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 42DB43A1D95 for ; Mon, 22 Nov 2004 23:13:42 +0000 (GMT) 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 18620-06 for ; Mon, 22 Nov 2004 23:13:31 +0000 (GMT) Received: from boutiquenumerique.com (boutiquenumerique.com [82.67.9.10]) by svr1.postgresql.org (Postfix) with ESMTP id AFD373A1D8E for ; Mon, 22 Nov 2004 23:13:32 +0000 (GMT) Received: (qmail 18860 invoked from network); 23 Nov 2004 00:13:59 +0100 Received: from unknown (HELO musicbox) (boutiquenumerique-lists@192.168.0.2) by boutiquenumerique.com with SMTP; 23 Nov 2004 00:13:59 +0100 Date: Tue, 23 Nov 2004 00:15:26 +0100 To: pgsql-performance@postgresql.org Subject: Re: scalability issues on win32 References: <6EE64EF3AB31D5448D0007DD34EEB3412A7532@Herge.rcsinc.local> From: =?iso-8859-15?Q?Pierre-Fr=E9d=E9ric_Caillaud?= Organization: =?iso-8859-15?Q?La_Boutique_Num=E9rique?= Content-Type: text/plain; format=flowed; delsp=yes; charset=iso-8859-15 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-ID: In-Reply-To: <6EE64EF3AB31D5448D0007DD34EEB3412A7532@Herge.rcsinc.local> User-Agent: Opera M2/7.54 (Linux, build 751) 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: 200411/341 X-Sequence-Number: 9295 > Test platform: > Pentium 4 3.06 GHz/HT > 10k SATA Raptor > 1Gb memory > Windows XP Pro SP2/Redhat Fedora 3 (64 bit results coming soon) Could you please add information about... - filesystems ? - windows configured as "network server" or as "desktop box" ? - virtual memory In my experience you MUST deactivate virtual memory on a Windows box to avoid catastrophic competition between virtual memory and disk cache - respective pgsql configurations (buffers...) identical ? - explain analyze for the two, identical ? - client on same machine or via network (100Mb ? 1G ?) - size of the data set involved in query - first query time after boot (with nothing in the cache), and times for the next disk-cached runs ? - are the N users doing the same query or exercising different parts of the dataset ? You don't do any writes in your test do you ? Just big SELECTs ?