Controlling the Internet: a survey and some new results

In the past, significant progress has been made in the mathematical modelling of congestion control and congestion feedback mechanisms in the Internet. The resulting models have proved to be very useful in improving existing control and feedback mechanisms, and to make them scalable to networks that operate at very high speeds. Tools from convex optimization, control theory and stochastic processes have played a major role in the development of this Internet congestion control theory. In this paper, we focus on the control-theoretic aspects of the theory, and review some recent developments in the design of stable, scalable congestion control mechanisms. We also present a new scheme that can improve the performance of the Internet with minimal changes to the current architecture.