From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefan Priebe Subject: Re: improve speed with auth supported=none Date: Tue, 13 Nov 2012 09:06:55 +0100 Message-ID: <50A1FF9F.7010705@profihost.ag> References: <50A117B9.4060804@profihost.ag> <50A170BF.5030402@profihost.ag> <50A1F9F8.4030808@inktank.com> <50A1FC4D.8010000@profihost.ag> <50A1FF21.2020700@inktank.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.profihost.ag ([85.158.179.208]:42670 "EHLO mail.profihost.ag" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754290Ab2KMIG5 (ORCPT ); Tue, 13 Nov 2012 03:06:57 -0500 In-Reply-To: <50A1FF21.2020700@inktank.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Josh Durgin Cc: =?ISO-8859-1?Q?S=E9bastien_Han?= , "ceph-devel@vger.kernel.org" , Alexandre DERUMIER Am 13.11.2012 09:04, schrieb Josh Durgin: > On 11/12/2012 11:52 PM, Stefan Priebe wrote: >> Am 13.11.2012 08:42, schrieb Josh Durgin: >>> On 11/12/2012 01:57 PM, Stefan Priebe wrote: >>>> Thanks, >>>> >>>> this gives another burst for iops. I'm now at 23.000 iops ;-) So for >>>> random 4k iops ceph auth and especially the logging is a lot of >>>> overhead. >>> >>> How much difference did disabling auth make vs only disabling logging? >> >> disable debug logging: 3000 iops >> disable auth logging: 2000 iops >> >> Is anybody in the ceph team also interested in a call graph of kvm when >> VM is doing random 4k write io? > > I'm interested, although I'm not sure how easy it'll be to find > bottlenecks from it. It's worth a shot. *arg* kvm process segfaults when i use google perf tools to create / record one. Any software suggestion? Stefan