The CAP theorem, also known as Brewer’s theorem, was introduced by Eric Brewer in 1998 as a conjecture. It states that any distributed system cannot have consistency, availability, and partition tolerance simultaneously.
Consistency is a property that ensures that all nodes in a distributed system have a single, current, and identical copy of the data. Consistency is achieved using consensus algorithms in order to ensure that all nodes have the same copy of the data. This is also called state machine replication.
Availability means that the nodes in the system are up, accessible for use, and are accepting incoming requests and responding with data without any failures as and when required.
Partition tolerance ensures that if a group of nodes is unable to communicate with other nodes due to network failures, the distributed system continues to operate correctly.
A Venn diagram as shown below is commonly used to visualize the CAP theorem:
The preceding diagram shows that only two properties at a time can be achieved. Either AP, CA, or CP.
In summary:
- If we opt for CP (consistency and partition tolerance), we sacrifice availability.
- If we opt for AP (availability and partition tolerance), we sacrifice consistency.
- If we opt for AC (availability and consistency), we sacrifice partition tolerance.
In blockchains, consistency is sacrificed in favor of availability and partition tolerance. In this scenario, Consistency (C) on the blockchain is not achieved simultaneously with Partition tolerance (P) and Availability (A), but it is achieved over time.
This is called eventual consistency, where consistency is achieved as a result of validation from multiple nodes over time. It means that there can be a temporary disagreement between nodes on the final state, but it is eventually agreed upon.
For example, in Bitcoin, multiple transaction confirmations are required to achieve a good level of confidence that transactions may not be rolled back in the future and eventually a consistent view of transaction history is available to all nodes. Multiple confirmations of a transaction over time provide eventual consistency in Bitcoin.
Suryateja Pericherla, at present is a Research Scholar (full-time Ph.D.) in the Dept. of Computer Science & Systems Engineering at Andhra University, Visakhapatnam. Previously worked as an Associate Professor in the Dept. of CSE at Vishnu Institute of Technology, India.
He has 11+ years of teaching experience and is an individual researcher whose research interests are Cloud Computing, Internet of Things, Computer Security, Network Security and Blockchain.
He is a member of professional societies like IEEE, ACM, CSI and ISCA. He published several research papers which are indexed by SCIE, WoS, Scopus, Springer and others.
Leave a Reply