 |
 |
Archives of the TeradataForum
Message Posted: Wed, 18 Jun 2003 @ 16:35:15 GMT
Subj: | | Re: Performance Analyze |
|
From: | | Geoffrey Rommel |
| Sorry, I am new to Teradata and I would like to know if the following request is good or not on big tables (~ 10 000 000
rows) | |
Actually, your EXPLAIN plan is not too bad, since it will have to do a full-table scan anyway, but this might work better:
select
DATE_ARRT_INFO,CODE_GUIC,CODE_ORGN_FINN,REFR_PRDT,NUMR_ENTT_TITL,....
from DW_SUIV_PRDT_COLT where
DATE_ARRT_INFO >= date '2003-06-01' - INTERVAL '6' MONTH
/* another way of saying the same thing: */
DATE_ARRT_INFO >= add_months(date '2003-06-01', -6)
This avoids the intermediate table and the join.
| |