A hierarchical approach to autonomic network management

Recently, the autonomic communication networks paradigm has been introduced as a solution to the increasing management complexity of communication networks in the Future Internet. In order to encompass the large-scale nature of these networks, a general consensus has been reached that the supporting autonomic management architectures should be distributed for scalability reasons. However, several open issues related to the distribution of autonomic components remain to be solved. In this paper, we propose a novel approach to structuring distributed autonomic components in large-scale communication networks. The approach is generic and can be applied to many existing autonomic architectures and control loops. The autonomic components are structured in a hierarchy, which simplifies the interaction between components, and allows them to manage resources and govern child components in a more scalable manner. In addition to giving a detailed description of the hierarchical architecture, the advantages of the proposed approach are validated through analytical evaluation results.