CAP Theorem

The CAP theorem states that a distributed data store can provide at most two of three desirable guarantees at the same time: consistency, meaning every read receives the most recent write; availability, meaning every request receives a non-error response; and partition tolerance, meaning the system continues operating despite network communication failures between its nodes. Computer scientist Eric Brewer first proposed the idea as a conjecture in a 2000 conference keynote, and computer scientists Seth Gilbert and Nancy Lynch published a formal proof of the theorem in a 2002 paper. Because real-world distributed systems must tolerate network partitions, the theorem is most often applied as a practical choice between prioritizing consistency or availability specifically during a partition event, rather than as a strict three-way tradeoff under normal operation. The CAP theorem became a foundational reference point in the design of distributed databases and large-scale internet systems, shaping the tradeoffs behind many modern database architectures.