An efficient method for mutual exclusion in truly distributed systems

Many operations in a distributed system require mutual exclusion to guarantee correctness. Quorum methods have been widely proposed for implementing mutual exclusion. Majority voting is the best known quorum method. It has the merit of simplicity, but may incur high message overhead. The tree algorithm is an efficient structured quorum method for mutual exclusion problems. The quorums generated by a tree algorithm are smaller on average than those generated by a majority voting. However, the tree algorithm enforces a highly biased treatment to the nodes at different levels. This affects its performance in a distributed system where the nodes have similar characteristics. We propose a new structured quorum method which treats the nodes more evenly than the tree algorithm yet still preserves a satisfactory availability. We believe that this method is desirable for implementing mutual exclusion in a truly distributed system.<<ETX>>

[1]  Leslie Lamport,et al.  The Implementation of Reliable Distributed Multiprocess Systems , 1978, Comput. Networks.

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

[3]  Mostafa H. Ammar,et al.  Performance Characterization of Quorum-Consensus Algorithms for Replicated Data , 1989, IEEE Trans. Software Eng..

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

[5]  Ichiro Suzuki,et al.  A distributed mutual exclusion algorithm , 1985, TOCS.

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

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

[8]  Robert H. Thomas,et al.  A Majority consensus approach to concurrency control for multiple copy databases , 1979, ACM Trans. Database Syst..

[9]  Michel Raynal,et al.  A Distributed Algorithm for Mutual Exclusion in an Arbitrary Network , 1988, Comput. J..

[10]  Chienwen Wu,et al.  The triangular lattice protocol: a highly fault tolerant and highly efficient protocol for replicated data , 1992, [1992] Proceedings 11th Symposium on Reliable Distributed Systems.

[11]  Mostafa H. Ammar,et al.  The grid protocol: a high performance scheme for maintaining replicated data , 1990, [1990] Proceedings. Sixth International Conference on Data Engineering.

[12]  Fred B. Schneider,et al.  Synchronization in Distributed Programs , 1982, TOPL.

[13]  David K. Gifford,et al.  Weighted voting for replicated data , 1979, SOSP '79.

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

[15]  Jian Tang,et al.  Obtaining Coteries That Optimize the Availability of Replicated Databases , 1993, IEEE Trans. Knowl. Data Eng..

[16]  J. D. Day,et al.  A principle for resilient sharing of distributed resources , 1976, ICSE '76.

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

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