X-Original-To: pgsql-performance-postgresql.org@postgresql.org Received: from localhost (mx1.hub.org [200.46.208.251]) by postgresql.org (Postfix) with ESMTP id 5A0159FB388 for ; Wed, 12 Jul 2006 16:51:09 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.208.251]) (amavisd-new, port 10024) with ESMTP id 93196-03-2 for ; Wed, 12 Jul 2006 16:51:00 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from wolff.to (wolff.to [66.93.197.194]) by postgresql.org (Postfix) with SMTP id 2508A9FB47C for ; Wed, 12 Jul 2006 16:50:39 -0300 (ADT) Received: (qmail 7758 invoked by uid 500); 12 Jul 2006 19:25:39 -0000 Date: Wed, 12 Jul 2006 14:25:39 -0500 From: Bruno Wolff III To: "Koth, Christian \(DWBI\)" Cc: pgsql-performance@postgresql.org Subject: Re: Commit slower on faster PC Message-ID: <20060712192539.GD29575@wolff.to> Mail-Followup-To: Bruno Wolff III , "Koth, Christian (DWBI)" , pgsql-performance@postgresql.org References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200607/117 X-Sequence-Number: 19977 On Wed, Jul 12, 2006 at 10:16:40 -0600, "Koth, Christian (DWBI)" wrote: > > I have noticed a strange performance behaviour using a commit statement on two different machines. On one of the machines the commit is many times faster than on the other machine which has faster hardware. Server and client are running always on the same machine. > > Server version (same on both machines): PostgreSQL 8.1.3. (same binaries as well) > > PC1: > ---- > IDE-HDD (approx. 50 MB/s rw), fs: ext3 > > PC2: > ---- > SCSI-HDD (approx. 65 MB/s rw), fs: ext3 > > Both installations of the database have the same configuration, different from default are only the following settings on both machines: > > pgbench gives me the following results: > PC1: > ---- > tps = 293.625393 (excluding connections establishing) > > PC2: > ---- > tps = 46.519634 (excluding connections establishing) Have you checked to see if the ide drive is lying about having written the data to the platters?