X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 267553A4052 for ; Mon, 25 Oct 2004 14:51:39 +0100 (BST) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 91568-07 for ; Mon, 25 Oct 2004 13:51:30 +0000 (GMT) Received: from smtp105.rog.mail.re2.yahoo.com (smtp105.rog.mail.re2.yahoo.com [206.190.36.83]) by svr1.postgresql.org (Postfix) with SMTP id BD5633A402C for ; Mon, 25 Oct 2004 14:51:32 +0100 (BST) Received: from unknown (HELO phlogiston.dydns.org) (a.sullivan@rogers.com@65.49.125.184 with login) by smtp105.rog.mail.re2.yahoo.com with SMTP; 25 Oct 2004 13:51:33 -0000 Received: by phlogiston.dydns.org (Postfix, from userid 1000) id 7CAE44058; Mon, 25 Oct 2004 09:51:31 -0400 (EDT) Date: Mon, 25 Oct 2004 09:51:31 -0400 From: Andrew Sullivan To: PgSQL - Performance Subject: Re: Performance Anomalies in 7.4.5 Message-ID: <20041025135131.GA2886@phlogiston.dyndns.org> Mail-Followup-To: PgSQL - Performance References: <25228.1098395599@sss.pgh.pa.us> <20041022194058.GC22395@phlogiston.dyndns.org> <417977EE.8050207@zeut.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <417977EE.8050207@zeut.net> User-Agent: Mutt/1.5.6+20040722i X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.0 tagged_above=0.0 required=5.0 tests= X-Spam-Level: X-Archive-Number: 200410/409 X-Sequence-Number: 8885 On Fri, Oct 22, 2004 at 05:13:18PM -0400, Matthew T. O'Connor wrote: > Yes that is the long term goal, but the autovac in 8.0 is still all or > nothing. Yes, which is why I couldn't use the current iteration for production: the cost is too high. I think this re-inforces my original point, which is that taking away the ability of DBAs to thump the planner for certain tables -- even indirectly -- under certain pathological conditions is crucial for production work. In the ideal world, the wizards and genius planners and such like would work perfectly, and the DBA would never have to intervene. In practice, there are cases when you need to haul on a knob or two. While this doesn't mean that we should adopt the Oracle approach of having knobs which adjust the sensitivity of the knob that tunes the main knob-tuner, I'm still pretty leery of anything which smacks of completely locking the DBA's knowledge out of the system. A -- Andrew Sullivan | ajs@crankycanuck.ca The plural of anecdote is not data. --Roger Brinner