How to Determine Candidates for Multi-Level Partitioned Primary Index with DBQL
One of the more difficult challenges in database management and administration is determining where and how to implement a new RDBMS feature or function. In this presentation we’ll look at the DBQL data available for evaluation of tables and columns used within a workload and how this data can be leveraged for determining candidates for MLPPI (Multi-Level Partitioned Primary Index).
This presentation will discuss using DBQL data as the one source for gathering such key information as frequency of use, large scan and total query counts, as well as CPU and IO usage. In addition the presentation will review techniques for testing and validating conclusions with confidence. Finally, learn how using the wealth of information available in DBQL data can greatly facilitate taking advantage of powerful features and functions like MLPPI within the Teradata RDBMS to achieve significant performance gains.
Note: This was a 2010 Teradata Partners Conference session.
- David Hensel, Senior PS Consultant - Teradata America’s Professional Services
- Donna Becker, Director - Teradata America’s Professional Services
Audience: Data Warehouse Administrator, Data Warehouse Architect/Designer, Data Warehouse Technical Specialist