[one-users] A question about the utilization of VMs

Adnan Pasic pqado at yahoo.de
Thu Nov 24 12:16:29 PST 2011


Is there a way to circumvent that? For my diploma thesis I need to utilize
deployed VMs for a period of up to 12 hours. However, if KSM is active, the
physical host doesn't hold this utilization, but adjusts the used memory
pages.
Since I want to measure the power consumption of the physical host under
full VM utilization, I need the real memory to stay utilized as well, yet I
don't want to be forced to kill a VM in order to free the memory after 12
hours of load (as there would be no need to use opennebula then. I still
want be able to migrate my VMs :)).
Can you give me a hint?

-----Ursprüngliche Nachricht-----
Von: rsmontero at gmail.com [mailto:rsmontero at gmail.com] Im Auftrag von Ruben
S. Montero
Gesendet: Donnerstag, 24. November 2011 21:10
An: Adnan Pasic
Cc: users at lists.opennebula.org
Betreff: Re: [one-users] A question about the utilization of VMs

Well that's normal if not usinng balloning. The hypervisor (outside
the VM) allocates the whole memory till the VM is destroyed

On Thu, Nov 24, 2011 at 10:19 AM, Adnan Pasic <pqado at yahoo.de> wrote:
> Okay, here is what I do:
> - first I turned off the KSM function on the host that is going to hold my
> VM
> - then: onevm create vm_2GB.qemu
> - then: ssh user at vm_2GB
> - then: lookbusy -m 2000M
>  --> Via "top" or "free" I can see that this VM is fully utilized now!
>  --> In the physical machine I get the same result - 2GB utilized, which
is
> as expected!
> - then, after some time: killall lookbusy (from inside the VM)
>  --> "top" or "free" now say that the VM is empty in regard to memory
> utilization
> - However when I now execute the "free" command on my host, I get this:
>
>             total       used       free     shared    buffers     cached
> Mem:          3023       2933         89          0         21        675
> -/+ buffers/cache:       2236        786
> Swap:         6234          9       6225
>
> As you can see the utilized memory is not cached or whatsoever, but really
> utilized (as we can see in line 2: 2236 MB used!!!).
> Any ideas? :-(
>
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>



-- 
Dr. Ruben Santiago Montero
Associate Professor (Profesor Titular), Complutense University of Madrid

URL: http://dsa-research.org/doku.php?id=people:ruben
Weblog: http://blog.dsa-research.org/?author=7




More information about the Users mailing list