You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Josh Rosen (JIRA)" <ji...@apache.org> on 2019/07/09 15:10:00 UTC
[jira] [Reopened] (SPARK-11309) Clean up hacky use of MemoryManager
inside of HashedRelation
[ https://issues.apache.org/jira/browse/SPARK-11309?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Josh Rosen reopened SPARK-11309:
--------------------------------
I'm re-opening this issue because this problem is still relevant.
There's some discussion of this issue ongoing at GitHub: https://github.com/apache/spark/pull/25084#issuecomment-509681776
> Clean up hacky use of MemoryManager inside of HashedRelation
> ------------------------------------------------------------
>
> Key: SPARK-11309
> URL: https://issues.apache.org/jira/browse/SPARK-11309
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Josh Rosen
> Priority: Major
> Labels: bulk-closed
>
> In HashedRelation, there's a hacky creation of a new MemoryManager in order to handle broadcasting of BytesToBytesMap: https://github.com/apache/spark/blob/85e654c5ec87e666a8845bfd77185c1ea57b268a/sql/core/src/main/scala/org/apache/spark/sql/execution/joins/HashedRelation.scala#L323
> Something similar to this has existed for a while, but the code recently became much messier as an indirect consequence of my memory manager consolidation patch. We should see about cleaning this up and removing the hack.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org