Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.
Network tomography establishes linear relationships between the characteristics of individual links and those of end-to-end paths. It has been proved that these relationships can be used to infer the characteristics of links from end-to-end measurements, provided that links are not correlated, i.e., the status of one link is independent from the status of other links. In this paper, we consider the problem of identifying link characteristics from end-to-end measurements when links are "correlated," i.e., the status of one link may depend on the status of other links. There are several practical scenarios in which this can happen; for instance, if we know the network topology at the IP-link or at the domain-link level, then links from the same local-area network or the same administrative domain are potentially correlated, since they may be sharing physical links, network equipment, even management processes. We formally prove that, under certain well defined conditions, network tomography works when links are correlated, in particular, it is possible to identify the probability that each link is congested from end-to-end measurements. We also present a practical algorithm that computes these probabilities. We evaluate our algorithm through extensive simulations and show that it is accurate in a variety of realistic congestion scenarios.