A mean-field model for multiple TCP connections through a buffer implementing RED

Abstract

Active queue management schemes like RED (Random Early Detection) have been suggested when multiple TCP sessions are multiplexed through a bottleneck buffer. The idea is to detect congestion before the buffer overflows and packets are lost. When the queue length reaches a certain threshold RED schemes drop/mark incoming packets with a probability that increases as the queue size increases. The objectives are an equitable distribution of packet loss, reduced delay and delay variation and improved network utilization. Here we model multiple connections maintained in the congestion avoidance regime by the RED mechanism. The window sizes of each TCP session evolve like independent dynamical systems coupled by the queue length at the buffer. We introduce a mean-field approximation to one such RED system as the number of flows tends to infinity. The deterministic limiting system is described by a transport equation. The numerical solution of the limiting system is found to provide a good description of the evolution of the distribution of the window sizes, the average queue size, the average loss rate per connection and the total throughput. TCP with RED or tail-drop may exhibit limit cycles and this causes unnecessary packet delay variation and variable loss rates. The root cause of these limit cycles is the hysteresis due to the round trip time delay in reacting to a packet loss.

Wireless Channel Parameters Maximizing TCP Throughput

“…The result in Corollary 3 coincides with [3], which is expected since letting C approach ∞ is equivalent to considering random transmission losses only.…”

An Extension of the TCP Steady-State Throughput Equation for Parallel Flows and Its Application in MulTFRC

“…From (2) and (5), we have (12) We assume that at the end of a TD-period , flows experiencing loss in that TDP have the window size , and other flows that experience loss in the previous loss event have the window size , etc. The mean value of the window size of the cumulative flow is (13) From (11)- (13), we have (14) From (10), assuming that a loss occurs independently distributed over the size of the cumulative window in a loss round, hence , we have (15) and including (6), (8), (11), (12), and (14) (16) Solving this equation for , we get (17), shown at the bottom of the page, and including (14), we get (18), shown at the bottom of the page. From (1), (2), (6), (8), and (17), we have (19), shown at the bottom of the page.…”

Stability of a rate control system with averaged feedback and network delay

“…Kelly's model adopts a fluid model, and the feedback signal generated by a resource is a function of users' instantaneous rates. A fluid model is shown to be a good approximation when the number of flows is large and resource capacities are high [2], [13], [14], [15]. In practice, however, an instantaneous rate through a resource is difficult to measure accurately; in order to obtain a good estimate of the aggregate rate, a resource may need to rely on an average rate over a period.…”