@postgresql.org Subject: Re: Query Performance and IOWait Message-Id: <20041119151337.477645de.frank@wiles.org> In-Reply-To: <419CD858.1000201@fastcrypt.com> References: <419CD858.1000201@fastcrypt.com> X-Mailer: Sylpheed version 0.9.10 (GTK+ 1.2.10; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.0 tagged_above=0.0 required=5.0 tests= X-Spam-Level: X-Archive-Number: 200411/306 X-Sequence-Number: 9260 On Thu, 18 Nov 2004 12:14:00 -0500 Dave Cramer wrote: > Andrew, > > Dell's aren't well known for their disk performance, apparently most > of the perc controllers sold with dell's are actually adaptec > controllers. Also apparently they do not come with the battery > required to use the battery backed up write cache ( In fact according > to some Dell won't even sell the battery to you). Also Dell's > monitoring software is quite a memory hog. > > Have you looked at top ?, and also hdparm -Tt /dev/sd? I haven't seen any PERC controllers that were really Adaptec ones, but I for one quit buying Dell RAID controllers several years ago because of poor Linux support and performance. On one machine (not a PostgreSQL server) we saw a 20% speed improvement by switching to software raid. If you have a test machine, I would suggest moving the data to a box without a RAID controller and see if you get better results. --------------------------------- Frank Wiles http://www.wiles.org ---------------------------------