What is a quorum-based consensus algorithm, and where is it used in mission-critical systems?

What is a quorum-based consensus algorithm, and where is it used in mission-critical systems?

• A. They have no fault tolerance.
• B. Algorithms like Paxos or Raft require a majority to agree, used in distributed databases and configuration stores to ensure consistency.
• C. They are used only for load balancing.
• D. They require a single leader to decide all writes.

Answer: (B)

Quorum-based consensus means a value is considered chosen only after a majority of the replicas have agreed on it. This majority requirement gives the system fault tolerance: even if some nodes fail or messages are delayed, the cluster can keep progressing as long as more than half of the nodes are reachable. Algorithms like Paxos and Raft implement this idea, ensuring that all correct replicas converge on the same sequence of writes and that decisions remain safe despite failures. In mission-critical systems, this approach is used to keep replicated data in sync across machines, which is essential for maintaining strong consistency in distributed databases and configuration stores. For example, services like etcd rely on Raft to coordinate state across nodes. This isn’t about load balancing, and while a leader may coordinate actions, the final decision to commit a write requires agreement from a majority, not a single node alone.

Quorum-based consensus means a value is considered chosen only after a majority of the replicas have agreed on it. This majority requirement gives the system fault tolerance: even if some nodes fail or messages are delayed, the cluster can keep progressing as long as more than half of the nodes are reachable. Algorithms like Paxos and Raft implement this idea, ensuring that all correct replicas converge on the same sequence of writes and that decisions remain safe despite failures. In mission-critical systems, this approach is used to keep replicated data in sync across machines, which is essential for maintaining strong consistency in distributed databases and configuration stores. For example, services like etcd rely on Raft to coordinate state across nodes. This isn’t about load balancing, and while a leader may coordinate actions, the final decision to commit a write requires agreement from a majority, not a single node alone.