Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753093Ab3IEN3p (ORCPT ); Thu, 5 Sep 2013 09:29:45 -0400 Received: from Mycroft.westnet.com ([216.187.52.7]:38201 "EHLO mycroft.westnet.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751941Ab3IEN3m (ORCPT ); Thu, 5 Sep 2013 09:29:42 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <21032.34644.131559.360272@quad.stoffel.home> Date: Thu, 5 Sep 2013 09:29:56 -0400 From: "John Stoffel" To: Waiman Long Cc: John Stoffel , Alexander Viro , Linus Torvalds , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, "Chandramouleeswaran, Aswin" , "Norton, Scott J" Subject: Re: [PATCH] dcache: Translating dentry into pathname without taking rename_lock In-Reply-To: <5227E6AF.2040603@hp.com> References: <1378321523-40893-1-git-send-email-Waiman.Long@hp.com> <52278970.2080803@hp.com> <21031.39606.16374.72229@quad.stoffel.home> <5227E6AF.2040603@hp.com> X-Mailer: VM 8.2.0b under 23.2.1 (x86_64-pc-linux-gnu) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1793 Lines: 40 >>>>> "Waiman" == Waiman Long writes: Waiman> On 09/04/2013 04:40 PM, John Stoffel wrote: >>>>>>> "Waiman" == Waiman Long writes: Waiman> In term of AIM7 performance, this patch has a performance boost of Waiman> about 6-7% on top of Linus' lockref patch on a 8-socket 80-core DL980. >> Waiman> User Range | 10-100 | 200-10000 | 1100-2000 | Waiman> Mean JPM w/o patch | 4,365,114 | 7,211,115 | 6,964,439 | Waiman> Mean JPM with patch | 3,872,850 | 7,655,958 | 7,422,598 | Waiman> % Change | -11.3% | +6.2% | +6.6% | >> >> This -11% impact is worisome to me, because at smaller numbers of >> users, I would still expect the performance to go up. So why the big >> drop? >> >> Also, how is the impact of these changes on smaller 1 socket, 4 core >> systems? Just because it helps a couple of big boxes, doesn't mean it >> won't hurt the more common small case. >> >> John Waiman> I don't believe the patch will make it slower with less Waiman> user. It is more a result of run-to-run variation. The short Waiman> workload typically completed in a very short time. In the Waiman> 10-100 user range, the completion times range from Waiman> 0.02-0.11s. With a higher user count, it needs several seconds Waiman> to run and hence the results are more reliable. Can you then show the variation over multiple runs? I think you have a good justification for larger boxes to make this change, I just worry about smaller systems getting hit and losing performance. John -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/