Skip to main content
Log in

Fast Paxos

  • Original Article
  • Published:
Distributed Computing Aims and scope Submit manuscript


As used in practice, traditional consensus algorithms require three message delays before any process can learn the chosen value. Fast Paxos is an extension of the classic Paxos algorithm that allows the value to be learned in two message delays. How and why the algorithm works are explained informally, and a TLA+ specification of the algorithm appears as an appendix.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others


  1. Brasileiro, F., Greve, F., Mostefaoui, A., Raynal, M.: Consensus in one communication step. In: Malyshkin, V. (ed.) Parallel Computing Technologies (6th International Conference, PaCT 2001), Lecture Notes in Computer Science, vol. 2127, pp. 42–50. Springer, Berlin Heidelberg New York (2001)

  2. Castro, M., Liskov, B.: Practical byzantine fault tolerance. In: Proceedings of the Third Symposium on Operating Systems Design and Implementation, pp. 173–186. ACM, New York (1999)

  3. Charron-Bost, B., Schiper, A.: Uniform consensus is harder than consensus (extended abstract). Tech. Rep. DSC/2000/028, École Polytechnique Fédérale de , Switzerland (2000).

  4. De Prisco R., Lampson B., Lynch N. (2000) Revisiting the paxos algorithm. Theor. Comput. Sci. 243, 35–91

    Article  MathSciNet  Google Scholar 

  5. Fischer M.J., Lynch N., Paterson M.S. (1985) Impossibility of distributed consensus with one faulty process. J. ACM 32(2): 374–382

    Article  MATH  MathSciNet  Google Scholar 

  6. Lamport, L.: Introduction to TLA. SRC Technical Note 1994-001, Digital Systems Research Center (1994). Currently available from

  7. Lamport L. (1998) The part-time parliament. ACM Trans. Comput. Syst. 16(2): 133–169

    Article  Google Scholar 

  8. Lamport, L.: A summary of TLA+ (2000). Currently available from or by searching the Web for the 21-letter string obtained by removing the – characters from uid-lamport-tla-homepage

  9. Lamport L. (2001) Paxos made simple. ACM SIGACT News (Distributed Computing Column) 32(4): 18–25

    Google Scholar 

  10. Lamport, L.: Specifying Systems. Addison-Wesley, Boston (2003). Also available on the Web via a link at

  11. Lamport, L.: Lower bounds for asynchronous consensus. Tech. Rep. MSR-TR-2004-71, Microsoft Research (2004). Currently available from, or by searching the Web for the 23-letter string obtained by removing the – characters from all-lamports-pubs-onthe-web

  12. Lampson B.W. How to build a highly available system using consensus. In: Babaoglu O. Marzullo K. (eds) Distributed Algorithms. Lecture Notes in Computer Science, vol. 1151, pp. 1–17. Springer Berlin, Heidelberg New York (1996)

  13. Martin, J.P., Alvisi, L.: Fast byzantine consensus. In: Proceedings of the International Conference on Dependable Systems and Networks (DSN 2005). IEEE Computer Society, Yokohama (2005) (in press)

  14. Pedone F., Schiper A. (2002) Handling message semantics with generic broadcast. Distributed Computing 15(2): 97–107

    Article  Google Scholar 

  15. Pedone, F., Schiper, A., Urbán, P., Cavin, D.: Solving agreement problems with weak ordering oracles. In: Proceedings of the 4th European Dependable Computing Conference (EDCC-4). Lecture Notes in Computer Science, vol. 2485, pp. 44–61. Springer, Berlin Heidelberg New York (2002)

  16. Zielinski, P.: Optimistic generic broadcast. In: Fraigniaud, P. (ed.) DISC ’05: Proceedings of the 19th International Conference on Distributed Computing, Lecture Notes in Computer Science, vol. 3724, pp. 369–383. Springer, Berlin Heidelberg New York (2005)

Download references

Author information

Authors and Affiliations


Rights and permissions

Reprints and permissions

About this article

Cite this article

Lamport, L. Fast Paxos. Distrib. Comput. 19, 79–103 (2006).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: