This Article 
 Bibliographic References 
 Add to: 
How to Choose a Timing Model
October 2008 (vol. 19 no. 10)
pp. 1367-1380
When employing a consensus algorithm for state machine replication, should one optimize for the case that all communication links are usually timely, or for fewer timely links? Does optimizing a protocol for better message complexity hamper the time complexity? In this paper, we investigate these types of questions using mathematical analysis as well as experiments over PlanetLab (WAN) and a LAN. We present a new and efficient leader-based consensus protocol that has O(n) stable-state message complexity (in a system with n processes) and requires only O(n) links to be timely at stable times. We compare this protocol with several previously suggested protocols. Our results show that a protocol that requires fewer timely links can achieve better performance, even if it sends fewer messages.

[1] M.K. Aguilera, C. Delporte-Gallet, H. Fauconnier, and S. Toueg, “Stable Leader Election,” Proc. 15th Int'l Symp. Distributed Computing (DISC), 2001.
[2] T. Anker, D. Dolev, G. Greenman, and I. Shnayderman, “Evaluating Total Order Algorithms in WAN,” Proc. Int'l Workshop Large-Scale Group Comm. (IWLSGC), 2003.
[3] O. Bakr and I. Keidar, “Evaluating the Running Time of a Communication Round over the Internet,” Proc. ACM Symp. Principles of Distributed Computing (PODC), 2002.
[4] A. Bavier, M. Bowman, B. Chun, D. Culler, S. Karlin, S. Muir, L. Peterson, T. Roscoe, T. Spalink, and M. Wawrzoniak, Operating System Support for Planetary-Scale Network Services. Princeton Univ. Press, 2004.
[5] N. Cardwell, S. Savage, and T. Anderson, Modeling the Performance of Short TCP Connections, 1998.
[6] T.D. Chandra, V. Hadzilacos, and S. Toueg, “The Weakest Failure Detector for Solving Consensus,” J. ACM, vol. 43, no. 4, pp. 685-722, July 1996.
[7] H. Chernoff, “A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on a Sum of Observations,” Annals of Math. Statistics, vol. 23, pp. 493-507, 1952.
[8] F. Cristian and C. Fetzer, “The Timed Asynchronous Distributed System Model,” IEEE Trans. Parallel and Distributed Systems, June 1999.
[9] D. Dobre, M. Majuntke, and N. Suri, “CoReFP: Contention-Resistant Fast Paxos for WANs,” technical report, TU Darmstadt, 2006.
[10] P. Dutta and R. Guerraoui, “Fast Indulgent Consensus with Zero Degradation,” Proc. Fourth European Dependable Computing Conf. (EDCC '02), Oct. 2002.
[11] P. Dutta, R. Guerraoui, and I. Keidar, “The Overhead of Consensus Failure Recovery,” Technical Report 200456, Ecole Polytechnique Federale de Lausanne (EPFL), 2004.
[12] P. Dutta, R. Guerraoui, and I. Keidar, “The Overhead of Indulgent Failure Recovery,” Distributed Computing, 2006.
[13] P. Dutta, R. Guerraoui, and L. Lamport, “How Fast Can Eventual Synchrony Lead to Consensus?” Proc. Int'l Conf. Dependable Systems and Networks (DSN '05), pp. 22-27, 2005.
[14] C. Dwork, N.A. Lynch, and L. Stockmeyer, “Consensus in the Presence of Partial Synchrony,” J. ACM, vol. 35, no. 2, pp. 288-323, Apr. 1988.
[15] A. Fernandez, E. Jimenez, and M. Raynal, “Eventual Leader Election with Weak Assumptions on Initial Knowledge, Communication Reliability, and Synchrony,” Proc. Int'l Conf. Dependable Systems and Networks (DSN), 2006.
[16] R. Guerraoui, “Revisiting the Relationship between Non Blocking Atomic Commitment and Consensus Problems,” Proc. Ninth Int'l Workshop Distributed Algorithms (WDAG '95), pp. 87-100, Sept. 1995.
[17] R. Guerraoui, “Indulgent Algorithms,” Proc. ACM Symp. Principles of Distributed Computing (PODC), 2000.
[18] R. Guerraoui and M. Raynal, “The Information Structure of Indulgent Consensus,” IEEE Trans. Computers, vol. 53, no. 4, pp.453-466, 2004.
[19] I. Keidar and A. Shraer, “Timeliness, Failure-Detectors, and Consensus Performance,” Proc. ACM Symp. Principles of Distributed Computing (PODC), 2006.
[20] L. Lamport, “The Implementation of Reliable Distributed Multiprocess Systems,” Computer Networks, vol. 2, 1978.
[21] L. Lamport, “The Part-Time Parliament,” ACM Trans. Computer Systems, vol. 16, no. 2, pp. 133-169, May 1998.
[22] M. Larrea, A. Fernández, and S. Arévalo, “Optimal Implementation of the Weakest Failure Detector for Solving Consensus,” Proc. Symp. Reliable Distributed Systems (SRDS '00), pp. 52-59, 2000.
[23] N. Lynch and M. Tuttle, “An Introduction to Input/Output Automata,” CWI Quarterly, vol. 2, no. 3, pp. 219-246, 1989.
[24] D. Malkhi, F. Oprea, and L. Zhou, “Omega Meets Paxos: Leader Election and Stability without Eventual Timely Links,” Proc. Int'l Symp. Distributed Computing (DISC '05), pp. 199-213, Sept. 2005.
[25] U. Schmid, “Failure Model Coverage under Transient Link Failures,” Research Report 2/2004, Technische Univ. Wien, 2004.
[26] P. Urban, I. Shnayderman, and A. Schiper, “Comparison of Failure Detectors and Group Membership: Performance Study of Two Atomic Broadcast Algorithms,” Proc. Int'l Conf. Dependable Systems and Networks (DSN), 2003.

Index Terms:
Distributed Systems, synchrony assumptions, eventual synchrony, failure detectors, consensus algorithms, FT Middleware
Idit Keidar, Alexander Shraer, "How to Choose a Timing Model," IEEE Transactions on Parallel and Distributed Systems, vol. 19, no. 10, pp. 1367-1380, Oct. 2008, doi:10.1109/TPDS.2008.16
Usage of this product signifies your acceptance of the Terms of Use.