DISC 2003: Distributed Computing pp 352-366 | Cite as

Overcoming the Majority Barrier in Large-Scale Systems

  • Haifeng Yu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2848)

Abstract

In asynchronous environments, the majority barrier prevents us from achieving consensus when more than half of the nodes fail. We argue that the majority barrier significantly limits the availability of practical systems such as peer-to-peer systems. To overcome the barrier realistically, this paper proposes a witness model that probabilistically strengthens the traditional asynchronous model for large-scale systems. The model is motivated by the prevalence of fuzzy partitions in today’s Internet, namely, reachability is far from transitive. Our assumptions in the model are strictly weaker than those in previous approaches. To show that the model is realistic, we carefully use real Internet measurements to validate our strengthening assumptions. On the other hand, we design three consensus protocols under the witness model that can tolerate n-1 failures, showing that the model is strong enough to overcome the majority barrier.

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

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Andersen, D., Balakrishnan, H., Kaashoek, F., Morris, R.: Resilient Overlay Networks. In: Proceedings of the 18th Symposium on Operating Systems Principles (SOSP) (October 2001)Google Scholar
  2. 2.
    Aspnes, J., Herlihy, M.: Fast Randomized Consensus Using Shared Memory. Journal of Algorithms 11(3), 441–461 (1990)MATHCrossRefMathSciNetGoogle Scholar
  3. 3.
    Aspnes, J., Waarts, O.: Randomized Consensus in Expected O(nlog 2 n) Operations per Processor. SIAM Journal on Computing 25(5), 1024–1044 (1996)MATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Attiya, H., Bar-Noy, A., Dolev, D.: Sharing Memory Robustly in Message-Passing Systems. Journal of the ACM, 124–142 (January 1996)Google Scholar
  5. 5.
    Attiya, H., Dwork, C., Lynch, N., Stockmeyer, L.: Bounds on the Time to Reach Agreement in the Presence of Timing Uncertainty. Journal of the ACM 41(1), 122–152 (1994)MATHCrossRefMathSciNetGoogle Scholar
  6. 6.
    Barbara, D., Garcia-Molina, H.: The Vulnerability of Vote Assignments. ACM Transactions on Computer Systems (August 1986)Google Scholar
  7. 7.
    Ben-Or, M.: Another Advantage of Free Choice: Completely Asynchronous Agreement Protocols. In: Proceedings of the 2nd Annual ACM Symposium on Principles of Distributed Computing, pp. 27–30 (1983)Google Scholar
  8. 8.
    Bracha, G., Toueg, S.: Asynchronous Consensus and Broadcast Protocols. Journal of the ACM, 824–840 (October 1985)Google Scholar
  9. 9.
    Chandra, B., Dahlin, M., Gao, L., Nayate, A.: End-to-End WAN Service Availability. In: Proceedings of the 3rd Usenix Symposium on Internet Technologies and Systems (January 2001)Google Scholar
  10. 10.
    Chandra, T., Toueg, S.: Unreliable Failure Detectors for Reliable Distributed Systems. Journal of the ACM, 225–267 (March 1996)Google Scholar
  11. 11.
    Chor, B., Israeli, A., Li, M.: On Processor Coordination Using Asynchronous Hardware. In: Symposium on Principles of Distributed Computing, pp. 86–97 (1987)Google Scholar
  12. 12.
    Cohen, R., Erez, K., ben Avraham, D., Havlin, S.: Resilience of the Internet to Random Breakdowns. Physical Review Letters 85(21) (November 2000)Google Scholar
  13. 13.
    Dwork, C., Lynch, N., Stockmeyer, L.: Consensus in the Presence of Partial Synchrony. Journal of the ACM 35(2), 288–323 (1990)CrossRefMathSciNetGoogle Scholar
  14. 14.
    Englert, B., Shvartsman, A.A.: Graceful Quorum Reconfiguration in a Robust Emulation of Shared Memory. In: Proceedings of the International Conference on Distributed Computer Systems, pp. 454–463 (2000)Google Scholar
  15. 15.
    Fischer, M., Lynch, N., Paterson, M.: Impossibility of Distributed Consensus with One Faulty Process. Journal of the ACM, 374–382 (1985)Google Scholar
  16. 16.
    Gafni, E., Lamport, L.: Disk Paxos. In: Proceedings of the International Symposium on Distributed Computing, pp. 330–344 (2000)Google Scholar
  17. 17.
    Gupta, I., Renesse, R.V., Birman, K.: A Probabilistically Correct Leader Election Protocol for Large Groups. In: Herlihy, M.P. (ed.) DISC 2000. LNCS, vol. 1914, p. 89. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  18. 18.
    Hennessy, J.: The Future of Systems Research. IEEE Computer 32(8), 27–33 (1999)Google Scholar
  19. 19.
    Lamport, L.: The Part-Time Parliament. ACM Transactions on Computer Systems 16, 133–169 (1998)CrossRefGoogle Scholar
  20. 20.
    Lee, E.K., Thekkath, C.A.: Petal: Distributed Virtual Disks. In: Proceedings of the 7th International Conference on Architectural Support for Programming Languages and Operating Systems (October 1996)Google Scholar
  21. 21.
    Lynch, N.: Distributed Algorithms. Morgan Kaufmann Publishers, San Francisco (1997)Google Scholar
  22. 22.
    Lynch, N., Shvartsman, A.: RAMBO: A Reconfigurable Atomic Memory Service for Dynamic Networks. In: Malkhi, D. (ed.) DISC 2002. LNCS, vol. 2508, Springer, Heidelberg (2002)CrossRefGoogle Scholar
  23. 23.
    Malkhi, D., Reiter, M., Wool, A., Wright, R.: Probabilistic Quorum Systems. The Information and Computation Journal 170(2) (November 2001)Google Scholar
  24. 24.
    Mostefaoui, A., Raynal, M.: Unreliable Failure Detectors with Limited Scope Accuracy and an Application to Consensus. In: Proceedings of the 19th International Conference on Foundations of Software Technology and Theoretical Computer Science (December 1999)Google Scholar
  25. 25.
    Muthitacharoen, A., Morris, R., Gil, T., Chen, B.: Ivy: A Read/Write Peer-to-peer File System. In: Proceedings of the 5th Symposium on Operating Systems Design and Implementation (December 2002)Google Scholar
  26. 26.
    Peleg, D., Wool, A.: The Availability of Quorum Systems. In: Information and Computation, pp. 210–223 (1995)Google Scholar
  27. 27.
    Rowstron, A., Druschel, P.: Pastry: Scalable, Distributed Object Location and Routing for Large-scale Peer-to-peer Systems. In: Proceedings of the 18th IFIP/ACM International Conference on Distributed Systems Platforms (Middleware 2001) (November 2001)Google Scholar
  28. 28.
    Saks, M., Shavit, N., Woll, H.: Optimal Time Randomized Consensus - Making Resilient Algorithms Fast in Practice. In: Proceedings of the Second Symposium on Discrete Algorithms, January 1991, pp. 351–362 (1991)Google Scholar
  29. 29.
    Savage, S., Anderson, T., Aggarwal, A., Becker, D., Cardwell, N., Collins, A., Hoffman, E., Snell, J., Vahdat, A., Voelker, G., Zahorjan, J.: Detour: A Case for Informed Internet Routing and Transport. IEEE Micro 19(1) (January 1999)Google Scholar
  30. 30.
    Stoica, I., Morris, R., Karger, D., Kaashoek, F., Balakrishnan, H.: Chord: A Scalable Peer-To-Peer Lookup Service for Internet Applications. In: Proceedings ofthe ACM SIGCOMM 2001, August 2001, pp. 149–160 (2001)Google Scholar
  31. 31.
    Yu, H.: Overcoming the Majority Barrier in Large-Scale Systems. Technical report, Duke University (2003), Technical Report CS-2003-05, Available at http://www.cs.duke.edu/~yhf/tr-2003-05.pdf.
  32. 32.
    Yu, H., Vahdat, A.: The Costs and Limits of Availability for Replicated Services. In: Proceedings ofthe 18th ACM Symposium on Operating Systems Principles (SOSP) (October 2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Haifeng Yu
    • 1
  1. 1.Computer Science DepartmentDuke UniversityDurhamUSA

Personalised recommendations