You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-dev@hadoop.apache.org by "Harsh J (Created) (JIRA)" <ji...@apache.org> on 2012/03/13 14:11:39 UTC

[jira] [Created] (MAPREDUCE-4001) Improve MAPREDUCE-3789's fix logic by looking at job's slot demands instead

Improve MAPREDUCE-3789's fix logic by looking at job's slot demands instead
---------------------------------------------------------------------------

                 Key: MAPREDUCE-4001
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4001
             Project: Hadoop Map/Reduce
          Issue Type: Improvement
          Components: contrib/capacity-sched
    Affects Versions: 1.1.0
            Reporter: Harsh J
            Assignee: Harsh J
            Priority: Minor


In MAPREDUCE-3789, the fix had unfortunately only covered the first time assignment scenario, and the test had not really caught the mistake of using the condition of looking at available TT slots (instead of looking for how many slots a job's task demands).

We should change the condition of reservation in such a manner:

{code}
          if ((getPendingTasks(j) != 0 &&
               !hasSufficientReservedTaskTrackers(j)) &&
-                (taskTracker.getAvailableSlots(type) !=
+                !(j.getNumSlotsPerTask(type) >
                  getTTMaxSlotsForType(taskTrackerStatus, type))) {
{code}

I had not realized during the earlier ticket that j.getNumSlotsPerTask(type) did exist.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira