Received: from localhost (mx1.hub.org [200.46.208.251]) by postgresql.org (Postfix) with ESMTP id 52F229FB1C2 for ; Wed, 25 Oct 2006 18:12:19 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.208.251]) (amavisd-new, port 10024) with ESMTP id 85826-09 for ; Wed, 25 Oct 2006 18:12:15 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from mail.metronet.co.uk (mail.metronet.co.uk [213.162.97.75]) by postgresql.org (Postfix) with ESMTP id 5677B9FA621 for ; Wed, 25 Oct 2006 18:12:15 -0300 (ADT) Received: from mainbox.archonet.com (84-51-143-99.archon037.adsl.metronet.co.uk [84.51.143.99]) by smtp.metronet.co.uk (MetroNet Mail) with ESMTP id 3679E749D5; Wed, 25 Oct 2006 22:11:58 +0100 (BST) Received: from [192.168.1.17] (client17.office.archonet.com [192.168.1.17]) by mainbox.archonet.com (Postfix) with ESMTP id 68D7015EDA; Wed, 25 Oct 2006 22:11:56 +0100 (BST) Message-ID: <453FD351.7000408@archonet.com> Date: Wed, 25 Oct 2006 22:12:49 +0100 From: Richard Huxton User-Agent: Thunderbird 1.5.0.7 (X11/20060909) MIME-Version: 1.0 To: Mark Lonsdale Cc: "Joshua D. Drake" , pgsql-performance@postgresql.org Subject: Re: Configuration Issue ? References: <673ECBB77DA91F47AD4902A5FD73BCE4BA3A94@mail3> In-Reply-To: <673ECBB77DA91F47AD4902A5FD73BCE4BA3A94@mail3> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.457 tagged_above=0 required=5 tests=AWL, FORGED_RCVD_HELO X-Spam-Level: X-Archive-Number: 200610/536 X-Sequence-Number: 21665 Mark Lonsdale wrote: > > Hi Josh > > Thanks for the feedback, that is most usefull. When you said one of the > settings was likely killing us, was it all of the settings for > max_fsm_relations, max_fsm_pages, and sort_mem or just the setting for > sort_mem ? > > Can you explain why the setting would be killing me :-) The sort_mem is crucial. It's memory *per sort*, which means one query can use several times that amount. > The long and short is you need to upgrade to at least 7.4, preferrably > 8.1. Joshua means this too. Upgrade to 7.3.16 within the next few days, then test out something more recent. You should see some useful performance gains from 8.1. -- Richard Huxton Archonet Ltd