Work

Paxos

paper · 1989

Distributed Systems Consensus Algorithms

Paxos is a family of protocols for solving consensus in a network of unreliable processors. Developed by Leslie Lamport, it became the foundation for building reliable distributed systems and influenced virtually all modern consensus algorithms.

The Consensus Problem

How can a group of computers agree on a single value when messages can be lost or delayed, and some computers might fail? This problem is fundamental to distributed systems—from database replication to distributed locks.

The Algorithm

Paxos uses a sequence of proposals and acceptances:

  1. Prepare: A proposer sends a proposal number to acceptors
  2. Promise: Acceptors promise not to accept older proposals
  3. Accept: The proposer asks acceptors to accept a value
  4. Learn: Once a majority accepts, the value is chosen

Presentation and Reception

Lamport first wrote about Paxos in 1989 using an allegory about a fictional Greek island parliament. The paper was considered too whimsical and wasn’t published until 1998. A more straightforward explanation, “Paxos Made Simple,” followed in 2001.

Impact

Paxos and its variants underpin modern distributed systems: Google’s Chubby and Spanner, Apache ZooKeeper, and countless other systems use Paxos-derived protocols. The Raft consensus algorithm was designed as a more understandable alternative to Paxos.