From a57aadae84aca27e5f02ac0bd64fd0ea34a64b61 Mon Sep 17 00:00:00 2001
From: Takuya UESHIN <ueshin@happy-camper.st>
Date: Thu, 12 May 2016 12:36:18 -0700
Subject: [PATCH] [SPARK-13902][SCHEDULER] Make DAGScheduler not to create
duplicate stage.
## What changes were proposed in this pull request?
`DAGScheduler`sometimes generate incorrect stage graph.
Suppose you have the following DAG:
```
[A] <--(s_A)-- [B] <--(s_B)-- [C] <--(s_C)-- [D]
\ /
<-------------
```
Note: [] means an RDD, () means a shuffle dependency.
Here, RDD `B` has a shuffle dependency on RDD `A`, and RDD `C` has shuffle dependency on both `B` and `A`. The shuffle dependency IDs are numbers in the `DAGScheduler`, but to make the example easier to understand, let's call the shuffled data from `A` shuffle dependency ID `s_A` and the shuffled data from `B` shuffle dependency ID `s_B`.
The `getAncestorShuffleDependencies` method in `DAGScheduler` (incorrectly) does not check for duplicates when it's adding ShuffleDependencies to the parents data structure, so for this DAG, when `getAncestorShuffleDependencies` gets called on `C` (previous of the final RDD), `getAncestorShuffleDependencies` will return `s_A`, `s_B`, `s_A` (`s_A` gets added twice: once when the method "visit"s RDD `C`, and once when the method "visit"s RDD `B`). This is problematic because this line of code: https://github.com/apache/spark/blob/8ef3399/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala#L289 then generates a new shuffle stage for each dependency returned by `getAncestorShuffleDependencies`, resulting in duplicate map stages that compute the map output from RDD `A`.
As a result, `DAGScheduler` generates the following stages and their parents for each shuffle:
| | stage | parents |
|----|----|----|
| s_A | ShuffleMapStage 2 | List() |
| s_B | ShuffleMapStage 1 | List(ShuffleMapStage 0) |
| s_C | ShuffleMapStage 3 | List(ShuffleMapStage 1, ShuffleMapStage 2) |
| - | ResultStage 4 | List(ShuffleMapStage 3) |
The stage for s_A should be `ShuffleMapStage 0`, but the stage for `s_A` is generated twice as `ShuffleMapStage 2` and `ShuffleMapStage 0` is overwritten by `ShuffleMapStage 2`, and the stage `ShuffleMap Stage1` keeps referring the old stage `ShuffleMapStage 0`.
This patch is fixing it.
## How was this patch tested?
I added the sample RDD graph to show the illegal stage graph to `DAGSchedulerSuite`.
Author: Takuya UESHIN <ueshin@happy-camper.st>
Closes #12655 from ueshin/issues/SPARK-13902.
---
.../apache/spark/scheduler/DAGScheduler.scala | 4 +-
.../spark/scheduler/DAGSchedulerSuite.scala | 47 +++++++++++++++++++
2 files changed, 50 insertions(+), 1 deletion(-)
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 4dfd532e93..5291b66366 100644
--- a/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
+++ b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
@@ -286,7 +286,9 @@ class DAGScheduler(
case None =>
// We are going to register ancestor shuffle dependencies
getAncestorShuffleDependencies(shuffleDep.rdd).foreach { dep =>
- shuffleToMapStage(dep.shuffleId) = newOrUsedShuffleStage(dep, firstJobId)
+ if (!shuffleToMapStage.contains(dep.shuffleId)) {
+ shuffleToMapStage(dep.shuffleId) = newOrUsedShuffleStage(dep, firstJobId)
+ }
}
// Then register current shuffleDep
val stage = newOrUsedShuffleStage(shuffleDep, firstJobId)
diff --git a/core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala b/core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala
index e3ed079e4e..088a476086 100644
--- a/core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala
+++ b/core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala
@@ -325,6 +325,53 @@ class DAGSchedulerSuite extends SparkFunSuite with LocalSparkContext with Timeou
assert(sparkListener.stageByOrderOfExecution(0) < sparkListener.stageByOrderOfExecution(1))
}
+ /**
+ * This test ensures that DAGScheduler build stage graph correctly.
+ *
+ * Suppose you have the following DAG:
+ *
+ * [A] <--(s_A)-- [B] <--(s_B)-- [C] <--(s_C)-- [D]
+ * \ /
+ * <-------------
+ *
+ * Here, RDD B has a shuffle dependency on RDD A, and RDD C has shuffle dependency on both
+ * B and A. The shuffle dependency IDs are numbers in the DAGScheduler, but to make the example
+ * easier to understand, let's call the shuffled data from A shuffle dependency ID s_A and the
+ * shuffled data from B shuffle dependency ID s_B.
+ *
+ * Note: [] means an RDD, () means a shuffle dependency.
+ */
+ test("[SPARK-13902] Ensure no duplicate stages are created") {
+ val rddA = new MyRDD(sc, 1, Nil)
+ val shuffleDepA = new ShuffleDependency(rddA, new HashPartitioner(1))
+ val s_A = shuffleDepA.shuffleId
+
+ val rddB = new MyRDD(sc, 1, List(shuffleDepA), tracker = mapOutputTracker)
+ val shuffleDepB = new ShuffleDependency(rddB, new HashPartitioner(1))
+ val s_B = shuffleDepB.shuffleId
+
+ val rddC = new MyRDD(sc, 1, List(shuffleDepA, shuffleDepB), tracker = mapOutputTracker)
+ val shuffleDepC = new ShuffleDependency(rddC, new HashPartitioner(1))
+ val s_C = shuffleDepC.shuffleId
+
+ val rddD = new MyRDD(sc, 1, List(shuffleDepC), tracker = mapOutputTracker)
+
+ submit(rddD, Array(0))
+
+ assert(scheduler.shuffleToMapStage.size === 3)
+ assert(scheduler.activeJobs.size === 1)
+
+ val mapStageA = scheduler.shuffleToMapStage(s_A)
+ val mapStageB = scheduler.shuffleToMapStage(s_B)
+ val mapStageC = scheduler.shuffleToMapStage(s_C)
+ val finalStage = scheduler.activeJobs.head.finalStage
+
+ assert(mapStageA.parents.isEmpty)
+ assert(mapStageB.parents === List(mapStageA))
+ assert(mapStageC.parents === List(mapStageA, mapStageB))
+ assert(finalStage.parents === List(mapStageC))
+ }
+
test("zero split job") {
var numResults = 0
var failureReason: Option[Exception] = None
--
GitLab