A Fully Distributed Quorum Consensus Method with High Fault-Tolerance and Low Communication Overhead

The main objective of data replication in a distributed database system is to provide high data availability for transaction processing. Quorum consensus (QC) methods are commonly applied to managing replicated data. In this paper, we present a new QC method. The proposed QC method is highly fault-tolerant, and fully distributed (i.e., each site in a distributed system is equally weighted). Further, we can show that the proposed QC method has a low message overhead: 1. (1) In the best case, each transaction operation process needs only to communicate with Ω(√n) remote sites to get permission (n is the number of sites storing replicated copies of the manipulating data item). 2. (2) In the worst case, each transaction operation process may be forced to communicate with Ω(√n log n) remote sites due to site failures. We also compare our method with the existing QC methods.

[1]  Akhil Kumar,et al.  Cost and availability tradeoffs in replicated data concurrency control , 1993, TODS.

[2]  Akhil Kumar,et al.  A High Availability \sqrt{N} Hierarchical Grid Algorithm for Replicated Data , 1991, Inf. Process. Lett..

[3]  Mostafa H. Ammar,et al.  The Grid Protocol: A High Performance Scheme for Maintaining Replicated Data , 1992, IEEE Trans. Knowl. Data Eng..

[4]  Divyakant Agrawal,et al.  Efficient solution to the distributed mutual exclusion problem , 1989, PODC '89.

[5]  Divyakant Agrawal,et al.  An efficient and fault-tolerant solution for distributed mutual exclusion , 1991, TOCS.

[6]  Hector Garcia-Molina,et al.  Consistency in a partitioned network: a survey , 1985, CSUR.

[7]  Satish K. Tripathi,et al.  A fault-tolerant algorithm for replicated data management , 1992, [1992] Eighth International Conference on Data Engineering.

[8]  Maria E. Orlowska,et al.  An Optimal Voting Scheme for Minimizing the Overall Communication Cost in Replicated Data Management , 1996, J. Parallel Distributed Comput..

[9]  Hao Chen,et al.  An efficient method for mutual exclusion in truly distributed systems , 1994, 14th International Conference on Distributed Computing Systems.

[10]  Mamoru Maekawa,et al.  A N algorithm for mutual exclusion in decentralized systems , 1985, TOCS.

[11]  Hector Garcia-Molina,et al.  How to assign votes in a distributed system , 1985, JACM.

[12]  Satish K. Tripathi,et al.  Average message overhead of replica control protocols , 1993, [1993] Proceedings. The 13th International Conference on Distributed Computing Systems.

[13]  Akhil Kumar,et al.  Hierarchical Quorum Consensus: A New Algorithm for Managing Replicated Data , 1991, IEEE Trans. Computers.