[
https://issues.apache.org/jira/browse/SPARK-3861?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14164418#comment-14164418
]
Apache Spark commented on SPARK-3861:
-------------------------------------
User 'rxin' has created a pull request for this issue:
https://github.com/apache/spark/pull/2722
> Avoid rebuilding hash tables on each partition
> ----------------------------------------------
>
> Key: SPARK-3861
> URL: https://issues.apache.org/jira/browse/SPARK-3861
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Reporter: Reynold Xin
> Assignee: Reynold Xin
>
> BroadcastHashJoin builds a new hash table for each partition. We can build it
> once per node and reuse the hash table.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
