[
https://issues.apache.org/jira/browse/FLINK-31843?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17746044#comment-17746044
]
Yangze Guo commented on FLINK-31843:
------------------------------------
master:
3efbae44b4a21c9735f1a3054c276c32bbc46cdd
f1a079f95e409d3a81ff5dd199ea3c0dc30470cc
a69f96514f3da893f6a048b107ea287411f58714
3cdf228191a073c7008f88faa0b088c6a2428012
72bff2a2d0072602e4e625476bf5480dc50dc76c
> Select slots from SlotPool#freeSlots in bulk
> --------------------------------------------
>
> Key: FLINK-31843
> URL: https://issues.apache.org/jira/browse/FLINK-31843
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Reporter: Weihua Hu
> Assignee: Weihua Hu
> Priority: Major
> Labels: pull-request-available
>
> we should also reduce the number of calls of "getFreeSlotInformations". In
> current implementation, the scheduler will batch request slots for tasks in
> the same pipeline region(ExecutionSlotAllocator#allocateSlotsFor), but the
> slot allocator will process these requests one by one, and call
> "getFreeSlotInformations" once for each request.
> We can optimize it to call "getFreeSlotInformations" once for a bulk (of slot
> requests), instead of once for each slot request.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
