You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Namit Jain (JIRA)" <ji...@apache.org> on 2012/09/07 09:07:07 UTC
[jira] [Created] (HIVE-3445) Skewed join optimization for different
datatypes
Namit Jain created HIVE-3445:
--------------------------------
Summary: Skewed join optimization for different datatypes
Key: HIVE-3445
URL: https://issues.apache.org/jira/browse/HIVE-3445
Project: Hive
Issue Type: Bug
Components: Query Processor
Reporter: Namit Jain
Assignee: Namit Jain
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-3445) Skewed join optimization for
different datatypes
Posted by "Namit Jain (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HIVE-3445?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13450434#comment-13450434 ]
Namit Jain commented on HIVE-3445:
----------------------------------
set hive.internal.ddl.list.bucketing.enable=true;
set hive.optimize.skewjoinoptimization = true;
-- testing skew on other data types - int
CREATE TABLE T1(key INT, val STRING) SKEWED BY (key) ON ((2));
CREATE TABLE T2(key STRING, val STRING) SKEWED BY (key) ON ((3)) STORED AS TEXTFILE;
EXPLAIN
SELECT a.*, b.* FROM T1 a JOIN T2 b ON a.key = b.key;
The join keys for the two tables are different.
Ideally, the skew join optimization would still work for this, since the udf converting the join keys has a
one-one mapping. Instead of hard-coding this logic in SkewJoinOptimizer, it would be cleaner to create a new
annotation oneToOne, which would be valid for the type conversion udfs.
> Skewed join optimization for different datatypes
> ------------------------------------------------
>
> Key: HIVE-3445
> URL: https://issues.apache.org/jira/browse/HIVE-3445
> Project: Hive
> Issue Type: Bug
> Components: Query Processor
> Reporter: Namit Jain
> Assignee: Namit Jain
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira