You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hive.apache.org by "okumin (Jira)" <ji...@apache.org> on 2020/07/10 05:15:00 UTC

[jira] [Created] (HIVE-23831) Enforce auto-parallelism of Tez even if the number of reducers can be 1

okumin created HIVE-23831:
-----------------------------

             Summary: Enforce auto-parallelism of Tez even if the number of reducers can be 1
                 Key: HIVE-23831
                 URL: https://issues.apache.org/jira/browse/HIVE-23831
             Project: Hive
          Issue Type: Improvement
          Components: Hive
            Reporter: okumin
            Assignee: okumin


Hive on Tez disables auto-reduce and uses the maximum number of partitions when the minimum number of reducers, `estimated number * min.partition.factor`, is less than or equal to 1.

[https://github.com/apache/hive/blob/f98e136bdd5642e3de10d2fd1a4c14d1d6762113/ql/src/java/org/apache/hadoop/hive/ql/parse/GenTezUtils.java#L121-L131]

https://issues.apache.org/jira/browse/HIVE-14200

 

This can make Tez lose chance to optimize # of reducers especially in an environment with small `hive.tez.min.partition.factor`.

I'd say it would be fine to introduce an option to enforce auto-reduce.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)