[
https://issues.apache.org/jira/browse/SPARK-39983?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17575506#comment-17575506
]
Apache Spark commented on SPARK-39983:
--------------------------------------
User 'alex-balikov' has created a pull request for this issue:
https://github.com/apache/spark/pull/37413
> Should not cache unserialized broadcast relations on the driver
> ---------------------------------------------------------------
>
> Key: SPARK-39983
> URL: https://issues.apache.org/jira/browse/SPARK-39983
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core
> Affects Versions: 3.3.0
> Reporter: Alex Balikov
> Priority: Minor
>
> In TorrentBroadcast.writeBlocks we store the unserialized broadcast object in
> addition to the serialized version of it -
> {code:java}
> private def writeBlocks(value: T): Int = {
> import StorageLevel._
> // Store a copy of the broadcast variable in the driver so that tasks run
> on the driver
> // do not create a duplicate copy of the broadcast variable's value.
> val blockManager = SparkEnv.get.blockManager
> if (!blockManager.putSingle(broadcastId, value, MEMORY_AND_DISK,
> tellMaster = false)) {
> throw new SparkException(s"Failed to store $broadcastId in
> BlockManager")
> }
> {code}
> In case of broadcast relations, these objects can be fairly large (60MB in
> one observed case) and are not strictly necessary on the driver.
> Add the option to not keep the unserialized versions of the objects.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
