From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ge0.mail1.hoer.dk.ip.fullrate.dk ([90.185.1.42]:54270 "EHLO smtp.fullrate.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S966133Ab0CPL1l (ORCPT ); Tue, 16 Mar 2010 07:27:41 -0400 Received: from arvinserver1.home.troels.arvin.dk (1608ds2-ksa.0.fullrate.dk [90.184.70.86]) by smtp.fullrate.dk (Postfix) with ESMTP id 713479CFD9 for ; Tue, 16 Mar 2010 12:27:39 +0100 (CET) Received: from localhost (arvinserver1.home.troels.arvin.dk [127.0.0.1]) by arvinserver1.home.troels.arvin.dk (Postfix) with ESMTP id 4D524108124 for ; Tue, 16 Mar 2010 12:27:36 +0100 (CET) Received: from arvinserver1.home.troels.arvin.dk ([127.0.0.1]) by localhost (arvinserver1.home.troels.arvin.dk [127.0.0.1]) (amavisd-new, port 10024) with LMTP id MuHr7gi2macA for ; Tue, 16 Mar 2010 12:27:31 +0100 (CET) Received: from [192.168.216.2] (sst-gw [147.29.225.156]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by arvinserver1.home.troels.arvin.dk (Postfix) with ESMTP id 297A81080EE for ; Tue, 16 Mar 2010 12:27:31 +0100 (CET) Message-ID: <4B9F6B25.7070807@arvin.dk> Date: Tue, 16 Mar 2010 12:27:33 +0100 From: Troels Arvin MIME-Version: 1.0 Subject: Re: Does fio write only 0x00s? References: <4B9E9D92.80701@arvin.dk> <20100315212219.GQ5768@kernel.dk> <4B9EC06E.2000101@arvin.dk> <20100316080528.GX5768@kernel.dk> In-Reply-To: <20100316080528.GX5768@kernel.dk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: fio-owner@vger.kernel.org List-Id: fio@vger.kernel.org To: fio@vger.kernel.org Hello, Den 16-03-2010 09:05, skrev Jens Axboe: > I think what you are missing is that the random writes will create a > large sparse file. The Output is 10G, and you are doing a lot of reads. > So you could end up writing only 30% of the 10G, the rest would be > sparse holes in the file. Yes, that part about the file sparseness was I had to grasp. Expressed in another way: If I filter out the null-bytes from fio's work-file (cat iometer.1.0 | tr -d '\0'), then the remainder doesn't compress at all using gzip with default options. So I'll stop worrying about storage-side compression/de-duplication when interpreting fio's results. Thanks. -- Regards, Troels Arvin http://troels.arvin.dk/