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

Posted to issues@spark.apache.org by "Apache Spark (Jira)" <ji...@apache.org> on 2021/11/23 11:55:00 UTC

[jira] [Commented] (SPARK-11792) SizeEstimator cannot provide a good size estimation of UnsafeHashedRelations

    [ https://issues.apache.org/jira/browse/SPARK-11792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17447961#comment-17447961 ] 

Apache Spark commented on SPARK-11792:
--------------------------------------

User 'pan3793' has created a pull request for this issue:
https://github.com/apache/spark/pull/34692

> SizeEstimator cannot provide a good size estimation of UnsafeHashedRelations
> ----------------------------------------------------------------------------
>
>                 Key: SPARK-11792
>                 URL: https://issues.apache.org/jira/browse/SPARK-11792
>             Project: Spark
>          Issue Type: Bug
>          Components: SQL
>            Reporter: Yin Huai
>            Assignee: Yin Huai
>            Priority: Blocker
>             Fix For: 1.6.0
>
>
> At the executor side, SizeEstimator sees that an UnsafeHashedRelation has a huge graph and produce a significant larger estimated size (compared with the size of data stored in the UnsafeHashedRelation). So, the memory manger will "think" we run out of memory after several broad cast joins.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org