Github user mridulm commented on a diff in the pull request:
https://github.com/apache/spark/pull/16989#discussion_r115410895
--- Diff: core/src/main/scala/org/apache/spark/scheduler/MapStatus.scala ---
@@ -193,8 +206,18 @@ private[spark] object HighlyCompressedMapStatus {
} else {
0
}
+ val hugeBlockSizes = ArrayBuffer[Tuple2[Int, Byte]]()
+ if (numNonEmptyBlocks > 0) {
+ uncompressedSizes.zipWithIndex.foreach {
+ case (size, reduceId) =>
+ if (size > 2 * avgSize) {
--- End diff --
This should be configurable in two respects.
* minimum size before we consider something a large block : if average is
10kb, and some blocks are > 20kb, spilling them to disk would be highly
suboptimal. (Unless I missed that check somewhere else).
* The fraction '2' should also be configurable - some deployments might be
ok with high memory usage (machines provisioned accordingly) while others might
need it to be more aggressive and lower.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
