Um Algoritmo para Diagnóstico de Redes de Topologia Arbitrária
Abstract
There is a growing demand for network management systems that are capable of effectively diagnosing faults and performance problems. To achieve this goal, it is important that those systems themselves be fault-tolerant. In this work we present a system-level diagnosis algorithm for general topology networks, that can be applied for network fault management. The algorithm allows link fault diagnosis, after which nodes compute network connectivity. It employs the minimum number of tests, i.e. one per link per testing interval. The latency of the algorithm is proportional to the diameter of the graph corresponding to the network. This approach is fault-tolerant in the sense that no matter which portion of the network is faulty, the fault-free nodes keep on monitoring the network continuously.
References
E.P. Duarte Jr., G. Mansfield, T. Nanya, and 5. Noguchi, "Non-Broadcast Network Fault Monitoring Based on System-Level Diagnosis", Proc. IEEE/IFIP IM'97, pp.597-609, San Diego, May 1997.
M.T. Rose, The Simple Book - An Introduction to Internet Management, 2nd ed., Prentice-Hall, Englewood Cliffs, NJ, 1994.
A. Bagchi, and S.L. Hakimi, "An Optimal Algorithm for Distributed System-Level Diagnosis", Proc. 21st Fault Tolerant Computing Symp., June, 1991.
M. Stahl, R. Buskens, and R. Bianchini, "Simulation of the Adapt On-Line Diagnosis Algorithm for General Topology Networks", Proc. IEEE 11th Symp. Reliable Distributed Systems, October 1992.
S. Rangarajan, A. T. Dabbura, and E. A. Ziegler, "A Distributed System-Level Diagnosis Algorithm for Arbitrary Network Topologies", IEEE Transactions on Computers, Vol.dd, pp. 3212-333, 1995.
