[
https://issues.apache.org/jira/browse/SPARK-13873?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Davies Liu resolved SPARK-13873.
--------------------------------
Resolution: Fixed
Fix Version/s: 2.0.0
Issue resolved by pull request 11740
[https://github.com/apache/spark/pull/11740]
> Avoid the copy in whole stage codegen when there is no joins
> ------------------------------------------------------------
>
> Key: SPARK-13873
> URL: https://issues.apache.org/jira/browse/SPARK-13873
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Davies Liu
> Assignee: Davies Liu
> Fix For: 2.0.0
>
>
> When we generate code for join, we copy the output row, because there could
> be multiple output row from single input row.
> We could avoid this copy when there is no join, or the join will not generate
> multiple output rows from single input row.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
