Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
0.6.0, 0.7.0, 0.8.0-alpha
-
None
Description
TezMerger$MergeQueue::next() shows up in profiles due a synchronized block in a tight loop.
Part of the slow operation was due to DataInputBuffer issues identified earlier in HADOOP-10694, but along with that approx 11% of my lock prefix calls were originating from the following line.
mergeProgress.set(totalBytesProcessed * progPerByte);
in two places within the core loop.
Attachments
Attachments
Issue Links
- relates to
-
TEZ-2605 [Umbrella] CPU optimizations for hotspots
- Open