A network management system based on ontology and slow intelligence system

Abstract

The last decade has witnessed an intense spread of computer networks that has been further accelerated with the introduction of wireless networks: this growth has increased significantly the problems of network management. Especially in small companies the management of such networks is often complex and faults have significant impacts on their businesses. A possible solution is the adoption of the Simple Network Management Network administrators can manage network performance, find and solve network problems, and plan for network growth by the use of the SNMP. Over the past years much efforts has been given to make more effective the Simple Network Management Protocol and new approaches has been developed. In particular a promising approach involves the use of Ontology. The ontology based network management has recently evolved from a theoretical proposal to a more mature technology and this is the starting point of this paper where a novel approach to the network management based on the use of the Slow Intelligence System methodologies and Ontology based techniques is proposed. The Slow Intelligence System is a general-purpose system characterized by being able to improve performance over time through a process involving various phases as enumeration, propagation, adaptation, elimination and concentration. Therefore, the proposed approach aims to develop a system able to acquire, according to the Simple Network Management Protocol, information from the various hosts that are in the managed networks and apply actions in order to solve problems. To check the feasibility of this approach and its performance an experimental campaign in a real scenario has been designed and the first experimental results in a real scenario are showed.