Received: from localhost (maia-5.hub.org [200.46.204.182]) by postgresql.org (Postfix) with ESMTP id 54E7E9FBBF4 for ; Tue, 26 Jun 2007 15:14:28 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.182]) (amavisd-maia, port 10024) with ESMTP id 41929-07 for ; Tue, 26 Jun 2007 15:14:16 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from westnet.com (westnet.com [216.187.52.2]) by postgresql.org (Postfix) with ESMTP id 187D49FBBD4 for ; Tue, 26 Jun 2007 15:14:23 -0300 (ADT) Received: from westnet.com (localhost [127.0.0.1]) by westnet.com (8.14.0/8.14.0) with ESMTP id l5QIEOw2018392 for ; Tue, 26 Jun 2007 14:14:24 -0400 (EDT) Received: from localhost (gsmith@localhost) by westnet.com (8.14.0/8.13.2/Submit) with ESMTP id l5QIEOr1018389 for ; Tue, 26 Jun 2007 14:14:24 -0400 (EDT) X-Authentication-Warning: westnet.com: gsmith owned process doing -bs Date: Tue, 26 Jun 2007 14:14:24 -0400 (EDT) From: Greg Smith X-X-Sender: gsmith@westnet.com To: pgsql-performance@postgresql.org Subject: Re: PostgreSQL Configuration Tool for Dummies In-Reply-To: <200706260826.28935.josh@agliodbs.com> Message-ID: References: <200706191254.09777.josh@agliodbs.com> <200706260826.28935.josh@agliodbs.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200706/601 X-Sequence-Number: 25634 On Tue, 26 Jun 2007, Josh Berkus wrote: > The problem is that there are no "safe, middle-of-the-road" values for some > things, particularly max_connections and work_mem. Your max_connections concern is one fact that haunts the idea of just giving out some sample configs for people. Lance's tool asks outright the expectation for max_connections which I think is the right thing to do. > Web app, typical machine: > work_mem = 256kb > default_statistics_target=100 > autovacuum=on > Reporting app, same machine: > work_mem = 32mb > default_statistics_target=500 > autovacuum=off I think people are stuck with actually learning a bit about work_mem whether they like it or not, because it's important to make it larger but we know going too high will be a problem with lots of connections doing sorts. As for turning autovacuum on/off and the stats target, I'd expect useful defaults for those would come out of how the current sample is asking about read vs. write workloads and expected database size. Those simple to understand questions might capture enough of the difference between your two types here. -- * Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD