X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id DEBCAD8E2A for ; Tue, 8 Nov 2005 09:45:18 -0400 (AST) 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 92820-02 for ; Tue, 8 Nov 2005 13:45:13 +0000 (GMT) X-Greylist: delayed 03:10:43.862651 by SQLgrey- Received: from svr2.postgresql.org (svr2.postgresql.org [65.19.161.25]) by svr1.postgresql.org (Postfix) with ESMTP id D39BAD893A for ; Tue, 8 Nov 2005 09:45:11 -0400 (AST) Received: from relay.icomedias.com (office.icomedias.com [62.99.232.80]) by svr2.postgresql.org (Postfix) with ESMTP id A4AD3F0D67 for ; Tue, 8 Nov 2005 10:34:27 +0000 (GMT) Received: from loki.icomedias.com ([10.192.17.128]) by relay.icomedias.com (8.13.4/8.13.4) with ESMTP id jA8AYGDS031660; Tue, 8 Nov 2005 11:34:16 +0100 From: Mario Weilguni To: pgsql-performance@postgresql.org Subject: Re: 8.1 iss Date: Tue, 8 Nov 2005 11:34:19 +0100 User-Agent: KMail/1.8.92 Cc: "PostgreSQL" References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200511081134.19495.mweilguni@sime.com> X-Scanned-By: MIMEDefang 2.43 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=0.269 required=5 tests=[AWL=0.269] X-Spam-Score: 0.269 X-Spam-Level: X-Archive-Number: 200511/139 X-Sequence-Number: 15396 Am Montag, 7. November 2005 18:22 schrieb PostgreSQL: > My most humble apologies to the pg development team (pg_lets?). > > I took Greg Stark's advice and set: > > shared_buffers = 10000 # was 50000 > work_mem = 1048576 # 1Gb - was 16384 > > Also, I noticed that the EXPLAIN ANALYZE consistently thought reads would > take longer than they actually did, so I decreased random_page_cost down to > 1 (the server has a SATA Raid at level 10). Don't do that, use 1.5 or 2, setting it to 1 will only work well if you have small databases fitting completly in memory.