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 A574CD1B52D for ; Tue, 28 Oct 2003 18:09:46 +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 34855-07 for ; Tue, 28 Oct 2003 14:09:16 -0400 (AST) Received: from mail.osdl.org (fw.osdl.org [65.172.181.6]) by svr1.postgresql.org (Postfix) with ESMTP id 5BD6DD1B538 for ; Tue, 28 Oct 2003 14:09:15 -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 h9SI8ao01456; Tue, 28 Oct 2003 10:08:37 -0800 Message-Id: <200310281808.h9SI8ao01456@mail.osdl.org> Date: Tue, 28 Oct 2003 10:08:33 -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: 200310/706 X-Sequence-Number: 4454 On 26 Oct, Bruce Momjian wrote: > Mark Wong wrote: >> > > Here are a pair of results where I just raise the load on the >> > > database, where increasing the load increases the area of the database >> > > touched in addition to increasing the transaction rate. The overall >> > > metric increases somewhat, but the response time for most of the >> > > interactions also increases significantly: >> > > >> > > http://developer.osdl.org/markw/dbt2-pgsql/158/ [baseline] >> > > - load of 100 warehouses >> > > - metric 1249.65 >> > > >> > > http://developer.osdl.org/markw/dbt2-pgsql/149/ >> > > - load of 140 warehouses >> > > - metric 1323.90 >> > >> > I looked at these charts and they looked normal to me. It looked like >> > your the load increased until your computer was saturated. Is there >> > something I am missing? >> >> I've run some i/o tests so I'm pretty sure I haven't saturated that. And it >> looks like I have almost 10% more processor time left. I do agree that it >> appears something might be saturated, I just don't know where to look... > > Could the 10% be context switching time, or is the I/O saturated? There are about 14,000 to 17,000 context switches/s according to the vmstat output. This is on a 1.5Ghz hyperthreaded Xeon processor. I don't know what I'm supposed to be able to expect in terms of context switching. I really doubt the i/o is saturated because I've run disktest (part of the Linux Test Project suite) and saw much higher throughput for various sequential/random read/write tests. I'm starting to collect oprofile data (and will hopefully have some results soon) to get an idea where the database is spending its time, just in case that may have something to do with it. Mark