Threshold-Crossing Theory Based Service Requests Analysis Method

The network service providers always prepare hardware design, maintenance and upgrading action plan via experience. This usually leads to large amount of waste resources. Although many complex models have been introduced into the research held of high performance network QOS evaluation, it is hard for such method to be applied into production environment because of the random and complex character of the network service requests and the heavy calculation tasks brought by the models. The threshold-crossing theory based service requests analysis method put forward in this paper is able to determine the necessary concurrent service requests that the server and network hardware have to handle based on the previous service requests data in production environment. This method can be applied into designing, maintaining and upgrading network equipment and provide basis for reducing resource waste, avoiding unnecessary investment and optimizing resources.