[
https://issues.apache.org/jira/browse/YUNIKORN-462?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17453792#comment-17453792
]
Weiwei Yang commented on YUNIKORN-462:
--------------------------------------
hi [[email protected]], [~wilfreds]
{noformat}
schedulingCycle:
- select queue
- select app
-select request
- check node condition (A)
- allocate (B)
- notify the shim
{noformat}
AssumePod happens at B when the core decides to allocate a pod onto a node. The
predicates checks happen at A. The reason to have A and B in the same loop is,
in the next scheduling cycle, when we do node condition check, i.e the A step,
the evaluation of the predicate functions will be based on the latest data. E.g
pod/node affinity/anti-affinity, volume bindings, ports, IP addresses, etc.
This needs to happen in a SYNC manner otherwise the next scheduling cycle's
result might be wrong. E.g pod X is anti-affinity to Y, if we skip calling
AssumePod at B, then we could end up allocating X and Y onto the same node.
> Streamline core to shim update on allocation change
> ---------------------------------------------------
>
> Key: YUNIKORN-462
> URL: https://issues.apache.org/jira/browse/YUNIKORN-462
> Project: Apache YuniKorn
> Issue Type: Sub-task
> Components: core - scheduler, shim - kubernetes
> Reporter: Wilfred Spiegelenburg
> Assignee: Manikandan R
> Priority: Major
>
> Currently in the scheduler we have two updates that get send to the shim when
> an allocation is added or released:
> * event to shim RM event handler to allocate
> * reconciler plugin to update the shim caches
> Before YUNIKORN-317 one update was made in the cace the other in the
> scheduler. Now they are both in the scheduler in quick succession. The cache
> update in the shim is needed to make sure that the predicates are seeing the
> correct info. The event does the real bind etc of the allocation on the node.
> We should be able to fold the two calls into one call. However this requires
> changes on both sides and might even impact the SI as it will likely become a
> synced event call.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
