Quo Vadis - A Framework for Intelligent Routing in Large Communication Networks

Modern communication networks contain hundreds if not thousands of interconnected nodes. Traffic management mechanisms must be able to support a cost-effective, responsive, flexible, robust, customer-oriented high speed communication environment while minimizing the overhead associated with management functions. Conventional traffic management mechanisms for routing and congestion control algorithms entail tremendous resource overhead in storage and update of network state information. Quo Vadis is an evolving framework for intelligent traffic management in very large communication networks. It is designed to exploit topological properties of large networks as well as their spatio-temporal dynamics to optimize multiple performance criteria through cooperation among nodes in the network. It employs a distributed representation of network state information using local load measurements supplemented by a less precise global summary. Routing decisions in Quo Vadis are based on parameterized heuristics designed to optimize various performance metrics in an anticipatory or pro-active as well as compensatory or reactive mode and to minimize the overhead associated with traffic management. The complexity of modern networks in terms of the number of entities, their interaction, and the resulting dynamics make an analytical study often impossible. Hence, we have designed and implemented an object oriented simulation toolbox to facilitate the experimental studies of Quo Vadis. Our efforts to design such a simulation environment were driven by the need to evaluate heuristic routing strategies and knowledge representation as employed by Quo Vadis. The results of simulation experiments within a grid network clearly demonstrate the ability of Quo Vadis to avoid congestion and minimize message delay under a variety of network load conditions. In order to provide a theoretical framework for the design and analytical study of decision mechanisms as employed by Quo Vadis, we draw upon concepts from the field of utility theory. Based on the concept of reward and cost incurred by messages in the network, utility functions which bias routing decisions so as to yield routes that circumvent congested areas have been designed. The existence of utility functions which yield minimum cost routes in uniform cost networks with a single congested node has been proven rigorously.