[
https://issues.apache.org/jira/browse/SPARK-4410?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14212729#comment-14212729
]
Apache Spark commented on SPARK-4410:
-------------------------------------
User 'marmbrus' has created a pull request for this issue:
https://github.com/apache/spark/pull/3268
> Support for external sort
> -------------------------
>
> Key: SPARK-4410
> URL: https://issues.apache.org/jira/browse/SPARK-4410
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Michael Armbrust
> Assignee: Michael Armbrust
> Priority: Critical
>
> When any given key is of too high cardinality the current sorting code can
> tip over (since it loads a whole partition into memory). I propose we add
> optional support for using sparks built in external sorting mechanism. It
> can be off by default, but if we determine this code path does not regress
> performance we can turn it on by default in the future.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
