Re: [Xen-devel] [PATCH 08 of 10 v2] libxl: enable automatic placement of guests on NUMA nodes

["Zhang, Yang Z" <yang.z.zhang@xxxxxxxxx>]

Dario Faggioli wrote on 2012-06-29:
> On Thu, 2012-06-28 at 15:41 +0300, Pasi KÃrkkÃinen wrote:
>>>> 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
>> 
> Yes, I know this slides, and I've also looked at most of the code
> they're referring to. That's all very interesting and it's well
> worthwhile to keep that in mind for future steps in the field of NUMA
> support. In particular, virtual topology awareness of the guest is
> something really important to have, and I'll be happy to
> discuss/work/contribute/help on it in the near future.
> 
> However, I'm still not sure I see how that ("guest NUMA") and live
> migration should affect (either now or during 4.3 dev cycle) the initial
> placement of a domain... That's what I was asking.

When doing migration, if the target machine is able to provide the same memory 
framework for the guest, we need to consider it firstly.

Best regards,
Yang

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel