Denormalizing the Physical Schema | Database Design | Teradata Vantage - Denormalizing the Physical Schema - Teradata Vantage - Analytics Database

Database Design

Deployment
VantageCloud
VantageCore
Edition
VMware
Enterprise
IntelliFlex
Product
Analytics Database
Teradata Vantage
Release Number
17.20
Published
June 2022
ft:locale
en-US
ft:lastEdition
2025-11-21
dita:mapPath
ogg1628096130566.ditamap
dita:ditavalPath
qkf1628213546010.ditaval
dita:id
zqc1472244571611
lifecycle
lifecycle
Product Category
Teradata Vantage™

This section describes some of the common ways to denormalize the physical implementation of a normalized model. The section also briefly describing the popular technique of dimensional modeling and shows how the most useful attributes of a dimensional model can be emulated for a normalized physical database implementation through the careful use of dimensional views.

The term denormalization describes any number of physical implementation techniques that enhance performance by reducing or eliminating the isomorphic mapping of the logical database design on the physical implementation of that design. The result of these operations is usually a violation of the design goal of making databases application-neutral. In other words, a “denormalized” database favors one or a few applications at the expense of all other possible applications.

Strictly speaking, these operations are not denormalization at all. The concept of database schema normalization is logical, not physical. Logical denormalization should be avoided. Develop a normalized design and then, if necessary, adjust the semantic layer of your physical implementation to provide the desired performance enhancement. Finally, use views to tailor the external schema to the usability needs of users and to limit their direct access to base tables (see Dimensional Views).