Github user kayousterhout commented on the pull request:
https://github.com/apache/spark/pull/1486#issuecomment-49483833
Does choice 1 (/this patch) change anything about the order in which things
are scheduled? My understanding (based on
https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala#L183)
is that we use a task's preferred locations to create a hash map of hosts to
tasks that can be run on them, and then when one of the hosts becomes free,
we'll schedule one of the tasks in the corresponding entry in the hash map.
So, I don't think the order of preferred Locations for a task has any effect on
scheduling order?
---
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.
---
