X-Original-To: pgsql-performance@postgresql.org Received: from ucsns.ucs.co.za (ucs.co.za [196.23.43.2]) by svr1.postgresql.org (Postfix) with ESMTP id 725E95E472D for ; Tue, 24 Aug 2004 10:30:19 -0300 (ADT) Received: from localhost (localhost.localdomain [127.0.0.1]) by ucsns.ucs.co.za (Postfix) with ESMTP id F059D2BD58 for ; Tue, 24 Aug 2004 15:30:17 +0200 (SAST) Received: from ucsns.ucs.co.za ([127.0.0.1]) by localhost (ucsns.ucs.co.za [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 03773-05 for ; Tue, 24 Aug 2004 15:30:14 +0200 (SAST) Received: from ucspost.ucs.co.za (mailgw1.ucs.co.za [196.23.43.253]) by ucsns.ucs.co.za (Postfix) with ESMTP id 6B6042BD78 for ; Tue, 24 Aug 2004 15:30:14 +0200 (SAST) Received: from jhb.ucs.co.za (jhb.ucs.co.za [172.31.1.3]) by ucspost.ucs.co.za (Postfix) with ESMTP id 95967D9E29 for ; Tue, 24 Aug 2004 15:30:14 +0200 (SAST) Received: from svb.ucs.co.za (svb.ucs.co.za [172.31.1.148]) by jhb.ucs.co.za (Postfix) with SMTP id BEE60975A0 for ; Tue, 24 Aug 2004 15:30:14 +0200 (SAST) Date: Tue, 24 Aug 2004 15:30:14 +0200 From: Stef To: pgsql-performance@postgresql.org Subject: Re: Query kills machine. Message-Id: <20040824153014.2a11d228@svb.ucs.co.za> In-Reply-To: <412B05A2.8070301@familyhealth.com.au> References: <20040824103240.4d909777@svb.ucs.co.za> <412B05A2.8070301@familyhealth.com.au> X-Mailer: Sylpheed-Claws 0.9.12 (GTK+ 1.2.10; i386-pc-linux-gnu) User-Agent: sillypheed-claws (anti-aliased) X-Face: GFoC95e6r)_TTG>n~=uFLojP=O~4W@Ms]>:.DMm/')(z3\Mwj^XP@? Q:3";lD.OM1"^mDu}2NJ@US:)dO:U*iY5EM50&Tx. X-Operating-System: sid X-X-X: _-^-_ Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at ucs.co.za X-Archive-Number: 200408/323 X-Sequence-Number: 7959 Christopher Kings-Lynne mentioned : => sort_mem = 4096 Reducing sort_mem to 4096 seems to make it run in a reasonable time again. Any idea why? The database does a whole lot of huge sorts every day, so I thought upping this parameter would help. A couple of queries do seem to run slower now that I reduced the sort_mem. The shared buffers still makes a significant difference when I increase it. Kind Regards Stefan