CAP Theorem

The CAP theorem says that a distributed data store can provide at most two of three guarantees at once: Consistency, Availability, and Partition tolerance. Since network partitions cannot be ruled out, the real choice during a partition is between consistency and availability.

Why it matters

CAP frames the design trade-offs behind distributed databases. A system that favors consistency (CP) may reject requests while a partition lasts. One that favors availability (AP) keeps answering but may return stale or conflicting data that must be reconciled later. Normal operation without a partition can offer both, so CAP describes behavior under failure.

Example

Two data centers lose their link.
CP choice: the minority side refuses writes -> consistent, less available
AP choice: both sides accept writes     -> available, may diverge

Common mistakes

• Reading it as "pick any two" at design time. Partition tolerance is not optional in a distributed system.
• Labeling a whole database CP or AP when many systems let you tune this per operation.
• Mixing up CAP consistency with the "C" in ACID, which concerns constraints.

Related

• Eventual consistency
• ACID
• Database replication
• NoSQL

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.