Archives of the TeradataForum

Message Posted: Thu, 08 Jan 2004 @ 10:36:22 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Ganga,

Thank you for the answers. I am trying to add them to the context you provided in the original posting and I realize I would like to know how large the output of the fast query is:

SELECT * FROM
OBC_MASTER.VW_RECONTACTS
WHERE OCID = 9 AND MID = 48 ;

In your first mail you said:

"OBC_MASTER.FA_OBC_MDN_MTD has 20 Million rows" and together with the new information that

select count(*)
from OBC_MASTER.FA_OBC_MDN_MTD
where ACTIVITY_CODE in (1,2) ;

returns about the same 20 million rows, we should assume that the WHERE clause AND ACT IN (1,2) in the slow query should not matter much for the optimizer.

In the original posting you also said:

My idea was to keep this set (as I said, it would be good to know its size) as a (physical) derived table but apparently that does not happen. This would suggest that the optimizer might be doing some in-memory derived table processing. From recent discussions here we know that this feature should be better turned off. Have you looked into this yet?

Regards,

Victor

	<Prev	Next>		<<First	<Prev	Next>	Last>>