From owner-freebsd-xen@FreeBSD.ORG Mon Mar 31 10:49:37 2014 Return-Path: Delivered-To: freebsd-xen@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 0C831CFE; Mon, 31 Mar 2014 10:49:37 +0000 (UTC) Received: from SMTP02.CITRIX.COM (smtp02.citrix.com [66.165.176.63]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "mail.citrix.com", Issuer "Cybertrust Public SureServer SV CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 2E05999C; Mon, 31 Mar 2014 10:49:35 +0000 (UTC) X-IronPort-AV: E=Sophos;i="4.97,764,1389744000"; d="scan'208";a="115219661" Received: from accessns.citrite.net (HELO FTLPEX01CL02.citrite.net) ([10.9.154.239]) by FTLPIPO02.CITRIX.COM with ESMTP; 31 Mar 2014 10:49:27 +0000 Received: from [IPv6:::1] (10.80.16.47) by smtprelay.citrix.com (10.13.107.79) with Microsoft SMTP Server id 14.2.342.4; Mon, 31 Mar 2014 06:49:26 -0400 Message-ID: <5339482E.9020301@citrix.com> Date: Mon, 31 Mar 2014 12:49:18 +0200 From: =?ISO-8859-1?Q?Roger_Pau_Monn=E9?= User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:24.0) Gecko/20100101 Thunderbird/24.4.0 MIME-Version: 1.0 To: Big Lebowski , Mark Felder Subject: Re: Terrible performance of XenServer 6.2 and FreeBSD 10.0-RELEASE guest References: <8FDB514D-1A94-48B2-AE19-62923F327938@FreeBSD.org> In-Reply-To: X-Enigmail-Version: 1.6 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-DLP: MIA1 Cc: freebsd-xen@freebsd.org X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 31 Mar 2014 10:49:37 -0000 On 29/03/14 14:30, Big Lebowski wrote: > This was enabled, but I've now disabled it. However, the network I/O seems > to be ok, on par with what I can achieve on Linux boxes. Its disk I/O that > makes the system unusable. I've done a `tar -Jxf src.txz` on identical boxes (which I think should generate a similar IO workload to what you are doing), one running FreeBSD natively, and the other running a FreeBSD PVHVM guest on top of Xen: Native: # time tar -Jxf src.txz real 2m15.280s user 0m10.021s sys 0m5.257s Xen guest: # time tar -Jxf src.txz real 3m13.366s user 0m11.430s sys 0m4.767s Which doesn't look that bad. When running under Xen I didn't experience any kind of noticeable slowness while running this workload on screen and switched to another console. If you think disk IO is the culprit, could you provide some synthetic workload with fio that can reproduce this? (Also the dmesg of the FreeBSD guest might be of interest) My test was done using OSS Xen 4.5 and a Linux Dom0 running kernel 3.13. Roger.