Github user StephanEwen commented on a diff in the pull request:
https://github.com/apache/incubator-flink/pull/254#discussion_r21558983
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/IntermediateResultPartition.java
---
@@ -0,0 +1,266 @@
+/*
+ * 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.io.network.partition;
+
+import com.google.common.base.Optional;
+import
org.apache.flink.runtime.deployment.IntermediateResultPartitionDeploymentDescriptor;
+import org.apache.flink.runtime.executiongraph.ExecutionAttemptID;
+import org.apache.flink.runtime.executiongraph.ExecutionVertex;
+import org.apache.flink.runtime.io.network.NetworkEnvironment;
+import org.apache.flink.runtime.io.network.buffer.Buffer;
+import org.apache.flink.runtime.io.network.buffer.BufferPool;
+import org.apache.flink.runtime.io.network.buffer.BufferPoolOwner;
+import org.apache.flink.runtime.io.network.buffer.BufferProvider;
+import org.apache.flink.runtime.io.network.buffer.LocalBufferPool;
+import
org.apache.flink.runtime.io.network.partition.queue.ConsumableOnceInMemoryOnlyPartitionQueue;
+import
org.apache.flink.runtime.io.network.partition.queue.IllegalQueueIteratorRequestException;
+import
org.apache.flink.runtime.io.network.partition.queue.IntermediateResultPartitionQueue;
+import
org.apache.flink.runtime.io.network.partition.queue.IntermediateResultPartitionQueueIterator;
+import org.apache.flink.runtime.jobgraph.IntermediateResultPartitionID;
+import org.apache.flink.runtime.jobgraph.IntermediateResultType;
+import org.apache.flink.runtime.jobgraph.JobID;
+import org.apache.flink.runtime.protocols.ConsumerNotificationProtocol;
+import org.apache.flink.runtime.protocols.ConsumerNotificationResult;
+
+import java.io.IOException;
+
+import static com.google.common.base.Preconditions.checkNotNull;
+import static com.google.common.base.Preconditions.checkState;
+
+public class IntermediateResultPartition implements BufferPoolOwner {
+
+ /**
+ * Note: This index needs to correspond to the index of the partition in
+ * {@link ExecutionVertex#resultPartitions}, which might be a little
+ * fragile as the data availability notifications use it.
+ */
+ private final int partitionIndex;
+
+ private final JobID jobId;
+
+ private final ExecutionAttemptID executionId;
+
+ private final IntermediateResultPartitionID partitionId;
+
+ private final IntermediateResultType partitionType;
+
+ private final BufferPool bufferPool;
+
+ private final ConsumerNotificationProtocol notificationProtocol;
+
+ private final IntermediateResultPartitionQueue[] queues;
+
+ private boolean hasNotifiedConsumers;
+
+ private boolean isReleased;
+
+ private boolean isFinished;
+
+ public IntermediateResultPartition(
+ int partitionIndex,
+ JobID jobId,
+ ExecutionAttemptID executionId,
+ NetworkEnvironment networkEnvironment,
+ IntermediateResultPartitionDeploymentDescriptor irpdd)
throws IOException {
+
+ this.partitionIndex = partitionIndex;
+ this.jobId = checkNotNull(jobId);
+ this.executionId = checkNotNull(executionId);
+
+ this.partitionId = checkNotNull(irpdd.getPartitionId());
+ this.queues = new
IntermediateResultPartitionQueue[irpdd.getNumberOfQueues()];
+ this.partitionType = irpdd.getPartitionType();
+
+ this.bufferPool =
networkEnvironment.getNetworkBufferPool().createBufferPool(irpdd.getNumberOfQueues(),
false);
+ this.notificationProtocol =
networkEnvironment.getNotificationProtocol();
+
+ // TODO The queues need to be created depending on the result
type
+ for (int i = 0; i < queues.length; i++) {
+ queues[i] = new
ConsumableOnceInMemoryOnlyPartitionQueue();
+ }
+ }
+
+ //
------------------------------------------------------------------------
+ // Properties
+ //
------------------------------------------------------------------------
+
+ public IntermediateResultPartitionID getPartitionId() {
+ return partitionId;
+ }
+
+ public JobID getJobId() {
+ return jobId;
+ }
+
+ public int getNumberOfQueues() {
+ return queues.length;
+ }
+
+ public BufferProvider getBufferProvider() {
+ return bufferPool;
+ }
+
+ public boolean isFinished() {
+ return isFinished;
+ }
+
+ //
------------------------------------------------------------------------
+ // Produce
+ //
------------------------------------------------------------------------
+
+ public void add(Buffer buffer, int targetQueue) throws IOException {
+ checkInProducePhase();
+
+ queues[targetQueue].add(buffer);
+
+ maybeNotifyConsumers(partitionType.isPipelined());
+ }
+
+ public void addToAllPartitions(Buffer buffer) throws IOException {
--- End diff --
I think this does not work for buffers that contain data frames. The
buffers are queued without incrementing the reference counter, so they will be
released as soon as the first queue discards them.
In general, I am skeptical if it is a good idea to have the same buffer
referenced from various queues. If we start spilling one of the queues, we may
not get any memory back, because references are held from other queues.
I think for the time being, broadcasts should really write disjoint
results. As a next step, we should create a single queue that is requested by
all consumers. That way, we need not duplicate data in memory, spilled, and
ease the reasoning about how much we get back when spilling data.
---
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.
---
