[jira] [Resolved] (SPARK-32384) repartitionAndSortWithinPartitions avoid shuffle with same partitioner

["wuyi (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SPARK-32384?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

wuyi resolved SPARK-32384.
--------------------------
    Fix Version/s: 3.2.0
         Assignee: zhengruifeng
       Resolution: Fixed

Resolved by https://github.com/apache/spark/pull/31480

> repartitionAndSortWithinPartitions avoid shuffle with same partitioner
> ----------------------------------------------------------------------
>
>                 Key: SPARK-32384
>                 URL: https://issues.apache.org/jira/browse/SPARK-32384
>             Project: Spark
>          Issue Type: Improvement
>          Components: Spark Core
>    Affects Versions: 3.1.0
>            Reporter: zhengruifeng
>            Assignee: zhengruifeng
>            Priority: Minor
>             Fix For: 3.2.0
>
>
> In {{combineByKeyWithClassTag}}, there is a check so that if the partitioner is the same as the one of the RDD:
> {code:java}
> if (self.partitioner == Some(partitioner)) {
>   self.mapPartitions(iter => {
>     val context = TaskContext.get()
>     new InterruptibleIterator(context, aggregator.combineValuesByKey(iter, context))
>   }, preservesPartitioning = true)
> } else {
>   new ShuffledRDD[K, V, C](self, partitioner)
>     .setSerializer(serializer)
>     .setAggregator(aggregator)
>     .setMapSideCombine(mapSideCombine)
> }
>  {code}
>  
> In {{repartitionAndSortWithinPartitions}}, this shuffle can also be skipped in this case.
>  
>  
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org