Re: [Xen-API] [RFC] Ballooning and live migration

[George Dunlap <George.Dunlap@xxxxxxxxxxxxx>]

Dave Scott wrote:
> > As for #2, I think that the following is simple enough:
> > * If free memory on migration target host >= VM memory target, just
> > transfer as is.
> > * If free memory on migration taget host < VM memory target, balloon
> > down to free memory.
> I think I agree that if the memory is completely free on the target then it's 
> sensible to just transfer as-is.
>
> If the memory isn't free on the target then we might have a choice about which 
> VMs to balloon: the VMs running on the target or the VM-to-be-migrated or both. 
> Do you have an opinion as to which is better? In the case of VM.start we figure 
> out what our target would be after the VM has started and host memory is 
> rebalanced, assuming no other VMs appear -- we could do a similar thing in 
> migrate perhaps.
Obviously that's a policy thing.  I chose the principles above because 
they were simple. :-)
Since we don't have any measurement of VM memory pressure at the moment, 
I suppose a logical thing to do would be to treat a migration on the 
target host similar to how we would treat  VM creation, wrt chosing new 
memory targets for running VMs, and the new VM.  It's basically the same 
problem, of deciding how much memory the new (to this host) VM should 
have, and taking it from other VMs.  The target can be chosen by the new 
host, passed to the old host, and if new_target < current_target, the VM 
can be ballooned down before migrating.
But I don't have an idea how complex that would be to implement. :-)

 -George

_______________________________________________
xen-api mailing list
xen-api@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/mailman/listinfo/xen-api