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 168699FB348 for ; Mon, 21 Aug 2006 18:43:39 -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 77390-08 for ; Mon, 21 Aug 2006 18:43:30 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from servata.com (ip-216-152-249-241.servata.com [216.152.249.241]) by postgresql.org (Postfix) with ESMTP id 4E0F89FAE4E for ; Mon, 21 Aug 2006 18:23:30 -0300 (ADT) Received: from [209.162.219.253] (helo=dogma.v10.wvs) by servata.com with esmtpa (Exim 4.50) id 1GFHEz-0001yz-7e; Mon, 21 Aug 2006 14:23:25 -0700 Subject: Re: How to get higher tps From: Jeff Davis To: Marty Jia Cc: pgsql-performance@postgresql.org In-Reply-To: <0B9A8C89DCC3AB488A78A4DE0FECDA90019E30D4@SITE3MAIL04.jeeves.ask.info> References: <0B9A8C89DCC3AB488A78A4DE0FECDA90019E30D4@SITE3MAIL04.jeeves.ask.info> Content-Type: text/plain Date: Mon, 21 Aug 2006 14:23:26 -0700 Message-Id: <1156195406.15743.28.camel@dogma.v10.wvs> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 (2.0.4-7) Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.294 tagged_above=0 required=5 tests=AWL, FORGED_RCVD_HELO, SPF_HELO_PASS X-Spam-Level: X-Archive-Number: 200608/321 X-Sequence-Number: 20506 On Mon, 2006-08-21 at 16:45 -0400, Marty Jia wrote: > I'm exhausted to try all performance tuning ideas, like following > parameters > > shared_buffers > fsync By "tuning" fsync, what do you mean? Did you turn it off? If you turned fsync off, that could compromise your data in case of any kind of crash or power failure. However, if you turn fsync off you should much higher TPS on pgbench than you're getting. > Dual Intel Xeon 2.8GHz > 6GB RAM > Linux 2.4 kernel > RedHat Enterprise Linux AS 3 > 200GB for PGDATA on 3Par, ext3 > 50GB for WAL on 3Par, ext3 Does your disk controller have battery-backed writeback cache? How much? > With PostgreSql 8.1.4 > > We don't have i/o bottle neck. > Well, chances are PostgreSQL is waiting for fsync, which means you do have an I/O bottleneck (however, you're not using all of your I/O bandwidth, most likely). Regards, Jeff Davis