Archives of the TeradataForum

Message Posted: Fri, 27 Sep 2002 @ 10:34:07 GMT

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

Hi there

I have encountered a case of what looks like strange optimizer behaviour.

In general:

- a fairly complex query is executed very differently (and with widely varying execution times) depending on the value in a predicate (filter) on an NUSI column

- collect statistics has been done (is done regularly, but here done just to be sure)

- the optimizer seems to use an estimate of 1 row in the one case, but a reasonable estimate in the other

The details are in a MS Word doc., if anyone can assist, please respond to me directly, and I will send the doc. to you.

Any assistance and/or suggestions will be valuable

Mike Philips

Professional Services / Pre-Sales
National Data Systems (NDS)
(NCR exclusive distributor)
N1 City
7463
South Africa

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