spark-instrumented-optimizer/core
yi.wu 0811666ab1 [SPARK-32878][CORE] Avoid scheduling TaskSetManager which has no pending tasks
### What changes were proposed in this pull request?

This PR proposes to avoid scheduling the (non-zombie) TaskSetManager which has no pending tasks.

### Why are the changes needed?

Currently, Spark always tries to schedule a (non-zombie) TaskSetManager even if it has no pending tasks. This causes notable problems for the barrier TaskSetManager: 1. `calculateAvailableSlots` can be called for multiple times for a launched barrier TaskSetManager; 2. user would see "Skip current round of resource offers for barrier stage" log message for
a launched barrier TaskSetManager all the time until the barrier TaskSetManager finishes, which is quite confused.

Besides, scheduling a TaskSetManager always involves many function invocations even if there're no pending tasks.

Therefore, I think we can skip those un-schedulable TasksetManagers to avoid the potential overhead.

### Does this PR introduce _any_ user-facing change?

No.

### How was this patch tested?

Pass existing tests.

Closes #29750 from Ngone51/filter-out-unschedulable-stage.

Authored-by: yi.wu <yi.wu@databricks.com>
Signed-off-by: Dongjoon Hyun <dongjoon@apache.org>
2020-09-14 21:15:06 -07:00
..
benchmarks [SPARK-32437][CORE] Improve MapStatus deserialization speed with RoaringBitmap 0.9.0 2020-07-25 08:07:28 -07:00
src [SPARK-32878][CORE] Avoid scheduling TaskSetManager which has no pending tasks 2020-09-14 21:15:06 -07:00
pom.xml [SPARK-31765][WEBUI][TEST-MAVEN] Upgrade HtmlUnit >= 2.37.0 2020-06-11 18:27:53 -05:00