From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mout.kundenserver.de ([212.227.126.187]:58774 "EHLO mout.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756119AbcCDEiC (ORCPT ); Thu, 3 Mar 2016 23:38:02 -0500 Message-ID: <56D91124.8050905@vlnb.net> Date: Thu, 03 Mar 2016 20:37:56 -0800 From: Vladislav Bolkhovitin MIME-Version: 1.0 Subject: Re: Fio high IOPS measurement mistake References: <56D525E1.6010407@vlnb.net> <56D66B1C.6050506@vlnb.net> <56D7A97F.9080800@vlnb.net> In-Reply-To: Content-Type: text/plain; charset=windows-1251 Content-Transfer-Encoding: 7bit Sender: fio-owner@vger.kernel.org List-Id: fio@vger.kernel.org To: Sitsofe Wheeler Cc: "fio@vger.kernel.org" Sitsofe Wheeler wrote on 03/02/2016 11:13 PM: > On 3 March 2016 at 07:10, Sitsofe Wheeler wrote: >> On 3 March 2016 at 03:03, Vladislav Bolkhovitin wrote: >>> For those who asked about perf profiling, it remained the same as before with the CPU >>> consumption is all about timekeeping and memset: >>> >>> - 55.74% fio fio [.] clock_thread_fn >>> clock_thread_fn >> >> Perhaps this is what is already included above but could you use the >> -g option on perf to collect it into a call-graph and post the top >> results? > > One extra question: do you see a difference between when you use > threads (thread=1) and when you use processes (thread=0)? Nothing serious (few %%), which could be easily explained, because I'm running number of jobs less, than CPU cores count, so each thread or process has a dedicated CPU core. Thanks, Vlad