From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?= Subject: Re: [PATCH 08 of 10 v2] libxl: enable automatic placement of guests on NUMA nodes Date: Thu, 28 Jun 2012 15:41:06 +0300 Message-ID: <20120628124106.GL2058@reaktio.net> References: <81f18379bb3d4d9397d1.1339779876@Solace> <1340878329.21008.2.camel@Solace> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1340878329.21008.2.camel@Solace> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Dario Faggioli Cc: Andre Przywara , Ian Campbell , Stefano Stabellini , George Dunlap , Juergen Gross , Ian Jackson , "xen-devel@lists.xen.org" , "Zhang, Yang Z" List-Id: xen-devel@lists.xenproject.org On Thu, Jun 28, 2012 at 12:12:09PM +0200, Dario Faggioli wrote: > On Thu, 2012-06-28 at 07:25 +0000, Zhang, Yang Z wrote: > > > This all happens internally to libxl, and no API for driving the mechanism is > > > provided for now. This matches what xend already does. > > we should consider the basic IONUMA. I mean when a guest with a device assigned, > > from the point of view of performance, it's better to allocated the memory from the node > > which the device belongs to. > > > That sure makes sense, and it would be a very useful extension! > > > Furthermore, we need consider the guest NUMA(I am working on it now) and live migration. > > > Great. Can you share, either here on or a separate thread some more > details about what you mean by both "guest NUMA" and "live migration" in > this context? > > As George said, that would be 4.3 material, but I'm still very > interested. :-) > For the mailinglist archives.. here are the Xen guest VM NUMA slides from XenSummit 2010: http://www.slideshare.net/xen_com_mgr/nakajima-numafinal http://www.slideshare.net/xen_com_mgr/dulloor-xensummit -- Pasi