Optimization Design Of Event Module And Research Of Fault Localization Algorithm Of SNMS Network Management System

With the development of computer technology and Internet, the network continues toexpand, so security and stability become particularly important. However, in the operation ofthe network, network elements will produce a variety of network events, these eventspotentially indicate the running state and behavior of elements, it’s an important gurantee fornetwork management to monitor these events effectively. Therefore, the establishment of asound network fault management system is very important to ensure efficient and stableoperation of computer network.By studying the primary demand and function of the network fault management, SNMPand EJB remote method calls and related technologies, this paper presents a design thatoptimizes the event module of the existing SNMS network management system, this designhas realized a fault localization algorithm based on network topology. This optimized systemis based on B/S structure and implements the browser interface. The technology used by thebrowser-side and server-side is jQuery and EJB, the communication technique between thesetwo is Ajax and Servlet. For alarm events occured in the network, a correlation analysis oftime and space domain is proposed in this paper to confirm the fault. The correlation analysisbased on time is about compressing, filtering, counting operation to reduce redundancy of thealarm information generated by the same devise within a certain time window. The correlationanalysis based on space is to confirm the real fault according to the realization of faultlocation based on network topology. This optimized system provides a visual operationinterface, it makes the network managers be able to check alert messages in time, check faultmessages that handle through correlation analysis, these alarm events can be deleted, or sendto network manager by E-mail and so on. About alarm events, this optimized system alsoprovides configuration parameters such as event category, event type, severity etc.The running results of the optimized system demonstrate that the optimized networkfault management sub-system can effectively support fault management of campus network,and the system has good extensibility and flexibility in deployment.