Data inconsistency can be prevented by preventing

Concept. Data redundancy means the same fact is physically stored in more than one place in a database. Data inconsistency is the state in which those duplicate copies hold different (conflicting) values for what should be a single fact, so the database no longer agrees with itself.

Why redundancy is the root cause: inconsistency can only occur when there is more than one copy to disagree. If a fact is stored only once, there is no second copy that can fall out of step when an update happens.

Application.

1. Suppose a customer's address is duplicated across several tables.

2. An update changes the address in one table but, by oversight, not in the others.

3. The copies now disagree — the database is inconsistent.

4. Remove the duplication (store the address exactly once, e.g. through normalization), and there is no second copy to update, so the inconsistency can never arise.

Cross-check / contrast.

• Redundancy — the duplication itself; eliminating it (normalization) is the standard way to prevent inconsistency.

• Metadata — "data about data" (schema, definitions, constraints); it describes the data rather than being the stored payload.

• Deadlocks — a concurrency-control problem where transactions wait on each other forever; a locking/scheduling issue.

• Multiprogramming — running several programs concurrently on the CPU; an operating-system scheduling idea.

Result: data inconsistency is prevented by preventing redundancy.