Abstract
Applications play an important role in the reliability evaluation of communication networks. In other words, the reliability of a network can be totally different when different applications are considered for the same network. However existing reliability evaluation methods, which are mostly based on the graph theory, give no or little consideration to applications. This paper proposes a concept of network application reliability and a Markov-based method for analyzing the proposed network application reliability measure. Furthermore, based on the reliability of each individual application, a method is proposed to evaluate the overall network reliability that incorporates effects of different applications running on the network. Both a case study and experiments are performed to illustrate the proposed concept and methods.