Archives of the TeradataForum

Message Posted: Tue, 30 Nov 2004 @ 13:23:23 GMT

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

Hello Ole

in answer to your questions

1) We are a retailer who use a normalised model (basket header and detail tables)

2) For the header table and the detail table the PI is basket id. Each basket detail row has a line number column for uniqueness

3) We do not use PPI's as we are not on V2r5 yet.

4) We do use a Secondary Index of Store and Transaction Date on the Basket Header to assist in query processing

The basket header and detail tables contain 65 weeks worh of data One thing we have found useful for affinity queries is having some subsets of the main data populated on a cyclical basis

These are

1) basket header and detail week to date

2) basket header and detail last elapsed week and same week last year

3) basket header and detail last 13 weeks and same weeks last year

These have had significant benefits for our marketing department as we have not had to build specific aggregates for them. We allow them and others to run complex affinity queries over these subsets of data to improve their analysis and protect the Data Warehouse environment

Regards

Phil Oldfield

Data Warehouse Teradata Senior Technical Specialist
Iceland Foods UK

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