[Openstack-operators] limit num instance-type per host
marcos.garcia at enovance.com
Wed Sep 24 15:18:12 UTC 2014
IMO it's a matter of capacity planning and design to minimize
fragmentation. I don't know of any mechanisms to filter this and solve
(i.e. re-balance) a posteriori. Blueprints exists, though, to allow
re-scheduling and re-balancing.
Maybe I'm wrong and there are indeed some scheduler filters out of the
Anyways, I recommend you to read this:
and open this spreadsheet:
https://github.com/noslzzp/cloud-resource-calculator , it will show you
the optimal flavor configuration for minimal fragmetation using
different variables (vcpu, RAM, disk)
On 2014-09-24 11:08 AM, Blair Bethwaite wrote:
> Hi all,
> I'm trying to wrap my head around whether it's possible with the
> existing scheduler filters, to put a limit per host on the number of
> instances per instance-type/flavor? I don't think this is possible
> with the existing filters or weights, but it seems like a fairly
> common requirement.
> The issue I'm thinking of using this for is that of instance-to-host
> fragmentation in homogenous deployments, where there is a tendency as
> the zone approaches capacity to hit a utilisation ceiling - there are
> rarely any "gaps" large enough for high vcpu count instances. I'm
> guessing that limiting the number of smaller instances per host would
> help to alleviate this.
> Looks like knocking up such a filter wouldn't be too hard, just want
> to check whether there is another way...?
> OpenStack-operators mailing list
> OpenStack-operators at lists.openstack.org
Technical Sales Engineer
*PHONE : *(514) – 907 - 0068 *EMAIL :*marcos.garcia at enovance.com
<mailto:marcos.garcia at enovance.com> - *SKYPE : *enovance-marcos.garcia**
*ADDRESS :*127 St-Pierre – Montréal (QC) H2Y 2L6, Canada *WEB :
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the OpenStack-operators