[jira] [Created] (HIVE-6614) Vectorized aggregates computed on map side diffe (hash mode) from values computed on reduce side (streaming mode)

[Remus Rusanu (JIRA)]

Remus Rusanu created HIVE-6614:
----------------------------------

             Summary: Vectorized aggregates computed on map side diffe (hash 
mode) from values computed on reduce side (streaming mode)
                 Key: HIVE-6614
                 URL: https://issues.apache.org/jira/browse/HIVE-6614
             Project: Hive
          Issue Type: Bug
            Reporter: Remus Rusanu
            Assignee: Remus Rusanu
            Priority: Critical


HIVE-6222 allows vectorized aggregates to operate on streaming mode, ie. flush 
after each key change and let the shuffle+reduce side to compute the final 
aggregate values. An error in patch .2 for HIVE-6222 shows that when the 
queries run in streaming mode, there are rounding diffs for some agg functions 
(VAR and friends). These occurred for non-decimal types, like ctinyint:

{code}
select csmallint, VAR_POP(ctinyint) from alltypesorc where csmallint = -75 
group by csmallint;
{code}

This produces  107.55555555555556 vs. 107.55555555555554.






--
This message was sent by Atlassian JIRA
(v6.2#6252)