|
Archives of the TeradataForumMessage Posted: Wed, 03 Aug 2005 @ 13:52:30 GMT
I will try to make my comments at the fifty-thousand foot level. Look into value compression as well as the priority scheduler which can make or break a warehouse. One for time the other for space. Look into resource partitioning so you can tweak weights and move jobs/sessions across resources. If you plan to use OLTP then make sure you have a sound backup/recovery strategy. Reporting tools will use alot of volatile tables. This makes query tuning for these a bit more tedious. Reorgs are called pack disk and spool space acts like a large temporary table. Don't forget that statistics are important and should be a big part of your strategy for tuning the optimizer. Don't rely on random amp samples. It would be a great to engage in technical services from NCR or a qualified consulting firm to avoid a long list of "lessons learned". Cheers, Mike Abbonizio
| ||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
Copyright 2016 - All Rights Reserved | ||||||||||||||||||||||||||||||||||||||||||||||||
Last Modified: 15 Jun 2023 | ||||||||||||||||||||||||||||||||||||||||||||||||