Github user kayousterhout commented on a diff in the pull request:
https://github.com/apache/spark/pull/6352#discussion_r31064802
--- Diff:
core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala ---
@@ -342,6 +342,35 @@ class DAGSchedulerSuite
assert(locs === Seq(Seq("hostA", "hostB"), Seq("hostB", "hostC"),
Seq("hostC", "hostD")))
}
+ /**
+ * +---+ shuffle +---+ +---+
+ * | A |<--------| B |<---| C |<--+
+ * +---+ +---+ +---+ | +---+
+ * +--| E |
+ * +---+ | +---+
+ * | D |<--+
+ * +---+
+ * Here, E has one-to-one dependencies on C and D. C is derived from A
by performing a shuffle
+ * and then a map. If we're trying to determine which ancestor stages
need to be computed in
+ * order to compute E, we need to figure out whether the shuffle A -> B
should be performed.
+ * If the RDD C, which has only one ancestor via a narrow dependency, is
cached, then we won't
+ * need to compute A, even if it has some unavailable output partitions.
The same goes for B:
+ * if B is 100% cached, then we can avoid the shuffle on A.
+ */
+ test("SPARK-7826: regression test for getMissingParentStages") {
+ val rddA = new MyRDD(sc, 1, Nil)
+ val rddB = new MyRDD(sc, 1, List(new ShuffleDependency(rddA, null)))
+ val rddC = new MyRDD(sc, 1, List(new OneToOneDependency(rddB))).cache()
+ val rddD = new MyRDD(sc, 1, Nil)
+ val rddE = new MyRDD(sc, 1,
+ List(new OneToOneDependency(rddC), new OneToOneDependency(rddD)))
+ cacheLocations(rddC.id -> 0) =
+ Seq(makeBlockManagerId("hostA"), makeBlockManagerId("hostB"))
+ val jobId = submit(rddE, Array(0))
+ val finalStage = scheduler.jobIdToActiveJob(jobId).finalStage
+ assert(scheduler.getMissingParentStages(finalStage).size === 0)
--- End diff --
It looks like what happens is that the call to submit() causes the first
set of missing parent stages to be submitted, so at that point, stage D is
submitted. Can you add a comment describing this?
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
