Github user CodingCat commented on a diff in the pull request:
https://github.com/apache/spark/pull/214#discussion_r10896811
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala ---
@@ -198,6 +201,13 @@ private[spark] class TaskSchedulerImpl(
*/
def resourceOffers(offers: Seq[WorkerOffer]): Seq[Seq[TaskDescription]]
= synchronized {
SparkEnv.set(sc.env)
+ // Make thread pool local for shutdown before the function returns
+ // This is for driver can exit normally which not call sc.stop or
sys.exit
+ val serializeWorkerPool = new ThreadPoolExecutor(
--- End diff --
I'm not sure if new a thread pool for every call of resourceOffer() is a
good choice, this function is called for every second by default...is this
overhead necessary? I think it's OK to build a threadpool running in the
lifecycle of taskscheduler, once the taskscheduler is shutdown,
(SparkContext.stop() is called), you can shutdown the thread pool
Glad to hear others' voice
---
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.
---
