From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [Xen-API] GSoC 2012 project brainstorming Date: Mon, 12 Mar 2012 18:48:04 +0100 Message-ID: <1331574484.8995.10.camel@Solace> References: <1330534437.4270.143.camel@zakaz.uk.xensource.com> <20120312170012.GI5369@phenom.dumpdata.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============6707244129216990381==" Return-path: In-Reply-To: <20120312170012.GI5369@phenom.dumpdata.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Konrad Rzeszutek Wilk Cc: Andre Przywara , xen-devel , Ian Campbell , xen-api@lists.xen.org List-Id: xen-devel@lists.xenproject.org --===============6707244129216990381== Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-nIKKP1ZXXwgO2c0Gyrt/" --=-nIKKP1ZXXwgO2c0Gyrt/ Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Mon, 2012-03-12 at 13:00 -0400, Konrad Rzeszutek Wilk wrote: > > HYPERVISOR > > ---------- > > - insmod Xen > > - event channel limits > > - NUMA >=20 > Yeah, we need that. > I'm looking into some aspects of this. Basically, I'm (trying to? :-D) putting some NUMA-aware VM placement logic in xl, basing the decision on how much free memory we have in each node. I'm also trying to have a meaningful set of benchmarks to better understand the performance implications of a sub-optimal placement. I'm hoping to have the implementation of this (memory-wise only for now) placemen logic soon, and I have a quite hard deadline for it placed at the end of the month, so... :-) As usual, any comment and ideas are more than welcome. > I think there were some patches posted for that.. whatever > happend to them? >=20 I've searched the list and the web for things like that and didn't find anything, but that could of course be my fault. Currently, my reference is what xm/xend does in order to deply VMs on the various nodes... I really hope the final result will look better than that, but that's enough for the first step. Finally, I'm not yet sure if there will be something suitable for GsoC here, but again I'm open to discuss any idea! :-) Regards, Dario --=20 <> (Raistlin Majere) ------------------------------------------------------------------- Dario Faggioli, http://retis.sssup.it/people/faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) PhD Candidate, ReTiS Lab, Scuola Superiore Sant'Anna, Pisa (Italy) --=-nIKKP1ZXXwgO2c0Gyrt/ Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEABECAAYFAk9eNtUACgkQk4XaBE3IOsTKHgCgpo3eelK/0GeQTcMGOGUQ4dGx 6wwAoJo8XwsvXtLWpF8hq5XDIOHxNLfs =ewQg -----END PGP SIGNATURE----- --=-nIKKP1ZXXwgO2c0Gyrt/-- --===============6707244129216990381== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============6707244129216990381==--