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 E002B52A35 for ; Thu, 28 Jul 2005 15:27:47 -0300 (ADT) 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 20591-06 for ; Thu, 28 Jul 2005 18:27:45 +0000 (GMT) Received: from mail.Mi8.com (mail.mi8.com [63.240.6.50]) by svr1.postgresql.org (Postfix) with ESMTP id 104B352A15 for ; Thu, 28 Jul 2005 15:27:44 -0300 (ADT) Received: from 172.16.1.110 by mail.Mi8.com with ESMTP (- GW05 Welcome to Mi8 Corporation www.Mi8.com); Thu, 28 Jul 2005 14:27:38 -0400 X-Server-Uuid: E0C866E6-C6CD-48B5-AE61-E57E73CF3CC7 Received: from MI8NYCMAIL06.Mi8.com ([172.16.1.175]) by D01SMTP01.Mi8.com with Microsoft SMTPSVC(6.0.3790.211); Thu, 28 Jul 2005 14:27:37 -0400 Received: from 67.103.45.218 ([67.103.45.218]) by MI8NYCMAIL06.Mi8.com ( [172.16.1.219]) via Exchange Front-End Server mi8owa.mi8.com ( [172.16.1.104]) with Microsoft Exchange Server HTTP-DAV ; Thu, 28 Jul 2005 14:27:37 -0500 User-Agent: Microsoft-Entourage/11.1.0.040913 Date: Thu, 28 Jul 2005 11:27:36 -0700 Subject: Re: Finding bottleneck From: "Luke Lonergan" To: "Kari Lavikka" , pgsql-performance@postgresql.org Message-ID: In-Reply-To: MIME-Version: 1.0 X-OriginalArrivalTime: 28 Jul 2005 18:27:37.0969 (UTC) FILETIME=[0817CE10:01C593A2] X-WSS-ID: 6EF7FC101M815461377-01-01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.577 tagged_above=0 required=5 tests=AWL, FORGED_RCVD_HELO, RCVD_NUMERIC_HELO X-Spam-Level: * X-Archive-Number: 200507/445 X-Sequence-Number: 13686 On 7/28/05 2:21 AM, "Kari Lavikka" wrote: There's a new profiling tool called oprofile: http://oprofile.sourceforge.net/download/ that can be run without instrumenting the binaries beforehand. To actually find out what the code is doing during these stalls, oprofile can show you in which routines the CPU is spending time when you start/stop the profiling. As an alternative to the "guess->change parameters->repeat" approach, this is the most direct way to find the exact nature of the problem. - Luke