[teiid-users] Undivided Join Query to HIVE

Ramesh Reddy rareddy at redhat.com
Mon May 14 10:40:09 EDT 2018


If both C and D tables are in Hive, yes it will send the joined query. You
need to see query plan as to why it is not sending a joined query.

On Mon, May 14, 2018 at 8:08 AM, JUNGHO PARK <j0106.park at samsung.com> wrote:

> Hello.
> I'm using Teiid 10.1.
>
> I observed Teiid divide multiple DBs Join SQL into each DB SQLs.
> This spesification is OK but it is unconveninet when using Hive because
> Hive execute multiple DBs join SQL by 1 mapreduce.
>
> Is there any way to avoid splitting multi DB Join SQL ?
>
> (Example)
> - Input to Teiid:
> SELECT c.id, d.id FROM a.c, b.d WHERE c.id = d.id;
>
> - Current Teiid output to Hive:
> 1. SELECT c.id FROM a.c
> 2. SELECT d.id FROM b.d
>
> - Expected output to Hive:
> SELECT c.id, d.id FROM a.c, b.d WHERE c.id = d.id;
>
> thanks.
>
> _______________________________________________
> teiid-users mailing list
> teiid-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/teiid-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/teiid-users/attachments/20180514/a2beede3/attachment.html 


More information about the teiid-users mailing list