Data Integrity

Data Integrity, when referring to databases, is a concept that data in the database must be correct and accurate.

The quality criteria of data:

* Accuracy - how accurate is the stored data compared to the real world data.
* Completeness - what data is missing from the real world data.
* Consistency - how consistent is the stored data regarding data constraints and related data.
* Currentness - how up to date is the stored data compared to the real world data.

Data Integrity has four components:

* Entity Integrity
* Domain Integrity
* Referential Integrity
* Business rules

Entity Integrity - means that there is a mechanism that will allow only uniquely identified rows in a table. This is done with either primary keys or unique keys that will prevent duplicate rows.

Domain Integrity - means that there is a mechanism that will restrict the data within certain boundaries. Besides the data types having their own limitations on the nature and size or range permitted there are also constraints. Constraints can be rules or checks that will only accept data with a certain format or within a certain range. The formula used by the rule or check is an expression that might use data from the table itself for its calculations.

Referential Integrity - means that there is a mechanism that will keep the data in related tables synchronized.

Business rules - rules that are particular to a specific application. Triggers can go a step further than checks and rules by allowing expressions that can get data from other tables and even call SP's or XP's, making them very useful for this situation.

above book excerpt is from:

Super SQL Server Systems
Turbocharge Database Performance with C++ External Procedures

ISBN: 0-9761573-2-2
Joseph Gama, P. J. Naughter