Re: [PATCH 1 of 3 v5/leftover] libxl: enable automatic placement of guests on NUMA nodes

[Dario Faggioli <raistlin@linux.it>]

[-- Attachment #1.1: Type: text/plain, Size: 1492 bytes --]

On Fri, 2012-07-20 at 11:39 +0200, Dario Faggioli wrote:
> > If we somehow determine that this approach doesn't work (no nodes with 
> > enough free memory or more vCPUs than CPUs-per-node) we should use the 
> > sophisticated algorithm.
> > 
> And again, you'll see the new code and will tell me what you think
> later, but I really think I turned it into something like that. The only
> thing 
> 
Some text got cut here.

What I wanted to ask is what you think we should do with guests that
have more vcpus than pcpus of a node. Right now, as I do for memory, I'm
using it as a reason to consider candidates with more than one node.
XenD did something similar, as it adds nodes (although it does that
afterwords) until the above stop being true.

Should we ignore that and stick to the number of nodes that gives us
enough memory? Should we do that but only up to a certain extent (e.g.,
by asking for a candidate to have at least half pcpus as the domain has
vcpus)? After all, even if we give the domain enough pcpus, he's not
going to have exclusive access to them anyway...

Again, just random thought seeking for a confirmation that is very hard
to find! :-)

Thanks and Regards,
Dario

-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://retis.sssup.it/people/faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)


[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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