| # |
| # 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. |
| # |
| |
| diff --git a/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala b/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala |
| index c4b619300b5..821a01985d9 100644 |
| --- a/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala |
| +++ b/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala |
| @@ -207,7 +207,9 @@ private[spark] class ExecutorAllocationManager( |
| // If dynamic allocation shuffle tracking or worker decommissioning along with |
| // storage shuffle decommissioning is enabled we have *experimental* support for |
| // decommissioning without a shuffle service. |
| - if (conf.get(config.DYN_ALLOCATION_SHUFFLE_TRACKING_ENABLED) || |
| + if (conf.isRssEnable()) { |
| + logInfo("Dynamic allocation will use remote shuffle service") |
| + } else if (conf.get(config.DYN_ALLOCATION_SHUFFLE_TRACKING_ENABLED) || |
| (decommissionEnabled && |
| conf.get(config.STORAGE_DECOMMISSION_SHUFFLE_BLOCKS_ENABLED))) { |
| logWarning("Dynamic allocation without a shuffle service is an experimental feature.") |
| diff --git a/core/src/main/scala/org/apache/spark/SparkConf.scala b/core/src/main/scala/org/apache/spark/SparkConf.scala |
| index 5f37a1abb19..af4bee1e1bb 100644 |
| --- a/core/src/main/scala/org/apache/spark/SparkConf.scala |
| +++ b/core/src/main/scala/org/apache/spark/SparkConf.scala |
| @@ -580,6 +580,10 @@ class SparkConf(loadDefaults: Boolean) extends Cloneable with Logging with Seria |
| Utils.redact(this, getAll).sorted.map { case (k, v) => k + "=" + v }.mkString("\n") |
| } |
| |
| + /** |
| + * Return true if remote shuffle service is enabled. |
| + */ |
| + def isRssEnable(): Boolean = get("spark.shuffle.manager", "sort").contains("RssShuffleManager") |
| } |
| |
| private[spark] object SparkConf extends Logging { |
| diff --git a/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala |
| index a82d261d545..72e54940ca2 100644 |
| --- a/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala |
| +++ b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala |
| @@ -2231,7 +2231,8 @@ private[spark] class DAGScheduler( |
| // if the cluster manager explicitly tells us that the entire worker was lost, then |
| // we know to unregister shuffle output. (Note that "worker" specifically refers to the process |
| // from a Standalone cluster, where the shuffle service lives in the Worker.) |
| - val fileLost = workerHost.isDefined || !env.blockManager.externalShuffleServiceEnabled |
| + val fileLost = (workerHost.isDefined || !env.blockManager.externalShuffleServiceEnabled) && |
| + !sc.getConf.isRssEnable() |
| removeExecutorAndUnregisterOutputs( |
| execId = execId, |
| fileLost = fileLost, |
| diff --git a/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala b/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala |
| index 3b72103f993..8e03754941e 100644 |
| --- a/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala |
| +++ b/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala |
| @@ -1015,7 +1015,8 @@ private[spark] class TaskSetManager( |
| // and we are not using an external shuffle server which could serve the shuffle outputs. |
| // The reason is the next stage wouldn't be able to fetch the data from this dead executor |
| // so we would need to rerun these tasks on other executors. |
| - if (isShuffleMapTasks && !env.blockManager.externalShuffleServiceEnabled && !isZombie) { |
| + if (isShuffleMapTasks && !env.blockManager.externalShuffleServiceEnabled && |
| + !isZombie && !conf.isRssEnable()) { |
| for ((tid, info) <- taskInfos if info.executorId == execId) { |
| val index = taskInfos(tid).index |
| // We may have a running task whose partition has been marked as successful, |
| diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/ShuffledRowRDD.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/ShuffledRowRDD.scala |
| index 47d61196fe8..98a5381bef4 100644 |
| --- a/sql/core/src/main/scala/org/apache/spark/sql/execution/ShuffledRowRDD.scala |
| +++ b/sql/core/src/main/scala/org/apache/spark/sql/execution/ShuffledRowRDD.scala |
| @@ -174,6 +174,9 @@ class ShuffledRowRDD( |
| } |
| |
| override def getPreferredLocations(partition: Partition): Seq[String] = { |
| + if (conf.isRssEnable()) { |
| + return Nil |
| + } |
| val tracker = SparkEnv.get.mapOutputTracker.asInstanceOf[MapOutputTrackerMaster] |
| partition.asInstanceOf[ShuffledRowRDDPartition].spec match { |
| case CoalescedPartitionSpec(startReducerIndex, endReducerIndex, _) => |
| -- |
| 2.32.0 |
| |