Labels, Routing, and Capacity: Bringing Theoretical Networking Closer to Practice

In this work we study a variety of problems, including network coordinate systems, compact routing, and wireless network capacity. The unifying thread is the observation that while strong theoretical properties are already known about all of these problems, the standard models used to prove these properties are not particularly realistic. We attempt to work in more realistic models, which are either new (slack guarantees for distance problems) or old but difficult to work in (the geometric SINR model for wireless network capacity). Sometimes this involves getting around theoretical lower bounds by weakening the model, and in other cases it involves strengthening the model to make it more realistic, and thus becoming unable to use existing theoretical algorithms. While obviously we do not claim to have the final word on these problems, we view this as a step towards bringing theory closer to practice.