Research And Application Of Network Management Based On Mobile Agent

As the development and popularization of Internet Technology, network scale is extending, network complexity is increasing and heterogeneity is enhanced .Under these circumstances, diversified network equipments should be adopted for the cunrrent network management system to process massive complicated data. The traditional centralized network management system based on C/S model couldn't meet the requirements; therefore the new network management and system architecture are developed. As a new distributed computing technology, Mobile Agent Technology is widely applied in E-business, network management, information collection & release based on web and parallel computing; especially network management based on Mobile Agent Technology has been researched by many people. Major contents in this paper are listed as following:(1) Mobile Agent related techniques, the current platform for Mobile Agent and the advantages of Mobile Agent in constructing the new distributed network management system, A network management model based on Mobile Agent; the work procedure and compare with the traditional model.(2) Location mechanism in Mobile Agent. The system is divided into different domins, network management station and domain cooperating is uesed to implement the Mobile Agent location service.The network management station is regarded as the location centre among network management system and all the Mobile Agents have to register to that when they are established or interdomin migrating. Then the Mobile Agent selects the first node as the agency's positioning server according to Mobile Agent's migratory path, and the node in this domian only need to record the corresponding position information. Experiments proved that the improved method could obtain a better effect.(3) Mobile Agent traveling routes. The Combination of Genetic and Ant Colony Algorithm (CGACA) based on ant colony algotithm is designed. The algorithm possesses the advantages of crossover and mutation in genetic algorithm. It can enhance the ability to search the most optimal solution in ant colony algorithm and accelerate convergence by incorporating crossover operator and mutation operator. Experiments testified that CGACA is more effective than the genetic algorithm and the ant colony algorithm.(4) The major threats Mobile Agent hosts face with at present and the causes of these threats and the security problems in Moble Agent.