zhuzhurk commented on a change in pull request #15310:
URL: https://github.com/apache/flink/pull/15310#discussion_r603166896
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/strategy/SchedulingPipelinedRegion.java
##########
@@ -23,10 +23,19 @@
import org.apache.flink.runtime.jobgraph.IntermediateResultPartitionID;
import org.apache.flink.runtime.topology.PipelinedRegion;
+import java.util.List;
+
/** Pipelined region on execution level, i.e., {@link ExecutionGraph} level. */
public interface SchedulingPipelinedRegion
extends PipelinedRegion<
ExecutionVertexID,
IntermediateResultPartitionID,
SchedulingExecutionVertex,
- SchedulingResultPartition> {}
+ SchedulingResultPartition> {
+ /**
+ * Get all blocking {@link ConsumedPartitionGroup}s.
+ *
+ * @return list of {@link ConsumedPartitionGroup}s
+ */
+ List<ConsumedPartitionGroup> getAllBlockingConsumedPartitionGroups();
Review comment:
>> I'm wondering should we change
SchedulingExecutionVertex#getConsumedPartitionGroups and
SchedulingResultPartition#getConsumerVertexGroups into Iterable
As long as it is not causing problems, I think it's fine to keep them as is.
`getAllBlockingConsumedPartitionGroups ()` should not return a list because it
is just complicating things (one have to convert a set to a list).
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/strategy/SchedulingPipelinedRegion.java
##########
@@ -23,10 +23,19 @@
import org.apache.flink.runtime.jobgraph.IntermediateResultPartitionID;
import org.apache.flink.runtime.topology.PipelinedRegion;
+import java.util.List;
+
/** Pipelined region on execution level, i.e., {@link ExecutionGraph} level. */
public interface SchedulingPipelinedRegion
extends PipelinedRegion<
ExecutionVertexID,
IntermediateResultPartitionID,
SchedulingExecutionVertex,
- SchedulingResultPartition> {}
+ SchedulingResultPartition> {
+ /**
+ * Get all blocking {@link ConsumedPartitionGroup}s.
+ *
+ * @return list of {@link ConsumedPartitionGroup}s
+ */
+ List<ConsumedPartitionGroup> getAllBlockingConsumedPartitionGroups();
Review comment:
>> I'm wondering should we change
SchedulingExecutionVertex#getConsumedPartitionGroups and
SchedulingResultPartition#getConsumerVertexGroups into Iterable
As long as it is not causing problems, I think it's fine to keep them as is.
`getAllBlockingConsumedPartitionGroups ()` should not return a list because it
is just complicating things (one have to convert a set to a list).
--
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]
