huwh commented on code in PR #22674:
URL: https://github.com/apache/flink/pull/22674#discussion_r1246229775
##########
flink-runtime/src/main/java/org/apache/flink/runtime/deployment/CachedShuffleDescriptors.java:
##########
@@ -54,10 +54,14 @@ public class CachedShuffleDescriptors {
/** Stores the mapping of resultPartitionId to index subscripts in
consumed partition group. */
private final Map<IntermediateResultPartitionID, Integer>
resultPartitionIdToIndex;
+ /** The number of consumers for {@link ConsumedPartitionGroup}. */
+ private final int numConsumers;
+
public CachedShuffleDescriptors(
ConsumedPartitionGroup consumedPartitionGroup,
ShuffleDescriptorAndIndex[] shuffleDescriptors) {
Review Comment:
There could be some incremental serialization of ShuffleDescriptors, and may
cause performance regression if there are few shuffle descriptors
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
