[
https://issues.apache.org/jira/browse/FLINK-22677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17375218#comment-17375218
]
Zhu Zhu commented on FLINK-22677:
---------------------------------
One thing need to mention is that I did not change {{AdaptiveScheduler}} to
support async partition registration, because otherwise we will need to
introduce async callback process to {{AdaptiveScheduler}}, which was
intentionally avoided. Given that customized {{ShuffleMaster}} is mainly for
batch jobs on external shuffle purpose and reactive mode is for streaming jobs
only, I think for reactive mode we can keep the assumption that the
{{ShuffleMaster}} will complete partition registration immediately.
WDYT? [~trohrmann]
> Scheduler should invoke ShuffleMaster#registerPartitionWithProducer by a real
> asynchronous fashion
> --------------------------------------------------------------------------------------------------
>
> Key: FLINK-22677
> URL: https://issues.apache.org/jira/browse/FLINK-22677
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Affects Versions: 1.14.0
> Reporter: Jin Xing
> Assignee: Zhu Zhu
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.14.0
>
>
> Current scheduler enforces a synchronous registration though the API of
> ShuffleMaster#registerPartitionWithProducer returns a CompletableFuture. In
> scenario of remote shuffle service, the talk between ShuffleMaster and remote
> cluster tends to be expensive. A synchronous registration risks to block main
> thread potentially and might cause negative side effects like heartbeat
> timeout. Additionally, expensive synchronous invokes to remote could
> bottleneck the throughput for applying shuffle resource, especially for batch
> jobs with complicated DAGs;
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
