Received: from localhost (unknown [200.46.204.183]) by developer.postgresql.org (Postfix) with ESMTP id 316CF2E008B for ; Wed, 28 May 2008 20:56:17 -0300 (ADT) Received: from developer.postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 69953-01-2 for ; Wed, 28 May 2008 20:56:08 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from davinci.ethosmedia.com (unknown [209.128.84.227]) by developer.postgresql.org (Postfix) with ESMTP id 070E22E00C0 for ; Wed, 28 May 2008 20:55:23 -0300 (ADT) X-EthosMedia-Virus-Scanned: no infections found Received: from [64.81.245.111] (account josh@agliodbs.com HELO temoku.sf.agliodbs.com) by davinci.ethosmedia.com (CommuniGate Pro SMTP 4.1.8) with ESMTP-TLS id 16800121 for pgsql-performance@postgresql.org; Wed, 28 May 2008 16:55:17 -0700 From: Josh Berkus Reply-To: josh@agliodbs.com Organization: Aglio Database Solutions To: pgsql-performance@postgresql.org Subject: 2GB or not 2GB Date: Wed, 28 May 2008 16:59:26 -0700 User-Agent: KMail/1.8 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200805281659.26945.josh@agliodbs.com> X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200805/360 X-Sequence-Number: 30328 Folks, Subsequent to my presentation of the new annotated.conf at pgCon last week, there's been some argument about the utility of certain memory settings above 2GB. I'd like to hash those out on this list so that we can make some concrete recomendations to users. shared_buffers: according to witnesses, Greg Smith presented at East that based on PostgreSQL's buffer algorithms, buffers above 2GB would not really receive significant use. However, Jignesh Shah has tested that on workloads with large numbers of connections, allocating up to 10GB improves performance. sort_mem: My tests with 8.2 and DBT3 seemed to show that, due to limitations of our tape sort algorithm, allocating over 2GB for a single sort had no benefit. However, Magnus and others have claimed otherwise. Has this improved in 8.3? So, can we have some test evidence here? And workload descriptions? -- --Josh Josh Berkus PostgreSQL @ Sun San Francisco