Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Invalid
-
0.5.0
-
None
-
None
Description
We'd like to accumulate the statistics (and get graphs) at the job level rather than the job level, because it will provide a lot more useful information. This likely implies that they need to be rolled up in the TaskStatus reported up to the JobTracker.
Attachments
Issue Links
- duplicates
-
HADOOP-481 Hadoop mapred metrics should include per job input/output statistics rather than per-task statistics
- Closed