[
https://issues.apache.org/jira/browse/YUNIKORN-468?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17276027#comment-17276027
]
Wilfred Spiegelenburg commented on YUNIKORN-468:
------------------------------------------------
I think we need to first sort out how and what we are going to do here. We need
a design on how we handle this and where the interactions are:
- what is the behaviour we want
- how does this relate to occupied resources of the node (non yunikorn
scheduled ones)
- what do we want to make configurable
- how does this interact with apps and queues and the rest of preemption
setting.
At the moment this is not a high priority as K8s can not change a node's size
as far as I know (i.e. nodes are not fixed in mem etc after creation). Before
we thus dive in lets figure out what we want/need.
> Node resource updates should trigger cleanup
> --------------------------------------------
>
> Key: YUNIKORN-468
> URL: https://issues.apache.org/jira/browse/YUNIKORN-468
> Project: Apache YuniKorn
> Issue Type: Improvement
> Components: core - common
> Reporter: Wilfred Spiegelenburg
> Priority: Minor
> Labels: pull-request-available
>
> Updating a node can set the overall size of a node and the resources
> allocated by a different scheduler (co-existence). The core just updates the
> node size and currently stops scheduling new resources even when the node is
> over subscribed.
> We should add a clean up in case that the node was shrunk and uses more than
> the resources available. the clean up should pre-empt enough resources to get
> it below the node size again.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
