As the Web application is world wide used, system' s performance, especially reliability, becomes more significant. Traditional performance testing tools such as QA Load and LoadRunner will generate the stress data with the fixed scale. But in the real time, network traffic is model-based. We focus on generating test data to simulate network traffic accurately for Web application reliability testing. The statistical results of network traffic show that the property of the self-similarity is ubiquitous in Web environment. So generating self-similar network traffic is demanded. But nowadays, there is a bottleneck in generating network traffic by single computer. We need a parallel method to solve this problem. In this paper we propose a distributed system based on a parallel algorithm to generate self-similar traffic using the Fraction Gaussian Noise (FGN) model. The experiment results show that the network traffic generated by the distributed system has self-similar property.