A comparison of the Byzantine Agreement problem and the Transaction Commit Problem
Transaction Commit algorithms and Byzantine Agreement algorithms solve the problem of multiple processes reaching agreement in the presence of process and message failures. This paper summarizes the computation and fault models of these two kinds of agreement and discusses the differences between them. In particular, it explains that Byzantine Agreement is rarely used in practice because it involves significantly more hardware and messages, yet does not give predictable behavior if there are more than a few faults.
Unable to display preview. Download preview PDF.
- Babaoglu, O., "on The Reliability of Consensus Based Fault Tolerant Distributed Computer Systems", ACM TOCS, V. 5.4, 1987.Google Scholar
- Gray, J., "Notes on Database Operating Systems", Operating Systems, An Advanced Course, Lecture Notes in Computer Science, V. 60, Springer Verlag, 1978.Google Scholar
- Gray, J., "Why Do Computers Stop, and What Can We Do About It?", Tandem TR: 85.7, Tandem Computers, 1985.Google Scholar
- Lamport, L., Shostak, R., Pease, M., "The Byzantine Generals Problem", ACM TPLS, V. 4.3, 1982.Google Scholar
- Lampson, B.W., Sturgis, H., "Atomic Transactions", Distributed Systems Architecture and Implementation; An Advanced Course, Lecture Notes in Computer Science, V. 105, Springer Verlag, 1981.Google Scholar
- Pease, M., Shostak, R., Lamport, L., "Reaching Agreement in the Presence of Faults", ACM Journal, V. 27.2, 1980.Google Scholar
- Rosenkrantz, D.J., R.D. Stearns, P.M. Lewis, "System Level Concurrency Control for Database Systems", ACM TODS, V. 3.2, 1977.Google Scholar
- Tay, Y.C., "The Reliability of (k,n)-Resilient Distributed Systems", Proc. 4th Symposium in Distributed Software and Database Systems, IEEE, 1984.Google Scholar