[GitHub] [flink] azagrebin commented on a change in pull request #8687: [FLINK-12612][coordination] Track stored partition on the TaskExecutor

[GitBox <gi...@apache.org>]

azagrebin commented on a change in pull request #8687: [FLINK-12612][coordination] Track stored partition on the TaskExecutor 
URL: https://github.com/apache/flink/pull/8687#discussion_r296750470
 
 

 ##########
 File path: flink-runtime/src/main/java/org/apache/flink/runtime/taskexecutor/partition/JobAwareShuffleEnvironmentImpl.java
 ##########
 @@ -0,0 +1,188 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.runtime.taskexecutor.partition;
+
+import org.apache.flink.api.common.JobID;
+import org.apache.flink.metrics.MetricGroup;
+import org.apache.flink.runtime.deployment.InputGateDeploymentDescriptor;
+import org.apache.flink.runtime.deployment.ResultPartitionDeploymentDescriptor;
+import org.apache.flink.runtime.executiongraph.ExecutionAttemptID;
+import org.apache.flink.runtime.executiongraph.PartitionInfo;
+import org.apache.flink.runtime.io.network.api.writer.ResultPartitionWriter;
+import org.apache.flink.runtime.io.network.partition.PartitionProducerStateProvider;
+import org.apache.flink.runtime.io.network.partition.ResultPartitionID;
+import org.apache.flink.runtime.io.network.partition.consumer.InputGate;
+import org.apache.flink.runtime.shuffle.ShuffleDescriptor;
+import org.apache.flink.runtime.shuffle.ShuffleEnvironment;
+import org.apache.flink.runtime.shuffle.ShuffleIOOwnerContext;
+import org.apache.flink.util.Preconditions;
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Collection;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+/**
+ * Wraps a {@link ShuffleEnvironment} to allow tracking of partitions per job.
+ */
+public class JobAwareShuffleEnvironmentImpl<G extends InputGate> implements JobAwareShuffleEnvironment<ResultPartitionWriter, G> {
+
+	private final ShuffleEnvironment<?, G> backingShuffleEnvironment;
+
+	/**
+	 * Releasing in-progress partitions while tasks are still running may lead to odd behaviors, hence we distinguish
+	 * between in-progress and finished partitions so that we can exclusively release finished partitions.
+	 */
+	private final PartitionTable inProgressPartitionTable = new PartitionTable();
+	private final PartitionTable finishedPartitionTable = new PartitionTable();
+
+	public JobAwareShuffleEnvironmentImpl(ShuffleEnvironment<?, G> backingShuffleEnvironment) {
+		this.backingShuffleEnvironment = Preconditions.checkNotNull(backingShuffleEnvironment);
+	}
+
+	@Override
+	public boolean hasPartitionsOccupyingLocalResources(JobID jobId) {
+		return inProgressPartitionTable.hasTrackedPartitions(jobId) || finishedPartitionTable.hasTrackedPartitions(jobId);
+	}
+
+	@Override
+	public void releaseFinishedPartitions(JobID jobId, Collection<ResultPartitionID> resultPartitionIds) {
+		// maybe double-check that all partitions are finished?
 
 Review comment:
   or for each task of job, schedule its partitions release after the termination future (in worst case double release or extra task state check). we can actually change `resultPartitionIds` to task `executionAttemptId` in the RPC API because this is how we use it and then immediately obtain task with its partitions locally. 

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services