From: Ted Ts'o Subject: Re: [Ocfs2-devel] [PATCH, RFC 0/3] *** SUBJECT HERE *** Date: Tue, 3 Aug 2010 18:57:47 -0400 Message-ID: <20100803225747.GG9453@thunk.org> References: <1280851315-9167-1-git-send-email-tytso@mit.edu> <20100803190703.GA15416@mail.oracle.com> <20100803200755.GD9453@thunk.org> <20100803211916.GB15416@mail.oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: Ext4 Developers List , ocfs2-devel@oss.oracle.com, Keith Maanthey , John Stultz , Eric Whitney Return-path: Received: from thunk.org ([69.25.196.29]:38080 "EHLO thunker.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757682Ab0HCW5r (ORCPT ); Tue, 3 Aug 2010 18:57:47 -0400 Content-Disposition: inline In-Reply-To: <20100803211916.GB15416@mail.oracle.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue, Aug 03, 2010 at 02:19:16PM -0700, Joel Becker wrote: > On Tue, Aug 03, 2010 at 04:07:55PM -0400, Ted Ts'o wrote: > > If you are set up to do some performance measurements on OCFS2, I'd > > appreciate if you could give it a try and let me know how the patches > > fare on OCFS2. > > We're lining up a 16-way box here. Do you have a happy dbench > command line or similar that you would like to see run? The problem with dbench is that the core VFS can end up being the bottleneck. OTOH, it's very easy to run. What would be great if you could try using the Flexible File System Benchmark (FFSB), which is what Eric and Steve used: http://free.linux.hp.com/~enw/ext4/2.6.34/ http://btrfs.boxacle.net/ It's a bit more work to set it up, though. - Ted