|
Archives of the TeradataForumMessage Posted: Fri, 24 May 2003 @ 00:02:28 GMT
My rule of thumb from experience using single data warehouses larger than 1 TB usable data Oracle, DB2, Sybase approx 5 DBAs This furthermore assumes that the queries are non predictable. If you have predictable queries, the effort on the traditional RDBMS side goes substantially down. Also, there are slight variations between ORCL, DB2 etc. but overall there is significant more need to handle the indices and details with DBs from the transactional side of the world than for traditional Data warehouses with TD. Taking similar numbers it is actually rather easy to guestimate the break even point for the higher capital investment for TD versus the runtime cost. Again, this things only work in order of magnitude, they are not scientifical formulas. Thomas Thomas Buehlmann
| ||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
Copyright 2016 - All Rights Reserved | ||||||||||||||||||||||||||||||||||||||||||||||||
Last Modified: 15 Jun 2023 | ||||||||||||||||||||||||||||||||||||||||||||||||