Re: [VOTE] Inclusion of memory overcommit manager
On 09/28/2011 08:49 AM, Michael D Day wrote:
board-bounces(a)ovirt.org wrote on 09/28/2011 05:44:56 AM:
>
> my main concern is if this is going to be "the" policy engine for vdsm
> going forward, is if it shouldn't be established on something which is a
> rule based technology (say, pacemaker).
I think this is way too much speculation. I don't think anyone has proposed
that MOM should be a general-purpose policy engine or that it should be
"the" policy engine for VDSM.
I think the reason for the speculation is that we aren't being formal enough in
our proposals. I think Fedora Features are a good model for making proposals
like this:
http://fedoraproject.org/wiki/Features/Policy/Proposals
I'll work on some ovirt wiki pages for a similar project inclusion template this
afternoon.
Regards,
Anthony Liguori
In fact Adam pointed out that in some actual
deployments MOM is used discretely from VDSM and that is a concrete
advantage of having a separate daemon. Further, redesigning the project or
proposing a re-implementation on a different technology base is way beyond
the discussion. In fact, that latter suggestion in particular is something
that the MOM project should consider. The former (inclusion into VDSM) is
something the VDSM project should consider.
+1
Thanks,
Mike
Mike Day
IBM Distinguished Engineer
Chief Virtualization Architect, Open Systems Development
Cell: +1 919 371-8786 | mdday(a)us.ibm.com
http://code.ncultra.org
_______________________________________________
Board mailing list
Board(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/board