[
https://issues.apache.org/jira/browse/SPARK-58138?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Nikolina Vraneš updated SPARK-58138:
------------------------------------
Description:
Make BinByExec report its output partitioning, so a child partitioning on a
pass-through column is preserved (and one on a scaled DISTRIBUTE column is
dropped). This avoids an unnecessary shuffle when a downstream operator is
clustered on such a column.
Sub-task of SPARK-57133.
was:
Make BinByExec report its output partitioning: preserve a child partitioning
whose keys are pass-through columns, and drop any partitioning on a scaled
DISTRIBUTE column (its output attribute carries a fresh ExprId).
This lets a downstream operator clustered on a pass-through column that the
child already partitions on avoid an unnecessary shuffle.
Sub-task of SPARK-57133.
> Preserve child output partitioning through BIN BY
> -------------------------------------------------
>
> Key: SPARK-58138
> URL: https://issues.apache.org/jira/browse/SPARK-58138
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 5.0.0
> Reporter: Nikolina Vraneš
> Priority: Major
> Labels: pull-request-available
>
> Make BinByExec report its output partitioning, so a child partitioning on a
> pass-through column is preserved (and one on a scaled DISTRIBUTE column is
> dropped). This avoids an unnecessary shuffle when a downstream operator is
> clustered on such a column.
> Sub-task of SPARK-57133.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
