Consistency in traditional RDBMS revolves around maintaining database-level rules during transactions, such as uniqueness constraints and default values. In distributed databases like Vitess, ensuring uniqueness across multiple shards adds complexity, requiring a re-implementation of these rules. This highlights the challenges of maintaining data integrity in a distributed environment.
