[
https://issues.apache.org/jira/browse/SPARK-35853?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
XiDuo You updated SPARK-35853:
------------------------------
Description:
In some queries, we might repartition by some columns with a large partition
number manually to make parallelism big enough. However if its output
partitioning satisfies some other node (e.g. join/aggregate), this shuffle can
not be optimized by AQE due to the shuffle origin.
was:
In some queries, we might repartition by some column with a large partition
number manually to make parallelism big enough. However if the output
partitioning satisfies some other node (e.g. join/aggregate), this shuffle can
not be optimized by AQE due to the shuffle origin.
> Remark the shuffle origin to ENSURE_REQUIREMENTS as far as possible
> -------------------------------------------------------------------
>
> Key: SPARK-35853
> URL: https://issues.apache.org/jira/browse/SPARK-35853
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 3.2.0
> Reporter: XiDuo You
> Priority: Major
>
> In some queries, we might repartition by some columns with a large partition
> number manually to make parallelism big enough. However if its output
> partitioning satisfies some other node (e.g. join/aggregate), this shuffle
> can not be optimized by AQE due to the shuffle origin.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
