Details
Description
Use SemaphoredDelegatingExecutor for each to stream to submit work, if possible, for better fairness in processes with many streams.
this also takes a DurationTrackerFactory to count how long was spent in the queue, something we would want to know
if cherrypicking, follow with HADOOP-18466 and HADOOP-18455
Attachments
Issue Links
- causes
-
HADOOP-18455 s3a prefetching Executor should be closed
- Resolved
- is depended upon by
-
HADOOP-18703 Backport S3A prefetching stream to branch-3.3
- Resolved
- links to