You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Junrui Lee (Jira)" <ji...@apache.org> on 2025/02/26 06:49:00 UTC

[jira] [Closed] (FLINK-37382) Early optimization for adaptive join when one input is below broadcast threshold

     [ https://issues.apache.org/jira/browse/FLINK-37382?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Junrui Lee closed FLINK-37382.
------------------------------
    Fix Version/s: 2.0.0
                   2.0.1
       Resolution: Done

master
69559fb5d231d704633fed807773cd1853601862

release-2.0
ae5fc93a69369409f4391538653a535147eb4f24

> Early optimization for adaptive join when one input is below broadcast threshold
> --------------------------------------------------------------------------------
>
>                 Key: FLINK-37382
>                 URL: https://issues.apache.org/jira/browse/FLINK-37382
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Runtime / Coordination
>    Affects Versions: 2.0.0
>            Reporter: xingbe
>            Assignee: xingbe
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 2.0.0, 2.0.1
>
>
> When the large table side has not yet initialized, if the small table side finishes early and the output data volume is smaller than the broadcast threshold, the adaptive join operator can be optimized to a broadcast hash join in advance. This can help reduce network shuffle overhead.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)