xintongsong commented on code in PR #20487:
URL: https://github.com/apache/flink/pull/20487#discussion_r939887668
##########
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/strategy/PipelinedRegionSchedulingStrategy.java:
##########
@@ -53,10 +53,20 @@ public class PipelinedRegionSchedulingStrategy implements
SchedulingStrategy {
private final Map<SchedulingPipelinedRegion, List<ExecutionVertexID>>
regionVerticesSorted =
new IdentityHashMap<>();
+ /** All downstream regions of one schedulingPipelinedRegion. */
+ private final Map<SchedulingPipelinedRegion,
Set<SchedulingPipelinedRegion>>
+ downstreamRegionsOfRegion = new IdentityHashMap<>();
Review Comment:
It should be good enough to store a mapping from the upstream region to the
downstream consumed partition group, as we already have another map helps
mapping the groups to downstream regions.
##########
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/strategy/PipelinedRegionSchedulingStrategy.java:
##########
@@ -53,10 +53,20 @@ public class PipelinedRegionSchedulingStrategy implements
SchedulingStrategy {
private final Map<SchedulingPipelinedRegion, List<ExecutionVertexID>>
regionVerticesSorted =
new IdentityHashMap<>();
+ /** All downstream regions of one schedulingPipelinedRegion. */
+ private final Map<SchedulingPipelinedRegion,
Set<SchedulingPipelinedRegion>>
+ downstreamRegionsOfRegion = new IdentityHashMap<>();
+
+ /** All downstream regions of one ExecutionVertexID. */
+ private final Map<ExecutionVertexID, Set<SchedulingPipelinedRegion>>
downstreamRegionsOfVertex =
+ new IdentityHashMap<>();
Review Comment:
Not sure if we need to store this. Finding out downstream regions of an
upstream vertex already has a `O(1)` complexity.
--
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]
