[
https://issues.apache.org/jira/browse/SPARK-49386?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18030135#comment-18030135
]
Dongjoon Hyun commented on SPARK-49386:
---------------------------------------
I added the label, `releasenotes`, not to forget this new feature in the
release notes. Thank you, [~dzcxzl] and [~attilapiros] .
> Add memory based thresholds for shuffle spill
> ---------------------------------------------
>
> Key: SPARK-49386
> URL: https://issues.apache.org/jira/browse/SPARK-49386
> Project: Spark
> Issue Type: Sub-task
> Components: Spark Core, SQL
> Affects Versions: 4.1.0
> Reporter: dzcxzl
> Assignee: dzcxzl
> Priority: Major
> Labels: releasenotes
> Fix For: 4.1.0
>
>
> We can only determine the number of spills by configuring
> {{{}spark.shuffle.spill.numElementsForceSpillThreshold{}}}. In some
> scenarios, the size of a row may be very large in the memory.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
