Advertisement

Asynchronous Reconfiguration for Paxos State Machines

  • Leander Jehl
  • Hein Meling
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8314)

Abstract

This paper addresses reconfiguration of a Replicated State Machine (RSM) in an asynchronous system. It is well known that consensus cannot be solved in an asynchronous system. Therefore an RSM providing strong consistency, cannot guarantee progress in an asynchronous system. However, we show that reconfiguring the RSM is possible in a purely asynchronous system. This differs from all existing reconfiguration methods which rely on consensus to choose a new configuration. Since a reconfiguration to a new set of machines or even a different datacenter can serve to restore synchrony between replicas, asynchronous reconfiguration can also serve to increase the availability of an RSM.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Schneider, F.B.: Implementing fault-tolerant services using the state machine approach: a tutorial. ACM Comput. Surv. 22(4), 299–319 (1990)CrossRefGoogle Scholar
  2. 2.
    Lamport, L.: The part-time parliament. ACM Trans. Comput. Syst. 16(2), 133–169 (1998)CrossRefGoogle Scholar
  3. 3.
    Lamport, L.: Paxos made simple. ACM SIGACT News (December 2001)Google Scholar
  4. 4.
    Shraer, A., Reed, B., Malkhi, D., Junqueira, F.: Dynamic reconfiguration of primary/backup clusters. USENIX ATC (2011)Google Scholar
  5. 5.
    Lamport, L., Malkhi, D., Zhou, L.: Vertical paxos and primary-backup replication. In: PODC (2009)Google Scholar
  6. 6.
    Fischer, M.J., Lynch, N.A., Paterson, M.S.: Impossibility of distributed consensus with one faulty process. J. ACM 32(2), 374–382 (1985)CrossRefzbMATHMathSciNetGoogle Scholar
  7. 7.
    Chandra, T.D., Hadzilacos, V., Toueg, S., Charron-Bost, B.: On the impossibility of group membership. In: PODC (1996)Google Scholar
  8. 8.
    Chandra, T.D., Toueg, S.: Unreliable failure detectors for reliable distributed systems. J. ACM 43, 225–267 (1996)CrossRefzbMATHMathSciNetGoogle Scholar
  9. 9.
    Lamport, L., Malkhi, D., Zhou, L.: Reconfiguring a state machine. SIGACT News 41(1), 63–73 (2010)CrossRefGoogle Scholar
  10. 10.
    Lorch, J.R., Adya, A., Bolosky, W.J., Chaiken, R., Douceur, J.R., Howell, J.: The smart way to migrate replicated stateful services. In: EuroSys (2006)Google Scholar
  11. 11.
    Aguilera, M.K., Keidar, I., Malkhi, D., Shraer, A.: Dynamic atomic storage without consensus. J. ACM 58(2), 7 (2011)CrossRefMathSciNetGoogle Scholar
  12. 12.
    Chockler, G.V., Keidar, I., Vitenberg, R.: Group communication specifications: a comprehensive study. ACM Comput. Surv. 33(4), 427–469 (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Leander Jehl
    • 1
  • Hein Meling
    • 1
  1. 1.Department of Electrical Engineering and Computer ScienceUniversity of StavangerNorway

Personalised recommendations