On the investigation of the joint use of self-* properties and Peer-To-Peer for network management

Abstract

Over the years, the network management community has been face to the need of designing new alternative management approaches able to support heterogeneity, scalability, reliability, and minor human intervention. Currently, there are two major alternatives that have been employed on the design and development of network management solutions. The first one uses autonomic computing and self-* properties, while the second one employs Peer-To-Peer (P2P) concepts and technologies. In the case of autonomic and self-* properties, there is a lack of investigations approximating the high level models to the management infrastructure, while P2P investigations suffer from the opposite problem. This thesis aims at bringing knowledge to issues involving the joint use of self-* properties and P2P. The major contributions of this thesis are: (i) a survey relating autonomic computing and self-* properties, P2P, and network and service management; (ii) the combination of techniques to explore parallel and cooperative behavior of management peers running the management algorithms; (iii) algorithms that embed managers into the managed environment instead of using managers on the borders of such environment; (iv) the change on the angle of network management solution development from morphological aspects (such as APIs, protocols, architectures, and frameworks) to the design of management algorithms.