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

Posted to issues@flink.apache.org by "Runkang He (Jira)" <ji...@apache.org> on 2022/12/01 02:23:00 UTC

[jira] [Comment Edited] (FLINK-30242) Push localHashAggregate pass the union node

    [ https://issues.apache.org/jira/browse/FLINK-30242?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17641107#comment-17641107 ] 

Runkang He edited comment on FLINK-30242 at 12/1/22 2:22 AM:
-------------------------------------------------------------

in flink olap scenario, we have already done this work and have a presentation in FFA 2022 a few days ago, we would like to consider to bring back it to the community.


was (Author: JIRAUSER280488):
We have already done this work, and have a presentation in FFA 2022 a few days ago, we will try to bring back it to the community.

> Push localHashAggregate pass the union node
> -------------------------------------------
>
>                 Key: FLINK-30242
>                 URL: https://issues.apache.org/jira/browse/FLINK-30242
>             Project: Flink
>          Issue Type: Improvement
>          Components: Table SQL / Planner
>            Reporter: Aitozi
>            Priority: Major
>
> The local hash aggregate after union will have an extra shuffle stage. We can swap it with the union node so the local hash aggregate can chain with the mapper stage saving the unnecessary shuffle, especially in the batch job.



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