Troubleshooting web sessions with CUSUM

A variety of factors may lead users to a poor quality of experience in a web browsing session, experiencing a high page load time. Without a clear explanation this can be annoying. In this paper, we present a novel algorithm and a whole redesigned architecture to provide an answer to the question “what´s wrong with this web site?”. In more detail, we propose the design and the implementation of a probe, running a novel diagnosis algorithm based on the original use of “classical” troubleshooting techniques merged together with statistical change point detection tools. Our proposed probe is able to correctly determine the root cause of poor web navigation experience, distinguishing, among the several portions of the network, the one responsible for the problem. The presented experimental results demonstrate the effectiveness of the proposed method.