Axiomatizing Congestion Control

Abstract

The overwhelmingly large design space of congestion control protocols, along with the increasingly diverse range of application environments, makes evaluating such protocols a daunting task. Simulation and experiments are very helpful in evaluating the performance of designs in specific contexts, but give limited insight into the more general properties of these schemes and provide no information about the inherent limits of congestion control designs (such as, which properties are simultaneously achievable and which are mutually exclusive). In contrast, traditional theoretical approaches are typically focused on the design of protocols that achieve to specific, predetermined objectives (e.g., network utility maximization), or the analysis of specific protocols (e.g., from control-theoretic perspectives), as opposed to the inherent tensions/derivations between desired properties. To complement today's prevalent experimental and theoretical approaches, we put forth a novel principled framework for reasoning about congestion control protocols, which is inspired by the axiomatic approach from social choice theory and game theory. We consider several natural requirements ("axioms'') from congestion control protocols -- e.g., efficient resource-utilization, loss-avoidance, fairness, stability, and TCP-friendliness -- and investigate which combinations of these can be achieved within a single design. Thus, our framework allows us to investigate the fundamental tradeoffs between desiderata, and to identify where existing and new congestion control architectures fit within the space of possible outcomes.