动态环境下的虚拟网故障诊断算法

On Virtual Network Fault Diagnosis Algorithm in Dynamic Environment

网络虚拟化的动态性特点容易造成网络模型知识不准确、网络信息传输噪声增大等问题,给故障诊断带来挑战。采用贝叶斯网对网络虚拟化环境下的故障诊断问题进行建模,提出一种动态环境下的虚拟网故障诊断算法,算法的主要过程包括时间信息处理、噪声处理、故障定位、计算后验概率和执行更新。仿真结果显示,该算法克服了网络虚拟化的动态性带来的问题,降低了诊断复杂度,提高了诊断性能。

The dynamic characteristics of network virtualization cause inaccurate network model knowledge, the increase of network information transmission noise, and bring a challenge to fault diagnosis. Using bayesian network to model the fault diagnosis problem in network virtualization environment, it puts forward a kind of virtual network fault diagnosis algorithm in dynamic environment, the main process of algorithm includes time information processing, noise processing, fault location,computing posteriori probability and update. The simulation result shows that the algorithm overcomes the dynamic problems of network virtualization, reduces the complexity, and improves the performance of diagnosis.