[ https://issues.apache.org/jira/browse/MAPREDUCE-1463?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12835076#action_12835076 ]
Arun C Murthy commented on MAPREDUCE-1463: ------------------------------------------ Actually, assuming we have a reasonable model, I'd do it on the JobTracker, maybe in JobInProgress.initTasks i.e. during job-initialization. > Reducer should start faster for smaller jobs > -------------------------------------------- > > Key: MAPREDUCE-1463 > URL: https://issues.apache.org/jira/browse/MAPREDUCE-1463 > Project: Hadoop Map/Reduce > Issue Type: Improvement > Components: jobtracker > Reporter: Scott Chen > Assignee: Scott Chen > Attachments: MAPREDUCE-1463-v1.patch, MAPREDUCE-1463-v2.patch, > MAPREDUCE-1463-v3.patch > > > Our users often complain about the slowness of smaller ad-hoc jobs. > The overhead to wait for the reducers to start in this case is significant. > It will be good if we can start the reducer sooner in this case. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.