From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757403Ab0EGRtG (ORCPT ); Fri, 7 May 2010 13:49:06 -0400 Received: from 64-131-60-146.usfamily.net ([64.131.60.146]:44465 "EHLO mail.sandeen.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756292Ab0EGRtD (ORCPT ); Fri, 7 May 2010 13:49:03 -0400 Message-ID: <4BE4528C.5000305@sandeen.net> Date: Fri, 07 May 2010 12:49:00 -0500 From: Eric Sandeen User-Agent: Thunderbird 2.0.0.24 (Macintosh/20100228) MIME-Version: 1.0 To: Bernd Eckenfels CC: linux-kernel@vger.kernel.org Subject: Re: ext4 is faster with LVM than without, and other filesystem benchmarks References: <201005071528.o47FSVas054141@neskaya.eckenfels.net> In-Reply-To: <201005071528.o47FSVas054141@neskaya.eckenfels.net> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Bernd Eckenfels wrote: > In article <20100507142310.GF13143@lh.kyla.fi> you wrote: >> I think the most surprising result of the benchmarks is that ext4 >> seems to be significantly faster under LVM than on the raw /dev/sdd1 >> partition. Without LVM the variance in run times is much larger. > > Sounds like the syncs/barriers never reach the disk with LVM. What mount > journal/sync options did you use? default? Is dmesg telling you that > barriers have been turned off for the lvm case? IF yo uwant it fast and > dirty try nobarrier,data=writeback :) LVM barriers should be working in that kernel, though... still, comparing mount -o nobarrier on each might yield a clue there. -Eric