Behavioral Analysis of IP QoS Management Mechanisms

International audienceThe Internet is willing to receive service convergence for the next generation networks. This new challenge will require from Internet infrastructures the ability to define and guarantee Quality of Services to multimedia services, in which each service may choose its own requirements. Modeling such network is not an easy task. Its dy-namicity and adaptive features require more intelligent behavior from network management. We argue that Multi-Agent Systems are able to model such dynamic environments. Additionally, it propitiates the design or employment of more intelligent and self-healing network man-agement mechanisms. This article presents a Behavioral Multi-Agent-based model for QoS-enabled Internet. Based on this behavioral ap-proach, we analyze network management mechanisms (or "elementary behaviors") in terms of performance and applicability profile. We use simulation to observe services performances when submitted to diverse QoS management elementary behaviors