[jira] [Commented] (CLOUDSTACK-6650) Reorder Cluster list in deployment planner to protect GPU enabled hosts from non-GPU VM deployment

["ASF subversion and git services (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CLOUDSTACK-6650?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14154700#comment-14154700 ] 

ASF subversion and git services commented on CLOUDSTACK-6650:
-------------------------------------------------------------

Commit 39fe766c2b6fb6edd4c1bf828625b29d9bb87719 in cloudstack's branch refs/heads/master from [~sanjay.tripathi]
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=39fe766 ]

CLOUDSTACK-6650: Reorder Cluster list in deployment planner to protect
GPU enabled hosts from non-GPU VM deployment.
Cluster reordering is based on the number of unique host tags in a cluster,
cluster with most number of unique host tags will put at the end of list.
Hosts with GPU capability will get tagged with implicit tags defined by
global config param 'implicit.host.tags' at the time os host discovery.

Also added FirstFitPlannerTest unit test file.


> Reorder Cluster list in deployment planner to protect GPU enabled hosts from non-GPU VM deployment
> --------------------------------------------------------------------------------------------------
>
>                 Key: CLOUDSTACK-6650
>                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-6650
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the default.) 
>         Environment: MS 4.4
> XS 620SP1
>            Reporter: Sanjay Tripathi
>            Assignee: Sanjay Tripathi
>             Fix For: 4.4.0
>
>
> If CS environment has GPU cards then for non-GPU enabled VM deployments, deployment planner should check the non-GPU hosts first and if they are out of capacity then deploy in  GPU enabled hosts.
> For now, this restriction is imposed at cluster level not at zone level, so to protect GPU resources, CS should give lower priority to clusters which has GPU enabled host and high priority to clusters which contains non-GPU enabled hosts.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)