Received: from localhost (unknown [200.46.204.183]) by postgresql.org (Postfix) with ESMTP id 29699650146 for ; Thu, 30 Oct 2008 18:49:05 -0300 (ADT) Received: from postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 24000-10 for ; Thu, 30 Oct 2008 18:49:02 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from skynet.simkin.ca (skynet.simkin.ca [72.51.27.117]) by postgresql.org (Postfix) with ESMTP id C248864FD96 for ; Thu, 30 Oct 2008 18:49:02 -0300 (ADT) Received: from charon.medialogik.com (unknown [10.10.2.229]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by skynet.simkin.ca (Postfix) with ESMTP id D45AB4CA8 for ; Thu, 30 Oct 2008 14:49:00 -0700 (PDT) From: Alan Hodgson Organization: Simkin Network Consulting To: pgsql-performance@postgresql.org Subject: Re: CPU utilization vs. IO wait, shared buffers? Date: Thu, 30 Oct 2008 14:48:58 -0700 User-Agent: KMail/1.9.10 References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200810301448.58941@hal.medialogik.com> X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests=none X-Spam-Level: X-Archive-Number: 200810/133 X-Sequence-Number: 31714 On Thursday 30 October 2008, "Oliver Johnson" wrote: > Another thing to note, we have VACUUM ANALYZE running on an hourly > interval and the switch from CPU to IO wait appears to always coincide > with a vacuum. > > What might cause this shift? The extra disk access caused by vacuum? That seems pretty obvious. Use auto-vacuum. There's no reason to vacuum your entire database every hour (doing so reads from disk the entirety of every table and index, and generates some write activity). -- Alan