Received: from localhost (unknown [200.46.204.182]) by postgresql.org (Postfix) with ESMTP id A51B82E2E61 for ; Mon, 21 Jan 2008 20:26:45 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.182]) (amavisd-maia, port 10024) with ESMTP id 78192-01-2 for ; Mon, 21 Jan 2008 20:26:32 -0400 (AST) Received: from westnet.com (westnet.com [216.187.52.2]) by postgresql.org (Postfix) with ESMTP id 108802E2E65 for ; Mon, 21 Jan 2008 20:26:15 -0400 (AST) Received: from westnet.com (localhost [127.0.0.1]) by westnet.com (8.14.0/8.14.0) with ESMTP id m0M0QEac013176 for ; Mon, 21 Jan 2008 19:26:14 -0500 (EST) Received: from localhost (gsmith@localhost) by westnet.com (8.14.0/8.13.2/Submit) with ESMTP id m0M0QE4Z013173 for ; Mon, 21 Jan 2008 19:26:14 -0500 (EST) X-Authentication-Warning: westnet.com: gsmith owned process doing -bs Date: Mon, 21 Jan 2008 19:26:14 -0500 (EST) From: Greg Smith X-X-Sender: gsmith@westnet.com To: pgsql-performance@postgresql.org Subject: Re: 8.3 synchronous_commit In-Reply-To: <1200959154.10057.533.camel@dogma.ljc.laika.com> Message-ID: References: <1200959154.10057.533.camel@dogma.ljc.laika.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200801/168 X-Sequence-Number: 28407 On Mon, 21 Jan 2008, Jeff Davis wrote: > He was referring to the CFQ I/O scheduler. I don't think that will > affect pgbench itself, because it doesn't read/write to disk, right? It does if you are writing latency log files but it shouldn't be in the cases given. But there's something weird about the interplay between the server disk I/O under CFQ and the time slices the pgbench client gets when they're all on the same system. pgbench is certainly a badly scaling (in regards to how it handles simulating many clients) single process utility and anything that starves its execution sometimes is deadly. I never dug deep into the exact scheduler issues, but regardless in this case it's kind of unrealistic for Hannes to expect to simulate 100 clients on a small system and still get a true view of how the server on that system performs. -- * Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD