SQL Server® 2008 Administration: Real World Skills for MCITP Certification and Beyond

Denormalizing a Database

After a dozen pages or so explaining how and why to normalize your databases, it may seem odd to begin discussing denormalization. After all, why would you want to intentionally structure a database that could be susceptible to anomalies? If you think back to the anomalies that were discussed, they all had to do with data modifications (inserts, updates, and deletes). Read anomalies or unexpected results rarely occur. You usually just get the data you ask for when querying a database. That which is sometimes called a read anomaly is more often a programming error in the application code. When users—particularly decision support staff—must perform ad hoc queries (custom queries built on-the-fly) against your data, working ...