X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id 910CDD1B55D for ; Wed, 29 Oct 2003 22:27:20 +0000 (GMT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 09587-06 for ; Wed, 29 Oct 2003 18:26:51 -0400 (AST) Received: from mail.osdl.org (fw.osdl.org [65.172.181.6]) by svr1.postgresql.org (Postfix) with ESMTP id 18D48D1B559 for ; Wed, 29 Oct 2003 18:26:48 -0400 (AST) Received: from osdl.org (markw@ibm-b.pdx.osdl.net [172.20.1.51]) by mail.osdl.org (8.11.6/8.11.6) with ESMTP id h9TMQEC26620; Wed, 29 Oct 2003 14:26:15 -0800 Message-Id: <200310292226.h9TMQEC26620@mail.osdl.org> Date: Wed, 29 Oct 2003 14:26:11 -0800 (PST) From: markw@osdl.org Subject: Re: analyzing postgresql performance for dbt-2 To: pgman@candle.pha.pa.us Cc: pgsql-performance@postgresql.org, osdldbt-general@lists.sourceforge.net In-Reply-To: <200310260436.h9Q4a9f09899@candle.pha.pa.us> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200310/734 X-Sequence-Number: 4482 I've done a better controlled series of tests where I restore the database before each test and have grabbed sar and oprofile data: http://developer.osdl.org/markw/dbt2-pgsql/176/ - load of 100 warehouses - metric 1234.52 http://developer.osdl.org/markw/dbt2-pgsql/177/ - load of 120 warehouses - metric 1259.43 http://developer.osdl.org/markw/dbt2-pgsql/178/ - load of 140 warehouses - metric 1244.33 For the most part our primary metric, and the vmstat and sar output look fairly close for each run. Here are a couple of things that I've found to be considerably different from run 176 to 178: - oprofile says postgresql calls to SearchCatCache increased ~ 20% - readprofile says there are 50% more calls in the linux kernel to do_signaction (in kernel/signal.c) Would these two things offer any insight to what might be throttling the throughput? Thanks, Mark