Cognitive Networks: Towards Self-Aware Networks

4.8. Conclusion

In this chapter, we have presented an overview of the concepts, theories and advances in self-managing networks, and discussed specific self-management problems. There is a strong tie between self-awareness and self-management, where self-awareness must be present to enable self-management and self-management in turn makes the system or network more self-aware. In fact, a large body of research in self-management deals with the problems of knowledge acquisition, representation, dissemination and analysis (Section 4.4). In the end, the road to self-management is full of difficult challenges both from general system designs and from the specific network management domains. In particular, we have brought forth two particular issues in self-management: benchmarking and validation; and the concept of self-stabilization. They will become major research focus in the coming years, because in contrast to the traditionally human-intensive management solutions, a much stronger emphasis must now be placed on system reliance and trustworthiness if the vision of self-management is to become a reality. Also because of this degree of reliance and trust we must place on self-managed systems, their design has to follow a rigorous and formal approach that emphasizes verifiable theories and formal design methodologies. The largely distributed nature of self-managing systems offers a rich ground for the exploration of novel theories that favor simple local interactions leading to complex ...