Archives of the TeradataForum
Message Posted: Thu, 28 Aug 2003 @ 14:07:16 GMT
| Subj: | | Re: Join Order In Teradata |
| |
| From: | | Dieter Noeth |
Dave Harper wrote:
| | I had a general question regarding how/why Teradata is join order specific. I hadn't been aware that table ordering in my join
condition would have an impact on the query plan that is generated -- I know it isn't like that in SQL Server and I believe Oracle. Is this
done for MPP so that each spool can be merged more easily dependant upon the data it retrieves? | |
The plan should be the same regardless of the join order.
Can you please post an example?
Are you using "tab1 JOIN tab2 ON ..." syntax or "tab1, tab2 WHERE ..."? There are some optimizations which are not applied when using
the first one (V2R5.0.2 fixes that).
Dieter
| 
