You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Stephan Ewen (Jira)" <ji...@apache.org> on 2020/11/04 14:25:00 UTC

[jira] [Commented] (FLINK-16848) Improve the performance of the RecordsWithSplitIds queue.

    [ https://issues.apache.org/jira/browse/FLINK-16848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17226075#comment-17226075 ] 

Stephan Ewen commented on FLINK-16848:
--------------------------------------

[~becket_qin] This has already been solved in another ticket / refactoring, right?

If yes, could you close this ticket?

> Improve the performance of the RecordsWithSplitIds queue.
> ---------------------------------------------------------
>
>                 Key: FLINK-16848
>                 URL: https://issues.apache.org/jira/browse/FLINK-16848
>             Project: Flink
>          Issue Type: Sub-task
>          Components: API / DataStream
>            Reporter: Jiangjie Qin
>            Assignee: Jiangjie Qin
>            Priority: Major
>
> Currently there is just one queue in the SourceReader which may become a performance bottleneck due to lock contention when there are many SplitFetcher threads. This ticket is supposed to improve the performance.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)