Details
-
Sub-task
-
Status: Resolved
-
Major
-
Resolution: Implemented
-
None
-
None
Description
Some minor optimizations for write path, particularly for small writes use-cases like hbase:
- Usage Collection.parallelStream() is premature and bring more cost than benefit. The Fork/Join model of parallelStream is just paranoid for small collection processing.
- Remove unnecessary UUID creation in ChunkBufferImplWithByteBuffer.
- Remove a unnecessary lock in KeyOutputStream.
Attachments
Attachments
Issue Links
- links to