X-Original-To: pgsql-performance-postgresql.org@postgresql.org Received: from localhost (mx1.hub.org [200.46.208.251]) by postgresql.org (Postfix) with ESMTP id 397C99FA147 for ; Wed, 30 Aug 2006 07:30:44 -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 13544-07-2 for ; Wed, 30 Aug 2006 07:30:31 -0300 (ADT) X-Greylist: delayed 01:00:03.640238 by SQLgrey- Received: from frankenstein.isux.com (mailbox.isux.com [203.173.10.51]) by postgresql.org (Postfix) with SMTP id D307B9FA5DC for ; Wed, 30 Aug 2006 07:30:30 -0300 (ADT) Received: from [203.173.10.53] by frankenstein.isux.com (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTPA id <0J4T004DQ0T61600@frankenstein.isux.com> for pgsql-performance@postgresql.org; Wed, 30 Aug 2006 19:24:46 +1000 (EST) Date: Wed, 30 Aug 2006 19:29:35 +1000 From: Matthew Sullivan Subject: performance problems. To: pgsql-performance@postgresql.org Message-id: <44F55A7F.40305@sorbs.net> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7BIT User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.4) Gecko/20060516 SeaMonkey/1.0.2 X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=1.069 tagged_above=0 required=5 tests=SPF_NEUTRAL X-Spam-Level: * X-Archive-Number: 200608/466 X-Sequence-Number: 20651 All, Got a little bit of a performance problem I hope that can be resolved. All the files/info I believe you are going to ask for are here: http://www.au.sorbs.net/~matthew/postgres/30.8.06/ The odd thing was it originally was fast (1-2 seconds) which is all I need - the query is a permissions check and I added a permissions caching engine to the client code. However, I was testing part of my new interface and added and "expired" some rows in the permissions, and authorisation tables (taking the row count to ~15) the performance dropped to 86seconds (ish) which is unusable... :-( Unfortunately I do not have a query plan from before the performance issue. work_mem has been adjusted from 512 to 8192, 65536 and 1000000 with no apparent effect. random_page_cost has been 4 and 2 - 2 results in 89seconds for the query. The hardware is a Compaq 6400r with 4G of EDO RAM, 4x500MHz Xeons and a Compaq RAID 3200 in RAID 5 configuration running across 3 spindles (34G total space). The OS is FreeBSD 5.4-RELEASE-p14 The PG Version is 8.1.3 Solutions/tips greatly appreciated. Regards, Mat