Similar articles

Abstract: Rapid developments in network systems of business service have resulted in more reliance on distributed computing, typified by "subscriber/push" architectures. Unfortunately, frequent and unexpectable network failures were routine, and downtime was not in hours, but in days. High availability has become the most important factor decreasing business risk and improving Quality of Service. Cluster technology has solved the non-stop problem on Local Area Network. However, most technologies including cluster today fail to ensure the non-stop Internet service based on Routers. With good performance on high availability and fault tolerance, quorum systems are very suitable for application to distributed business service networks. In this work, we modeled and developed a non-stop Internet service system based on a new quorum system, circle quorum system, for Boston Mutual Fund Broker, US. With five protocols, it provided highly available data services for clients on Internet.

[1] Ahamad, M., Ammar, M.H., 1980. Performance characterization of quorum-consensus algorithms for replicated data.IEEE Trans.Software Eng.,15(4):492-495.

[2] Fu, A.W.C., Wong, M., Yat, S., 2002. Diamond Quorum Consensus for High Capacity and Efficient in a Replicated Data System. Distributed and Parallel Databases, p.1-25.

[3] Kafri, N., Janecek, J., 2002. Dynamic Behavior of the Distributed Tree Quorum Algorithm. 22nd International Conference on Distributed Computing Systems(ICDCS'2002), Vienna, Austria.

[4] Kumar, A., 2002. An Efficient Super Grid Protocol for High Availability and Load Balancing. IEEE Transactions On Computer, p.1126-1133.

[5] Lee, C.M., Tam, A., Wang, C.L., 1998. Directed Point: An Efficient Communication Subsystem for Cluster Computing. Proceedings of the10th IASTED International Conference on Parallel and Distributed Computing and Systems, Las Vegas, Nevada, USA, p.662-675.

[6] Malkhi, D., 2000. Quorum Systems.In: The Encyclopedia of Distributed Computing, Joseph Urban and Partha Dasgupta editors. Kluwer Academic Publishers, Philip Drive Norwell, USA.

[7] Malkhi, D., Reiter, M., 2000. An architecture for survivable coordination in large distributed systems.IEEE Transactions on Knowledge and Data Engineering,12(2):187-202.

[8] Malkhi, D., Reiter, M.K., Alonso, G., Kemme, B., 2001. The load and availability of byzantine quorum systems.SIAM J. Computer,29(6):1889-1906.

[9] Malkhi, L., Alvisi, P., Reiter, E.M., 1999. Fault Detection for Byzantine Quorum Systems. Proceedings of 7th International Working Conference on Dependable Computing for Critical Applications, California, USA, p.357-372.

[10] Martin, J.P., Dahlin, M., 2002. Small Byzantine Quorum Systems. Proceedings of the International Conference on Dependable Systems and Networks (DSN 2002 and FTCS 32), DCC Track, Washington, DC, p.374-383.

[11] Mohan, R.B., Parmon, B.G., 1998. PARMON: A Comprehensive Cluster Monitoring System. The Australian Users Group for UNIX and Open Systems Conference and Exhibition, AUUG'98Open Systems: The Common Thread. Baulkham Hills, Australia.

[12] Peleg, D., Wool, A., 1997. Crumbling walls: a class of practical and efficient quorum systems.Distributed Computing,10(2):87-98.

[13] Peris, R.J., Martnez, P.M., Alonso, G., Kemme, B., 2001. How to Select A Replication Protocol According to Scalability, Availability, and Communication Overhead. Proc. of the Int. Symp. on Reliable Distributed Systems (SRDS), New Orleans, Louisiana.

[14] Tsuchiya, T., Kikuno, T., 2002. Byzantine quorum systems with maximum availability.InformationProcessing Letters,83:71-77.

[15] Yin, J., Martin, J.P., Alvisi, V., Dahlin, M., 2002. Fault-Tolerant Confidentiality (updated), International Workshop on Future Directions in Distributed Computing (IWFDDC'2002), Bertinoro, Italy.