Archives of the TeradataForum

Message Posted: Mon, 20 Feb 2006 @ 11:29:16 GMT

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

Hi,

We are working on a staging table that keeps 15 days of data. Every day around 300 millions records are getting loaded into this table. There is UPI (includes 8 columns) defined on this table partitioned on date column (PARTITION BY RANGE_N(START_DATE BETWEEN '2001-01-01' AND '2020-12- 31' EACH INTERVAL '1' DAY).

Statistics has been collected on this table on the date column. But there is no statistics on Index column. Even the simplest query (for eg. Select start_date, count(*) from table1 where start_date between '2006-02-05' and '2006-02-15') on this big table takes more than hour during day time when the system is also overloaded. System load could be one of the reason for this performance issue. But, is there any other way so that the performance can be increased? Will collecting statistics on "index" help in this case?

I tried collecting statistics on this table on date column again. And, this itself is running for 3 hrs and still running.

Thanks

Rishi

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