Research On Distributed Network Fault Monitoring Technology

As the network now operates in a much larger scale and more flexible, sustaining a stable and reliable network becomes an urgent problem for both managers and engineers. Conventional network fault monitoring models are inadequate in multi-resource load and scalability. This paper designs and implements a distributed network fault monitoring model and core modules are deeply studied.Firstly, key technologies about fault monitoring are discussed, and advantages and disadvantages of conventional fault monitoring models are also analyzed. On this basis,with the ICE(Internet Communication Engine) technology, a dynamic distributed fault self monitoring named IDDM(ICE Dynamic Distributed Model) is designed.Then a scheduling algorithm for monitoring tasks is put forward and implemented, based on Artificial Neural Networks and dynamic load.And a dynamic loadable task module is achieved.Finally, a multithreading message processing model is accomplished, making use of the Leader/Follower design pattern.The experimental results show that the system achieves a stratified performance in reducing both system load and latency of response. The design of model is proved to be advanced and of practical value in terms of implementation methods and integrative utility.