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

Posted to dev@hive.apache.org by xiaoyuan <yu...@163.com> on 2016/08/31 14:39:43 UTC

Why beeline use BufferedRows by default instead of IncrementalRows ?

Hi,

When I use beeline to execute a select * from a table with 4.2 GB data, beeline failed with OutOfMemoryError: GC overhead limit exceeded:


I also noticed that beeline use BufferedRows by default, If i change the default to IncrementalRows, the query will work. 


So why don’t we just change the default value to IncrementalRows ?

Thanks.

Re: Why beeline use BufferedRows by default instead of IncrementalRows ?

Posted by Sergio Pena <se...@cloudera.com>.

Hi,

There is a JIRA that was just fixed that sets incremental rows as default.
https://issues.apache.org/jira/browse/HIVE-14170

You should see this change on the next release Hive 2.2

On Wed, Aug 31, 2016 at 9:39 AM, xiaoyuan <yu...@163.com> wrote:

> Hi,
>
> When I use beeline to execute a select * from a table with 4.2 GB data,
> beeline failed with OutOfMemoryError: GC overhead limit exceeded:
>
> I also noticed that beeline use BufferedRows by default, If i change the
> default to IncrementalRows, the query will work.
>
> So why don’t we just change the default value to IncrementalRows ?
>
> Thanks.
>
>
>