Run sampled queries (ie. one mapper only) to allow testing and estimation of 
data volume expansion/contraction for mapred.reduce.tasks. 
----------------------------------------------------------------------------------------------------------------------------------------

                 Key: HIVE-304
                 URL: https://issues.apache.org/jira/browse/HIVE-304
             Project: Hadoop Hive
          Issue Type: Improvement
            Reporter: Venky Iyer


It should be possible to run a sample of a query so that users can verify the 
output looks reasonable (for testing), and to estimate various parameters about 
how the data volume changes, for intelligent allocation of cluster resources. 
One way to do this is to allow tagging of the query with some 'tag string' so 
that the same estimated parameters can be re-used for all queries run with that 
'tag string'. 


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to