CAP theorem

The CAP theorem states that any distributed data store can only provide two of three guarantees: Consistency (all nodes see the same data), Availability (every request gets a response, possibly stale), and Partition Tolerance (system works despite network failures). In a network partition, a choice must be made between consistency and availability. Normal operations can satisfy all three. The theorem was proposed by Eric Brewer in 2000 and formally proved in 2002. Systems like RDBMS choose consistency, while many NoSQL databases choose availability, though MongoDB and Redis prioritize consistency. Brewer later clarified that the 'two out of three' framing is misleading since partitions are the only scenario requiring trade-offs. The PACELC theorem extends CAP by considering latency-consistency trade-offs even without partitions.