Adaptive routing method for a dynamic network

Abstract

A routing system and method utilizes a highly-adaptive, loop-free, distributed routing algorithm for dynamic networks. The basic, underlying method is neither a distance-vector nor a link-state method; the invention employs an algorithm which is one of a family of algorithms which are called “link reversal” algorithms. The protocol's reaction is structured as a sequence of diffusing computations, each computation consisting of a sequence of directed link reversals. This behavior is achieved, in part, through the use of a “physical or logical clock” to establish the temporal order of topological change events. A key concept in the protocol's design; is an attempt to decouple (to the greatest extent possible) the generation of far-reaching control message propagation from the dynamics of the network topology. These design characteristics make the protocol highly-adaptive, efficient and scalable-being best-suited for use in large, dynamic, bandwidth-constrained networks. In such networks, the protocol's reaction to link failures typically involves only a single pass of the distributed algorithm. The results of a simulation study indicate that for a given available bandwidth, as either the size of the network or the rate of topological changes increases, the performance the invention eventually exceeds that of ILS.