From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964811AbWGSM7D (ORCPT ); Wed, 19 Jul 2006 08:59:03 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S964812AbWGSM7D (ORCPT ); Wed, 19 Jul 2006 08:59:03 -0400 Received: from thunk.org ([69.25.196.29]:8348 "EHLO thunker.thunk.org") by vger.kernel.org with ESMTP id S964811AbWGSM7B (ORCPT ); Wed, 19 Jul 2006 08:59:01 -0400 Date: Wed, 19 Jul 2006 08:58:53 -0400 From: Theodore Tso To: joel Cc: linux-kernel@vger.kernel.org Subject: Re: filesystem tuning hints? Message-ID: <20060719125853.GA7093@thunk.org> Mail-Followup-To: Theodore Tso , joel , linux-kernel@vger.kernel.org References: <44BDAD5C.5020209@mainphrame.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <44BDAD5C.5020209@mainphrame.com> User-Agent: Mutt/1.5.11 X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: tytso@thunk.org X-SA-Exim-Scanned: No (on thunker.thunk.org); SAEximRunCond expanded to false Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org In order to answer your question about tuning a filesystem, it's a requirement to know something about the workload, and you haven't provided that. In general if you haven't been using the latest e2fsprogs creating ext3 with a larger journal size will help --- unless you don't have the requisite memory to support having extra pinned buffers, since your application may end up getting pushed out to swap. Many journalling filesystems can get much better performance by putting the journal on a separate external device, especially if that device is a battery-backed non-volitile ramdisk. With ext3, depending on your workload (if it is fsync-intensive, for example), using data journalling with can help a lot, especially with an external journal device. Finally, be warned that many filesystem benchmarks may be quite misleading, especially dbench, which doesn't match very many real-life workloads at all. It's designed to be half (the disk part) of an artificial web-based benchmark that was in itself a pretty bad benchmark that was easily subject to gaming and not very reflective of real-world workloads. Its main virtue is that it is easy to run and can be useful to developers as a good stress/smoke test to detect that new kernels haven't done something stupid, as long as you ignore its numbers. :-) In many cases, using your actual workload will be the best benchmark. Regards, - Ted