X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 69250D8433 for ; Thu, 1 Sep 2005 20:23:49 -0300 (ADT) 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 55773-04 for ; Thu, 1 Sep 2005 23:23:47 +0000 (GMT) Received: from smtp3.adl2.internode.on.net (smtp3.adl2.internode.on.net [203.16.214.203]) by svr1.postgresql.org (Postfix) with ESMTP id AA197D83AD for ; Thu, 1 Sep 2005 20:23:45 -0300 (ADT) Received: from pluto.mirrabooka.com (ppp188-133.static.internode.on.net [150.101.188.133]) by smtp3.adl2.internode.on.net (8.12.9/8.12.6) with ESMTP id j81NNiIE055604 for ; Fri, 2 Sep 2005 08:53:48 +0930 (CST) (envelope-from alison@mirrabooka.com) Received: by pluto.mirrabooka.com (Postfix, from userid 506) id 6CFDE2B123; Fri, 2 Sep 2005 09:23:44 +1000 (EST) Date: Fri, 02 Sep 2005 09:23:44 +1000 To: pgsql-performance@postgresql.org Subject: Re: lots of updates on small table Message-ID: <43178D80.nail2T711S8W7@pluto.mirrabooka.com> References: <42D5F34E.nail9QE1195IE@pluto.mirrabooka.com> <20050714162114.GA19778@alvh.no-ip.org> <42D6F854.nail3K81VIICH@pluto.mirrabooka.com> <5893.1121385447@sss.pgh.pa.us> <42D71EC1.nail8BK11QC2Q@pluto.mirrabooka.com> In-Reply-To: <42D71EC1.nail8BK11QC2Q@pluto.mirrabooka.com> User-Agent: nail 11.6 9/7/04 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit From: alison@mirrabooka.com (Alison Winters) X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.05 required=5 tests=[AWL=-0.000, FORGED_RCVD_HELO=0.05] X-Spam-Level: X-Archive-Number: 200509/38 X-Sequence-Number: 14346 Just to dig up an old thread from last month: In case anyone was wondering we finally got a free day to put in the new version of the software, and it's greatly improved the performance. The solutions we employed were as follows: - recompile everything with ecpg -t for auto-commit - vacuum run by cron every 15 minutes - vacuum full analyze AND a reindex of the table in question run by cron once per day I also went through and double-checked everywhere where we opened a cursor to make sure it was always closed when we were done. This seems to have fixed up the endlessly growing indexes and slower and slower updates. Thanks to everyone who helped out. Alison