[SPARK-24386][SS] coalesce(1) aggregates in continuous processing #21560
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,51 @@ | ||
| /* | ||
| * 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.spark.sql.execution.streaming.continuous | ||
|
|
||
| import java.util.UUID | ||
|
|
||
| import org.apache.spark.{HashPartitioner, SparkEnv} | ||
| import org.apache.spark.rdd.RDD | ||
| import org.apache.spark.sql.catalyst.InternalRow | ||
| import org.apache.spark.sql.catalyst.expressions.{Attribute, UnsafeRow} | ||
| import org.apache.spark.sql.catalyst.plans.physical.{Partitioning, SinglePartition} | ||
| import org.apache.spark.sql.execution.SparkPlan | ||
| import org.apache.spark.sql.execution.streaming.continuous.shuffle.{ContinuousShuffleReadPartition, ContinuousShuffleReadRDD} | ||
|
|
||
| /** | ||
| * Physical plan for coalescing a continuous processing plan. | ||
| * | ||
| * Currently, only coalesces to a single partition are supported. `numPartitions` must be 1. | ||
| */ | ||
| case class ContinuousCoalesceExec(numPartitions: Int, child: SparkPlan) extends SparkPlan { | ||
| override def output: Seq[Attribute] = child.output | ||
|
|
||
| override def children: Seq[SparkPlan] = child :: Nil | ||
|
|
||
| override def outputPartitioning: Partitioning = SinglePartition | ||
|
|
||
| override def doExecute(): RDD[InternalRow] = { | ||
| assert(numPartitions == 1) | ||
| new ContinuousCoalesceRDD( | ||
| sparkContext, | ||
| numPartitions, | ||
| conf.continuousStreamingExecutorQueueSize, | ||
| sparkContext.getLocalProperty(ContinuousExecution.EPOCH_INTERVAL_KEY).toLong, | ||
| child.execute()) | ||
| } | ||
| } |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,136 @@ | ||
| /* | ||
| * 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.spark.sql.execution.streaming.continuous | ||
|
|
||
| import java.util.UUID | ||
|
|
||
| import org.apache.spark._ | ||
| import org.apache.spark.rdd.RDD | ||
| import org.apache.spark.sql.catalyst.InternalRow | ||
| import org.apache.spark.sql.catalyst.expressions.UnsafeRow | ||
| import org.apache.spark.sql.execution.streaming.continuous.shuffle._ | ||
| import org.apache.spark.util.ThreadUtils | ||
|
|
||
| case class ContinuousCoalesceRDDPartition( | ||
| index: Int, | ||
| endpointName: String, | ||
| queueSize: Int, | ||
| numShuffleWriters: Int, | ||
| epochIntervalMs: Long) | ||
| extends Partition { | ||
| // Initialized only on the executor, and only once even as we call compute() multiple times. | ||
| lazy val (reader: ContinuousShuffleReader, endpoint) = { | ||
| val env = SparkEnv.get.rpcEnv | ||
| val receiver = new RPCContinuousShuffleReader( | ||
| queueSize, numShuffleWriters, epochIntervalMs, env) | ||
| val endpoint = env.setupEndpoint(endpointName, receiver) | ||
|
|
||
| TaskContext.get().addTaskCompletionListener { ctx => | ||
| env.stop(endpoint) | ||
| } | ||
| (receiver, endpoint) | ||
| } | ||
| // This flag will be flipped on the executors to indicate that the threads processing | ||
| // partitions of the write-side RDD have been started. These will run indefinitely | ||
| // asynchronously as epochs of the coalesce RDD complete on the read side. | ||
| private[continuous] var writersInitialized: Boolean = false | ||
| } | ||
|
|
||
| /** | ||
| * RDD for continuous coalescing. Asynchronously writes all partitions of `prev` into a local | ||
| * continuous shuffle, and then reads them in the task thread using `reader`. | ||
| */ | ||
| class ContinuousCoalesceRDD( | ||
| context: SparkContext, | ||
| numPartitions: Int, | ||
| readerQueueSize: Int, | ||
| epochIntervalMs: Long, | ||
| prev: RDD[InternalRow]) | ||
| extends RDD[InternalRow](context, Nil) { | ||
|
|
||
| // When we support more than 1 target partition, we'll need to figure out how to pass in the | ||
| // required partitioner. | ||
| private val outputPartitioner = new HashPartitioner(1) | ||
|
|
||
| private val readerEndpointNames = (0 until numPartitions).map { i => | ||
| s"ContinuousCoalesceRDD-part$i-${UUID.randomUUID()}" | ||
| } | ||
|
|
||
| override def getPartitions: Array[Partition] = { | ||
| (0 until numPartitions).map { partIndex => | ||
| ContinuousCoalesceRDDPartition( | ||
| partIndex, | ||
| readerEndpointNames(partIndex), | ||
| readerQueueSize, | ||
| prev.getNumPartitions, | ||
| epochIntervalMs) | ||
| }.toArray | ||
| } | ||
|
|
||
| private lazy val threadPool = ThreadUtils.newDaemonFixedThreadPool( | ||
| prev.getNumPartitions, | ||
| this.name) | ||
|
|
||
| override def compute(split: Partition, context: TaskContext): Iterator[InternalRow] = { | ||
| val part = split.asInstanceOf[ContinuousCoalesceRDDPartition] | ||
|
|
||
| if (!part.writersInitialized) { | ||
| val rpcEnv = SparkEnv.get.rpcEnv | ||
|
|
||
| // trigger lazy initialization | ||
| part.endpoint | ||
| val endpointRefs = readerEndpointNames.map { endpointName => | ||
| rpcEnv.setupEndpointRef(rpcEnv.address, endpointName) | ||
| } | ||
|
|
||
| val runnables = prev.partitions.map { prevSplit => | ||
| new Runnable() { | ||
| override def run(): Unit = { | ||
| TaskContext.setTaskContext(context) | ||
|
|
||
| val writer: ContinuousShuffleWriter = new RPCContinuousShuffleWriter( | ||
| prevSplit.index, outputPartitioner, endpointRefs.toArray) | ||
|
|
||
| EpochTracker.initializeCurrentEpoch( | ||
| context.getLocalProperty(ContinuousExecution.START_EPOCH_KEY).toLong) | ||
| while (!context.isInterrupted() && !context.isCompleted()) { | ||
| writer.write(prev.compute(prevSplit, context).asInstanceOf[Iterator[UnsafeRow]]) | ||
| // Note that current epoch is a non-inheritable thread local, so each writer thread | ||
| // can properly increment its own epoch without affecting the main task thread. | ||
| EpochTracker.incrementCurrentEpoch() | ||
| } | ||
| } | ||
| } | ||
| } | ||
|
|
||
| context.addTaskCompletionListener { ctx => | ||
| threadPool.shutdownNow() | ||
| } | ||
|
|
||
| part.writersInitialized = true | ||
|
|
||
| runnables.foreach(threadPool.execute) | ||
| } | ||
|
|
||
| part.reader.read() | ||
| } | ||
|
|
||
| override def clearDependencies(): Unit = { | ||
| throw new IllegalStateException("Continuous RDDs cannot be checkpointed") | ||
| } | ||
| } |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What if we have multiple repartitions which one meets the case and others are not? I'm not sure we are restricting repartition operations to be only once.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think there's any particular reason we need to. There's no reason we couldn't execute multiple repartitions if the optimizer isn't smart enough to combine them.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh wait, I see what you mean. Repartition(5, ...) would never be matched by this rule, since it only applies to Aggregate.