"This book should satisfy those who want a different perspective than the official Oracle documentation.
It will cover all important aspects of a data warehouse while giving the necessary examples to make the reading a lively experience." - Tim Donar, Author and Systems Architect for Enterprise Data WarehousesTuning a data warehouse database focuses on large transactions, mostly requiring what is known as throughput.
Throughput is the passing of large amounts of information through a server, network and Internet environment, backwards and forwards, constantly!
The ultimate objective of a data warehouse is the production of meaningful and useful reporting, from historical and archived data.
The trick is to make the reports print within an acceptable time frame. A data model contains tables and relationships between tables.
Tuning a data model involves Normalization and Denormalization.
Different approaches are required depending on the application, such as OLTP or a Data Warehouse.
Inappropriate database design can make SQL code impossible to tune.
Poor data modeling can have a most profound effect on database performance since all SQL code is constructed from the data model.