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

Posted to user@hive.apache.org by Wil - <wi...@yahoo.com> on 2011/03/01 00:58:39 UTC

Dynamic partition - support for distribute by

Hi,

Reading the wiki on dynamic partition, there is best practice example to solve 
the issue of creating too many dynamic partitions on a specific node. However, 
the query does not work. 
(http://wiki.apache.org/hadoop/Hive/Tutorial#Dynamic-partition_Insert)


Is this form of query support?


FROM table_a
INSERT OVERWRITE TABLE table_b PARTITION(dt)
SELECT * FROM (
 SELECT *, dt
 DISTRIBUTE BY dt
) T;

Thanks,
--wil

Re: Dynamic partition - support for distribute by

Posted by Wil - <wi...@yahoo.com>.

Thanks, the query works as expected. I guess the query on the wiki is out of 
date.


----- Original Message ----
From: Thiruvel Thirumoolan <th...@yahoo-inc.com>
To: "user@hive.apache.org" <us...@hive.apache.org>
Sent: Tue, March 1, 2011 3:26:13 AM
Subject: Re: Dynamic partition - support for distribute by

Not sure about that, but this is supported:

FROM (SELECT *, dt from table_a DISTRIBUTE BY dt) T
        INSERT OVERWRITE TABLE table_b PARTITION(dt)
                SELECT * ;

On Mar 1, 2011, at 5:28 AM, Wil - wrote:

> Hi,
> 
> Reading the wiki on dynamic partition, there is best practice example to solve 

> the issue of creating too many dynamic partitions on a specific node. However, 

> the query does not work. 
> (http://wiki.apache.org/hadoop/Hive/Tutorial#Dynamic-partition_Insert)
> 
> 
> Is this form of query support?
> 
> 
> FROM table_a
> INSERT OVERWRITE TABLE table_b PARTITION(dt)
> SELECT * FROM (
> SELECT *, dt
> DISTRIBUTE BY dt
> ) T;
> 
> Thanks,
> --wil
> 
> 
>

Re: Dynamic partition - support for distribute by

Posted by Thiruvel Thirumoolan <th...@yahoo-inc.com>.

Not sure about that, but this is supported:

FROM (SELECT *, dt from table_a DISTRIBUTE BY dt) T
        INSERT OVERWRITE TABLE table_b PARTITION(dt)
                SELECT * ;

On Mar 1, 2011, at 5:28 AM, Wil - wrote:

> Hi,
> 
> Reading the wiki on dynamic partition, there is best practice example to solve 
> the issue of creating too many dynamic partitions on a specific node. However, 
> the query does not work. 
> (http://wiki.apache.org/hadoop/Hive/Tutorial#Dynamic-partition_Insert)
> 
> 
> Is this form of query support?
> 
> 
> FROM table_a
> INSERT OVERWRITE TABLE table_b PARTITION(dt)
> SELECT * FROM (
> SELECT *, dt
> DISTRIBUTE BY dt
> ) T;
> 
> Thanks,
> --wil
> 
> 
>