jiangxb1987 commented on a change in pull request #24485: [SPARK-27590][CORE]
do not consider skipped tasks when scheduling speculative tasks
URL: https://github.com/apache/spark/pull/24485#discussion_r280194979
##########
File path: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala
##########
@@ -1038,7 +1035,11 @@ private[spark] class TaskSetManager(
val minFinishedForSpeculation = (speculationQuantile *
numTasks).floor.toInt
logDebug("Checking for speculative tasks: minFinished = " +
minFinishedForSpeculation)
- if (tasksSuccessful >= minFinishedForSpeculation && tasksSuccessful > 0) {
+ // It's possible that a task is marked as completed by the scheduler, then
the size of
+ // `successfulTaskDurations` may not equal to `tasksSuccessful`. Here we
should only count the
+ // tasks that are submitted by this `TaskSetManager` and are completed
successfully.
+ val numSuccessfulTasks = successfulTaskDurations.size()
+ if (numSuccessfulTasks >= minFinishedForSpeculation && numSuccessfulTasks
> 0) {
Review comment:
+1
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
