Short-lived TCP traffic (e.g., web mice) composes the majority of the current Internet traffic. Accurate traffic modeling of a large number of short-lived TCP flows is extremely difficult due to (i) the interaction between session, transport, and network layers; and (ii) the explosion of the size of state space when the number of flows is large. Typically, ad-hoc assumptions are required for the analysis to be tractable under a certain regime.

We introduce a stochastic model of a bottleneck ECN/RED gateway under a large number of competing TCP flows. Our main result shows that as the number of flows becomes large, the queue dynamics and the aggregate traffic are simplified and can be accurately described by simple statistical recursions. These recursions can be evaluated independently of the number of flows, and hence the resulting traffic model is scalable. Furthermore, the limiting model is also consistent with other previously proposed models in their respective regime. Simulation results are also presented to confirm the results.