Congestion Pricing: Paying Your Way in Communication Networks

N etwork congestion is a fundamental problem facing Internet users today. Many common complaints — “Why is this streaming video clip so jerky?” “Why are these Web pages taking so long to download compared to yesterday?” and “Why are my pings so high in Quake?” — usually reflect congestion somewhere in the network. Traditional approaches to congestion control have viewed the Internet as a cooperative network. Transport protocols such as TCP were designed such that sources would adapt to network congestion by backing off and thus allow all sources to continue sending through the network, but at a reduced rate. Such congestion control is successful, however, only if users agree to cooperate in the manner mandated by the protocol designers. This has worked so far, perhaps because implementing a new transport protocol entails kernel programming that is slightly harder than viruswriting or other antisocial activities. In addition, some Internet service providers (ISPs) have forced their user traffic to conform to TCP-like behavior, such as that described in RFC 2581.1 On the other hand, denial-of-service attacks and demands for higher quality of service (QoS) are just two cases where network users prefer to act in their own self-interest rather than cooperate with others. A network where users are selfish, and thus reluctant to defer to other users, may result in the famous “tragedy of the commons,” where — in the absence of controls — a shared resource is overconsumed by individuals who consider only their personal costs and not the cost to society as a whole.2 In terms of the Internet, the “tragedy” could be viewed as congestive collapse, resulting from overconsumption of the shared network resource.