The Implementation of a Distributed System Based on a Parallel Algorithm for Self-Similar Network Traffic Simulation

With the development of Web application, network traffic is constantly increasing. In order to do the performance testing of Web applications, a precise simulation of network traffic is necessary. However, traditional traffic model based on Poisson process can not characterize the behavior of real network traffic. Measurements of network traffic show that the property of the self-similarity is ubiquitous in Web environment. Because self-similarity is a common characteristic in todaypsilas network traffic, simulations which can generate the same degree of self-similarity as the original traffic are demanded. But nowadays, network traffic simulation has limitation by single computer. So this paper proposes a high-speed parallel algorithm to generate self-similar traffic based on the fraction Gaussian noise (FGN) model, We also give an architecture of a distributed system for running the parallel algorithm. The results show that the network traffic generated by the distributed system has self-similar property.