From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964898AbVHaRfV (ORCPT ); Wed, 31 Aug 2005 13:35:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S964902AbVHaRfU (ORCPT ); Wed, 31 Aug 2005 13:35:20 -0400 Received: from ns.virtualhost.dk ([195.184.98.160]:40127 "EHLO virtualhost.dk") by vger.kernel.org with ESMTP id S964898AbVHaRfT (ORCPT ); Wed, 31 Aug 2005 13:35:19 -0400 Date: Wed, 31 Aug 2005 19:35:26 +0200 From: Jens Axboe To: Holger Kiehl Cc: Vojtech Pavlik , linux-raid , linux-kernel Subject: Re: Where is the performance bottleneck? Message-ID: <20050831173525.GJ4018@suse.de> References: <20050829202529.GA32214@midnight.suse.cz> <20050831071126.GA7502@midnight.ucw.cz> <20050831072644.GF4018@suse.de> <20050831120714.GT4018@suse.de> <20050831162053.GG4018@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-IMAPbase: 1124875140 17 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Aug 31 2005, Holger Kiehl wrote: > ># ./oread /dev/sdX > > > >and it will read 128k chunks direct from that device. Run on the same > >drives as above, reply with the vmstat info again. > > > Using kernel 2.6.12.5 again, here the results: [snip] Ok, reads as expected, like the buffered io but using less system time. And you are still 1/3 off the target data rate, hmmm... With the reads, how does the aggregate bandwidth look when you add 'clients'? Same as with writes, gradually decreasing per-device throughput? -- Jens Axboe