From: Eric Whitney Subject: 3.1-rc2 and 3.0 filesystem scalability measurements Date: Fri, 09 Sep 2011 19:35:58 -0400 Message-ID: <4E6AA2DE.2060903@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: Ext4 Developers List , linux-fsdevel@vger.kernel.org Return-path: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org (Several filesystem developers have asked that I post this on the fsdevel as well the ext4 mailing lists.) I've posted the results of some 3.1-rc2 / 3.0 ext4 scalability measurements and comparisons on a 48 core x86_64 server at: http://free.linux.hp.com/~enw/ext4/3.1-rc2 This includes throughput and CPU efficiency graphs for five simple workloads, the raw data for same, and lockstats as well. The data have been useful in improving ext4 scalability in the past. The data cover ext4 filesystems with and without journals. For reference, ext3, xfs, and btrfs are included as well. The most notable improvement in this data is a significant scalability gain for xfs when running the large_file_creates workload (the lockstats suggest a behavioral change). This was first visible in 3.1-rc1. ext4 remains relatively unchanged. Dave Chinner has asked that I make some changes to the way I collect xfs data, and I'll try to address those before I post again. Thanks, Eric