You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Matt McCline (JIRA)" <ji...@apache.org> on 2016/05/07 08:28:12 UTC
[jira] [Commented] (HIVE-13713) We miss vectorization in a case of
count(*) when aggregation mode is COMPLETE
[ https://issues.apache.org/jira/browse/HIVE-13713?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15275153#comment-15275153 ]
Matt McCline commented on HIVE-13713:
-------------------------------------
With the fix, vectorization_limit.q does vectorize Reducer 2:
{code}
Reducer 2
Execution mode: vectorized
Reduce Operator Tree:
Group By Operator
keys: KEY._col0 (type: tinyint), KEY._col1 (type: double)
mode: mergepartial
outputColumnNames: _col0, _col1
Statistics: Num rows: 6144 Data size: 1320982 Basic stats: COMPLETE Column stats: NONE
Group By Operator
aggregations: count(_col1)
keys: _col0 (type: tinyint)
mode: complete
outputColumnNames: _col0, _col1
Statistics: Num rows: 3072 Data size: 660491 Basic stats: COMPLETE Column stats: NONE
Limit
Number of rows: 20
Statistics: Num rows: 20 Data size: 4300 Basic stats: COMPLETE Column stats: NONE
File Output Operator
compressed: false
Statistics: Num rows: 20 Data size: 4300 Basic stats: COMPLETE Column stats: NONE
table:
input format: org.apache.hadoop.mapred.SequenceFileInputFormat
output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat
serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
{code}
> We miss vectorization in a case of count(*) when aggregation mode is COMPLETE
> -----------------------------------------------------------------------------
>
> Key: HIVE-13713
> URL: https://issues.apache.org/jira/browse/HIVE-13713
> Project: Hive
> Issue Type: Bug
> Components: Hive
> Reporter: Matt McCline
> Assignee: Matt McCline
> Priority: Critical
>
> E.g. vectorization_limit.q doesn't vectorize Reducer 2 for the query:
> {code}
> "select ctinyint, count(distinct(cdouble)) from alltypesorc group by ctinyint order by ctinyint limit 20"
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)