[jira] [Commented] (HIVE-19113) Bucketing: Make CLUSTERED BY do CLUSTER BY if no explicit sorting is specified

["Jesus Camacho Rodriguez (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-19113?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16883393#comment-16883393 ] 

Jesus Camacho Rodriguez commented on HIVE-19113:
------------------------------------------------

[~gopalv], correct me if I am wrong. Currently the same behavior could be obtained by clustering and sorting by same column:

{code}
CREATE EXTERNAL TABLE bucketed (key int)
CLUSTERED BY(key) SORTED BY(key) INTO 4 BUCKETS;
{code}

It is just that {{CLUSTERED}} here acts as {{DISTRIBUTED}} and you are saying that it is misleading since {{CLUSTERED BY != CLUSTER BY}}?


> Bucketing: Make CLUSTERED BY do CLUSTER BY if no explicit sorting is specified
> ------------------------------------------------------------------------------
>
>                 Key: HIVE-19113
>                 URL: https://issues.apache.org/jira/browse/HIVE-19113
>             Project: Hive
>          Issue Type: Improvement
>          Components: Logical Optimizer
>    Affects Versions: 3.0.0
>            Reporter: Gopal V
>            Assignee: Deepak Jaiswal
>            Priority: Major
>
> The user's expectation of 
> "create external table bucketed (key int) clustered by (key) into 4 buckets stored as orc;"
> is that the table will cluster the key into 4 buckets, while the file layout does not do any actual clustering of rows.
> In the absence of a "SORTED BY", this can automatically do a "SORTED BY (key)" to cluster the keys within the file as expected.



--
This message was sent by Atlassian JIRA
(v7.6.14#76016)