Network Tomography and Improved Methods for Delay Distribution Inference

Network monitoring, prediction and diagnosis are of great importance for network administrators to manage their network. But direct measurement of network characteristics is sometimes infeasible. Network tomography comes out as an effective solution for this problem. It makes inference for network internal parameters with end-to-end measurement, and introduces probabilistic reasoning idea into network measurement. However, despite of this advantage, its computational overhead caused by the inverse inference is still a problem for practical use. In this paper, we make an overview of network tomography techniques and point out three methods to reduce computational complexity of delay distribution inference in large scale network monitoring.