Ngone51 commented on a change in pull request #29452:
URL: https://github.com/apache/spark/pull/29452#discussion_r472663605
##########
File path: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala
##########
@@ -1051,15 +1049,19 @@ private[spark] class TaskSetManager(
logDebug("Task length threshold for speculation: " + threshold)
for (tid <- runningTasksSet) {
var speculated = checkAndSubmitSpeculatableTask(tid, time, threshold)
- if (!speculated && tidToExecutorKillTimeMapping.contains(tid)) {
- // Check whether this task will finish before the exectorKillTime
assuming
- // it will take medianDuration overall. If this task cannot finish
within
- // executorKillInterval, then this task is a candidate for
speculation
- val taskEndTimeBasedOnMedianDuration = taskInfos(tid).launchTime +
medianDuration
- val canExceedDeadline = tidToExecutorKillTimeMapping(tid) <
- taskEndTimeBasedOnMedianDuration
- if (canExceedDeadline) {
- speculated = checkAndSubmitSpeculatableTask(tid, time, 0)
+ if (!speculated && executorDecommissionKillInterval.nonEmpty) {
+ val taskInfo = taskInfos(tid)
+ val decomState =
sched.getExecutorDecommissionState(taskInfo.executorId)
+ if (decomState.nonEmpty) {
+ // Check whether this task will finish before the exectorKillTime
assuming
Review comment:
We don't have `exectorKillTime` anymore after removing
`tidToExecutorKillTimeMapping`. Shall we reword the comment a little bit?
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
