Routing begins to have an important place in the context of high performance distributed systems, with an increasingly notable influence on the overall performance of the system under analysis. While many global algorithms have been proposed for the routing problem, in this paper we demonstrate how a relatively simple agent-based approach, based on ideas inspired from complex systems and reinforcement learning, can generate a highly complex set of local interactions between individual agents, whose emergent behaviour results in the desired routing effect.
emergence, routing, intelligent agents, mobile computing
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).