Introducing Automated Reasoning in Network Management

This paper proposes the adoption of Artificial Intelligencetechniques in the field of network management and monitoring.In order to allow automated reasoning on networkingtopics, we constructed an accurate ontological modelcapable of fitting as more as possible networking concepts.The thoroughly representation of the domain knowledge isused by a Logical Reasoner, which is an expert system capableof performing management tasks typically executedby human experts. The Logical Reasoner is integrated ina distributed multi-agent architecture for network management,which exploits the dynamic reasoning capabilities ofthe Situation Calculus formalism to provide a powerful systemcapable of performing high-level management tasks inorder to deal with network fault situations. The system exploitsprogrammable network technology to make possiblethe deployment of code which implements teleo-reactiveagents, distributed across the whole network. The informationrelated to network events, generated by programmablesensors deployed on the network devices, is collected bythe logical entity where it is merged with general domainknowledge, with a view to identifying the root causes offaults, and to decide on reparative actions.