pgandhi999 commented on a change in pull request #23677:
[SPARK-26755][SCHEDULER] : Optimize Spark Scheduler to dequeue speculative
tasks…
URL: https://github.com/apache/spark/pull/23677#discussion_r301754350
##########
File path: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala
##########
@@ -238,84 +228,41 @@ private[spark] class TaskSetManager(
index: Int,
resolveRacks: Boolean = true,
speculative: Boolean = false): Unit = {
- if (speculative) {
- for (loc <- tasks(index).preferredLocations) {
- loc match {
- case e: ExecutorCacheTaskLocation =>
- pendingSpeculatableTasks.forExecutor.getOrElseUpdate(
- e.executorId, new ArrayBuffer) += index
- case _ =>
- }
- pendingSpeculatableTasks.forHost.getOrElseUpdate(loc.host, new
ArrayBuffer) += index
- for (rack <- sched.getRackForHost(loc.host)) {
- pendingSpeculatableTasks.forRack.getOrElseUpdate(rack, new
ArrayBuffer) += index
- }
- }
-
- if (tasks(index).preferredLocations == Nil) {
- pendingSpeculatableTasks.noPrefs += index
- }
-
- pendingSpeculatableTasks.anyPrefs += index
- } else {
-
- for (loc <- tasks(index).preferredLocations) {
- loc match {
- case e: ExecutorCacheTaskLocation =>
- pendingTasks.forExecutor.getOrElseUpdate(e.executorId, new
ArrayBuffer) += index
- case e: HDFSCacheTaskLocation =>
- val exe = sched.getExecutorsAliveOnHost(loc.host)
- exe match {
- case Some(set) =>
- for (e <- set) {
- pendingTasks.forExecutor.getOrElseUpdate(e, new ArrayBuffer)
+= index
- }
- logInfo(s"Pending task $index has a cached location at
${e.host} " +
- ", where there are executors " + set.mkString(","))
- case None => logDebug(s"Pending task $index has a cached
location at ${e.host} " +
- ", but there are no executors alive there.")
- }
- case _ =>
- }
- pendingTasks.forHost.getOrElseUpdate(loc.host, new ArrayBuffer) +=
index
-
- if (resolveRacks) {
- sched.getRackForHost(loc.host).foreach { rack =>
- pendingTasks.forRack.getOrElseUpdate(rack, new ArrayBuffer) +=
index
+ val pendingTaskSetToAddTo = if (speculative) pendingSpeculatableTasks else
pendingTasks
+ // ... mostly the original code from `addPendingTask` here, just adding
+ // into pendingTaskSetToAddTo
Review comment:
Done, removed the comment
----------------------------------------------------------------
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]
