Dynamic pricing as congestion control in ATM networks

Several researchers have advocated dynamic pricing mechanisms like the smart market. This paper explores how dynamic state-dependent pricing and explicit congestion control can both be used to avoid and alleviate congestion. We show that dynamic pricing has significant advantages for heterogeneous traffic, although it reduces the raw throughput somewhat. When the propagation delay is non-trivial, a slow-reacting version of dynamic pricing is preferable. This paper also advocates the use of novel stream-oriented best-effort ATM services, with which a stream´s arrival process is declared to the network before transmission begins and then policed, although there are no performance guarantees and best-effort streams are never blocked. With this approach, applications have incentive to decrease traffic burstiness, and to reveal important information about their packet streams, making mechanisms like slow-reacting dynamic pricing more practical