[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH 1 of 3] libxl: take node distances into account during NUMA placement



On Fri, 2012-10-19 at 20:02 +0200, Dario Faggioli wrote: 
> I am. In fact, I think they answer George's question which was "by how
> much we are increasing the complexity of each step?" and not "by how
> much we are increasing the overall complexity?".
> 
> That of course doesn't mean I don't care about the overall complexity,
> just that I don't think this is making things worse than what we have
> (which is, btw, the result of the long discussion we had pre 4.2
> release).
> 
> > It seems to me that 
> > doing Nnodes^2 for each candidate multiplies the cost by the number of
> > candidates, rather than adding Nnodes^2.
> > 
> That is definitely true. Again the key is the difference between talking
> about "each candidate", as we were, and total.
> 
Which, BTW, make me thinking that you don't like the fact that, for each
candidate, we go through the list of domains to check what vcpus are
bound to such candidate.

That was requested during review of the automatic placement series,
reviewed itself, tested on an 8 nodes box with 50-60 domains and became
cs 4165d71479f9.

Anyway, if you want, I think I can change that too. :-)

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: signature.asc
Description: This is a digitally signed message part

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

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.