Optimal switch-aided flooding operations in ATM networks

We continue our previous work on switch-aided flooding (SAF) for use in ATM networks. The SAF-based protocols take advantage of hardware-supported cell relay and cell duplication in order to reduce the time needed to disseminate changes in network topology and resource availability. The SAF protocols use a spanning multipoint connection (SMC), which is a hardware-switched network spanning tree, but revert to conventional link-by-link flooding when the SMC is unavailable or under construction. The SAF protocol developed here further constructs a virtual ring topology to provide reliability more efficiently than our earlier protocols. The result is a robust flooding protocol that is guaranteed to deliver a broadcast message to all network switches reachable from the source node, and is (complexity-wise) as efficient as unreliable, hardware-based broadcast. The performance of the protocol in the presence of adverse network topology change events is studied via simulation. The results of the simulation reveal that the proposed flooding protocol delivers network updates several times faster than conventional approaches in normal situations, and is still, on average, twice as fast as conventional approaches in presence of component failures. Furthermore, the bandwidth consumed by a flooding operation is also significantly reduced