Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
3.0.0-alpha1
-
None
-
None
Description
In testing a high-write-throughput workload, I see the DN maintain good performance most of the time, except that occasionally one thread will block for a few seconds in finalizeReplica. It does so holding the FSDatasetImpl lock, which causes all other writer threads to block behind it. HDFS-1148 (making it a rw lock) would help here, but a bigger help would be to go do finer-grained locking (eg per block or per-subdir).
Attachments
Issue Links
- is related to
-
HDFS-1148 Convert FSDataset to ReadWriteLock
- Patch Available