Randomized Consensus in Wireless Environments: A Case Where More is Better

In many emerging wireless scenarios, consensus among nodes represents an important task that must be accomplished in a timely and dependable manner. However, the sharing of the radio medium and the typical communication failures of such environments may seriously hinder this operation. In the paper, we perform a practical evaluation of an existing randomized consensus protocol that is resilient to message collisions and omissions. Then, we provide and analyze an extension to the protocol that adds an extra message exchange phase. In spite of the added time complexity, the experiments confirm that our extension and some other implementation heuristics non-trivially boost the speed to reach consensus. Furthermore, we show that the speed-up holds also under particularly bad network conditions. As a consequence, our contribution turns out to be a viable and energy-efficient alternative for critical applications.

[1]  Miguel Correia,et al.  Intrusion Tolerance in Wireless Environments: An Experimental Evaluation , 2007, 13th Pacific Rim International Symposium on Dependable Computing (PRDC 2007).

[2]  Miguel Correia,et al.  Randomization Can Be a Healer: Consensus with Dynamic Omission Failures , 2009, DISC.

[3]  Mike Hibler,et al.  An integrated experimental environment for distributed systems and networks , 2002, OSDI '02.

[4]  Michael O. Rabin,et al.  Randomized byzantine generals , 1983, 24th Annual Symposium on Foundations of Computer Science (sfcs 1983).

[5]  Sy-Yen Kuo,et al.  Examining Web-Based Spyware Invasion with Stateful Behavior Monitoring , 2007 .

[6]  Nancy A. Lynch,et al.  Impossibility of distributed consensus with one faulty process , 1985, JACM.

[7]  James Aspnes,et al.  Randomized protocols for asynchronous consensus , 2002, Distributed Computing.

[8]  Gabriel Bracha,et al.  An asynchronous [(n - 1)/3]-resilient consensus protocol , 1984, PODC '84.

[9]  Nicola Santoro,et al.  Time is Not a Healer , 1989, STACS.

[10]  Victor Shoup,et al.  Random Oracles in Constantinople: Practical Asynchronous Byzantine Agreement Using Cryptography , 2000, Journal of Cryptology.

[11]  Miguel Correia,et al.  Experimental Comparison of Local and Shared Coin Randomized Consensus Protocols , 2006, 2006 25th IEEE Symposium on Reliable Distributed Systems (SRDS'06).

[12]  Michael Ben-Or,et al.  Another advantage of free choice (Extended Abstract): Completely asynchronous agreement protocols , 1983, PODC '83.