Research and Design on Web Server Optimization Model

This paper gives a new performance analysis method, which is through the Markov chain and queuing network model for Web server services thread and queue modeling, computing performance level coefficient, the different performance measurement data through the normalized integration, and presents an intuitive, quantitative results to match the best server configuration parameters. And on this basis we design a new Web server optimization model, with the aid of the historical experience and feedback mechanism to reconfigure the Web server and to achieve its performance adaptive optimization.