MULTIHOP LIGHTWAVE NETWORKS - A COMPARISON OF STORE-AND-FORWARD AND HOT-POTATO ROUTING

Multihop networks with store-and-forward nodes were recently proposed and studied as possible architectures for realizing the vast capacity potential of multiuser lightwave networks subject to electro-optic speed constraints imposed at each access node. In this paper, we analytically determine the achievable aggregate capacity for a variant of the basic multihop approach in which minimum distance store-and-forward routing is replaced by a "hot-potato" routing algorithm. With hot-potato routing, all packets simultaneously arriving at a given node and not intended for reception at that node are immediately placed onto the outbound links leaving that node; if two or more packets contend for the same outgoing link to achieve minimum distance routing, then all but one will be misrouted to links which produce longer paths to the eventual destination. Our interest in studying such a scheme arises from the potential of optical pulse generation techniques to permit the formation of optical packets at data rates too high to permit electro-optic conversion, electronic packet routing, and store-and-forward buffering at each node. While not avoiding the electro-optic bottleneck for network access, each optical link would operate at a super-electronic data rate and contain time multiplexed packets from a multitude of sources. We confine our attention to the development of an analytical methodology for finding the probability distribution of the number of hops with hot potato routing for symmetric networks under uniform traffic load. From this, the expected number of hops and the aggregate network capacity can be found. Results show that the maximum throughput achievable with hot-potato routing can be as low as 25% of that for store-and-forward routing, and that the relative degradation increases as the number of nodes grows larger. This implies that the link speed-up needed to produce a significant overall capacity advantage with hot potato should be at least a factor of 10. The analytical methodology and results are quite general and applicability is not limited to lightwave networks.