[
https://issues.apache.org/jira/browse/SPARK-36419?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17393428#comment-17393428
]
Apache Spark commented on SPARK-36419:
--------------------------------------
User 'akpatnam25' has created a pull request for this issue:
https://github.com/apache/spark/pull/33643
> Move final aggregation in RDD.treeAggregate to executor
> -------------------------------------------------------
>
> Key: SPARK-36419
> URL: https://issues.apache.org/jira/browse/SPARK-36419
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core, Tests
> Affects Versions: 3.0.0
> Reporter: Aravind Patnam
> Priority: Minor
>
> For the last iteration in RDD.treeAggregate, spark relies on RDD.fold as an
> implementation detail.
> RDD.fold pulls all shuffle partitions to the driver to merge the result.
> There are two concerns with this:
> a) Shuffle machinery at executors is much more robust/fault tolerant compared
> to fetching results to driver.
> b) Driver is single point of failure in a spark application. When this
> results in nontrivial increase in memory pressure while pulling partitions to
> driver or increased memory usage as part of computing the aggregated state
> (in user code), it can result in driver failures.
> For treeAggregate, instead of relying on fold for the last iteration, we
> should (optionally) do the computation at a single reducer - and fetch the
> final result to driver.
> Additional cost: one extra stage with a single (resulting) partition.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
