From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933189AbYDQPSo (ORCPT ); Thu, 17 Apr 2008 11:18:44 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759859AbYDQPSf (ORCPT ); Thu, 17 Apr 2008 11:18:35 -0400 Received: from webmail1.posta.tim.it ([213.230.128.226]:56797 "EHLO fep07-svc.tim.it" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1759776AbYDQPSe (ORCPT ); Thu, 17 Apr 2008 11:18:34 -0400 Message-ID: <48076A41.5040806@unimore.it> Date: Thu, 17 Apr 2008 17:18:25 +0200 From: Paolo Valente User-Agent: Thunderbird 2.0.0.12 (Windows/20080213) MIME-Version: 1.0 To: Aaron Carroll CC: Fabio Checconi , Jens Axboe , Pavel Machek , linux-kernel@vger.kernel.org Subject: Re: [RESEND][RFC] BFQ I/O Scheduler References: <20080401152903.GB34860@gandalf.sssup.it> <20080416184441.GA3923@ucw.cz> <4806EACB.7040408@unimore.it> <20080417071012.GP12774@kernel.dk> <480709A2.7040606@unimore.it> <48072549.7040104@cse.unsw.edu.au> <20080417111407.GS62286@gandalf.sssup.it> <48073F15.7070502@cse.unsw.edu.au> In-Reply-To: <48073F15.7070502@cse.unsw.edu.au> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Aaron Carroll ha scritto: > You still end up with reduced global throughput as > you've shown in the ``Short-term time guarantees'' table. It is an > interesting case though... since the lower performance is not though > fault > of the process it doesn't seem fair to ``punish'' it. Just a note about that table. The lower aggregate throughput of bfq is due to the fact that, because of the higher number of movies being read, a higher percentage of not-that-profitable accesses is being performed under bfq wrt to cfq. As shown in the complete logs of the aggregate throughput in the raw results, the aggregate throughput with bfq and cfq is practically the same when the number of movies is the same. The figure in the "Aggregate throughput" subsection is probably best suited for a comparison of the performance of the two schedulers with sequential accesses under the same conditions (the figure refers to the 2, 128 MB long, files, but we got virtually the same results in all the other tests). I do agree on that these experiments should be repeated with different (faster) devices. Paolo > > -- Aaron > >