Question

According to the CAP theorem, if a distributed system must prioritize data consistency and network partition tolerance, what must it sacrifice during a network partition?

Accepted Answer

According to the CAP theorem, if a distributed system prioritizes consistency and partition tolerance, it must sacrifice availability. The CAP theorem states that any distributed data store can only provide two of three guarantees: consistency, availability, and partition tolerance. Consistency means that every read request receives the most recent write or an error. Availability means that every request receives a non-error response, without the guarantee that it contains the most recent write. Partition tolerance means the system continues to operate despite an arbitrary number of messages being dropped or delayed by the network between nodes. When a network partition occurs, a communication break prevents nodes from sharing information. To maintain consistency, the system must ensure that data remains uniform across all nodes. If the system cannot communicate to update all nodes during the partition, it must refuse to process requests to prevent the return of potentially stale or conflicting data. By choosing to return an error instead of out-of-date information, the system chooses consistency over availability. An example is a banking database that locks an account during a network failure to prevent the same funds from being withdrawn twice from different branches; by prioritizing the accuracy of the balance over the ability to perform a transaction, the system sacrifices availability.

Home → All Courses → Engineering and Technology Courses → Software Architecture and System Design → Flashcard

According to the CAP theorem, if a distributed system must prioritize data consistency and network partition tolerance, what must it sacrifice during a network partition?