celeborn/assets/spark-patch/Celeborn_Dynamic_Allocation_spark3_4.patch
Angerszhuuuu 6e35745736
[CELEBORN-753] Rename spark patch file name to make it more clear
### What changes were proposed in this pull request?
Rename spark patch file name to make it more clear

### Why are the changes needed?

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

Closes #1666 from AngersZhuuuu/CELEBORN-753.

Authored-by: Angerszhuuuu <angers.zhu@gmail.com>
Signed-off-by: mingji <fengmingxiao.fmx@alibaba-inc.com>
2023-06-30 11:41:12 +08:00

87 lines
5.0 KiB
Diff

# 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.
Subject: [PATCH] [CORE][SHUFFLE] Support enabling DRA with Apache Celeborn
---
Index: core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala b/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala
--- a/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala (revision efae5362cbeaf1594b18edf594e83b2cf72afce6)
+++ b/core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala (date 1685946134830)
@@ -209,7 +209,7 @@
} else if (decommissionEnabled &&
conf.get(config.STORAGE_DECOMMISSION_SHUFFLE_BLOCKS_ENABLED)) {
logInfo("Shuffle data decommission is enabled without a shuffle service.")
- } else if (!testing) {
+ } else if (!testing && !Utils.isCelebornEnabled(conf)) {
throw new SparkException("Dynamic allocation of executors requires the external " +
"shuffle service. You may enable this through spark.shuffle.service.enabled.")
}
Index: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
--- a/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala (revision efae5362cbeaf1594b18edf594e83b2cf72afce6)
+++ b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala (date 1685946134830)
@@ -2515,7 +2515,7 @@
// 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 = !Utils.isCelebornEnabled(sc.getConf) && (workerHost.isDefined || !env.blockManager.externalShuffleServiceEnabled)
removeExecutorAndUnregisterOutputs(
execId = execId,
fileLost = fileLost,
Index: core/src/main/scala/org/apache/spark/util/Utils.scala
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/core/src/main/scala/org/apache/spark/util/Utils.scala b/core/src/main/scala/org/apache/spark/util/Utils.scala
--- a/core/src/main/scala/org/apache/spark/util/Utils.scala (revision efae5362cbeaf1594b18edf594e83b2cf72afce6)
+++ b/core/src/main/scala/org/apache/spark/util/Utils.scala (date 1685946145650)
@@ -3271,6 +3271,9 @@
files.toSeq
}
+ def isCelebornEnabled(conf: SparkConf): Boolean =
+ conf.get("spark.shuffle.manager", "sort").contains("celeborn")
+
/**
* Return the median number of a long array
*
Index: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala b/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala
--- a/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala (revision efae5362cbeaf1594b18edf594e83b2cf72afce6)
+++ b/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala (date 1685946134830)
@@ -1055,7 +1055,7 @@
// data from this dead executor so we would need to rerun these tasks on other executors.
val maybeShuffleMapOutputLoss = isShuffleMapTasks &&
(reason.isInstanceOf[ExecutorDecommission] || !env.blockManager.externalShuffleServiceEnabled)
- if (maybeShuffleMapOutputLoss && !isZombie) {
+ if (maybeShuffleMapOutputLoss && !isZombie && !Utils.isCelebornEnabled(conf)) {
for ((tid, info) <- taskInfos if info.executorId == execId) {
val index = info.index
lazy val isShuffleMapOutputAvailable = reason match {