sarutak commented on a change in pull request #32786:
URL: https://github.com/apache/spark/pull/32786#discussion_r646731817
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/CollectMetricsExec.scala
##########
@@ -69,7 +69,7 @@ case class CollectMetricsExec(
// - Performance issues due to excessive serialization.
Review comment:
About No. 2 above, `TaskContext` is apparently task local and can be
shared among partitions even if a task can handle multiple partitions. So, I
wonder a simple way is to have a general purpose task local storage in
`TaskContext` and use it for task local accumulator.
No. 3 is too expensive to solve this problem isn't it?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
