A heterogeneous webserver such as an HTTP server should be able to currently deal with numerous users. To the end, it is inevitable to formally analyze web traffics as well as a webserver itself. In particular, as most systems adopt HTTP 1.1 protocol instead of HTTP 1.0 protocol, it is more difficult to represent the system as a simple analytic mode. In addition, since most of previous models missed the detailed processes of the server, it is unsuitable for the current server based on HTTP 1.1 to tune itself with its own system parameters. On the basis of HTTP 1.1 Protocol supporting persistent connections, we thus present an analytical end-to-end tandem queueing model considering specific hardware configurations inside the webserver, which ultimately covers from accepting the customer requests to completing the services.