Denormalization refers to introducing redundancy intentionally to improve performance.

• A. Denormalization reduces data redundancy by splitting tables.
• B. Denormalization refers to introducing redundancy intentionally to improve performance.
• C. Denormalization eliminates the need for data integrity constraints.
• D. Denormalization guarantees zero update anomalies.

Answer: (B)

Denormalization is the practice of duplicating data across tables to speed up read queries by reducing the number of joins. This intentional redundancy is done to boost performance, especially for workloads that read data more often than they write. By having redundant data in place, queries can be served more quickly since they don’t have to traverse multiple tables or perform complex joins.

That aligns with the statement because it explicitly describes introducing redundancy on purpose to improve performance. In contrast, normalization aims to minimize redundancy by splitting data into related tables, and it does not describe duplicating data for speed. Also, eliminating data integrity constraints or guaranteeing zero update anomalies are not defining aspects of denormalization; denormalization can make maintenance more complex and can introduce update anomalies if not managed carefully.