Received: from localhost (maia-1.hub.org [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id F35B49FB987 for ; Thu, 7 Jun 2007 10:32:53 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-maia, port 10024) with ESMTP id 00832-08 for ; Thu, 7 Jun 2007 10:32:44 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by postgresql.org (Postfix) with ESMTP id BDBA39FB914 for ; Thu, 7 Jun 2007 10:32:44 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.1/8.14.1) with ESMTP id l57DWVEf029878; Thu, 7 Jun 2007 09:32:31 -0400 (EDT) To: Kurt Overberg cc: pgsql-performance@postgresql.org Subject: Re: Weird 8.2.4 performance In-reply-to: References: <371D1936-E96B-4865-AEEF-C82AC41C2982@hotdogrecords.com> <46674AEF.6040803@paradise.net.nz> <4667CE87.30101@archonet.com> Comments: In-reply-to Kurt Overberg message dated "Thu, 07 Jun 2007 07:18:22 -0400" Date: Thu, 07 Jun 2007 09:32:31 -0400 Message-ID: <29877.1181223151@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200706/127 X-Sequence-Number: 25160 Kurt Overberg writes: > ... Turning off bitmap scans definitely seems > to help things, I really seriously doubt that. On queries like this, where each inner scan is fetching a couple hundred rows, the small extra overhead of a bitmap scan should easily pay for itself. I think you're looking entirely at caching effects that allow a re-read of the same data to go faster. You might try running the same query plan several times in a row and noting the lowest time, then repeat for the other query plan. This will get you comparable fully-cached times, which I bet will be very close to the same. regards, tom lane