Intelligent real-time reactive Network Management

摘要

A variety of sophisticated tools already exist for monitoring and analyzing various parameters and events in large enterprise networks. However, not only is there the high overhead cost of continuously running these tools, but there is also the need for constant human monitoring and decision making on how to react to various critical events and alarms. Ideally, we would like to have a system that allows the administrator to encode using policy & rules, the knowledge needed to automatically invoke only when needed the right monitoring and analysis tools only on relevant systems. In this paper we propose a framework for achieving the above. Based on this, we have developed a prototype implementation that harnesses existing tools. An illustrative example of intelligent real-time reactive network management achieved using this tool is also described.