Github user kayousterhout commented on a diff in the pull request:
https://github.com/apache/spark/pull/6652#discussion_r31777954
--- Diff: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
---
@@ -137,6 +137,15 @@ class DAGScheduler(
private[scheduler] val eventProcessLoop = new
DAGSchedulerEventProcessLoop(this)
taskScheduler.setDAGScheduler(this)
+ // Number of map, reduce tasks above which we do not assign preferred
locations
+ // based on map output sizes. We limit the size of jobs for which assign
preferred locations
+ // as sorting the locations by size becomes expensive.
+ private[this] val SHUFFLE_PREF_MAP_THRESHOLD = 1000
+ // NOTE: This should be less than 2000 as we use
HighlyCompressedMapStatus beyond that
+ private[this] val SHUFFLE_PREF_REDUCE_THRESHOLD = 1000
+ // Number of preferred locations to use for reducer tasks
+ private[scheduler] val NUM_REDUCER_PREF_LOCS = 5
--- End diff --
one more comment request here: can you add one sentence explaining this?
something like "making this smaller will focus on the locations where the most
data can be read locally, but may lead to lower scheduling efficiency or the
delay scheduling timers expiring if all of those locations are busy"
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
