In this paper we investigate the performances of the EFCI-based (Explicit F
orward Congestion Indication) and ER-based (Explicit Rate) algorithms for t
he rate-based flow control of the ABR (Available Bit Rate) traffic in an AT
M network. We consider the case of two switches in tandem. We present sever
al definitions of a bottleneck, and provide conditions that determine wheth
er the first, the second or both queues are bottleneck. We show that it is
not necessarily the queue with the slowest transmission rate in which conge
stion actually occurs. We derive analytic formulas for the maximum queue le
ngth. We compare our results to those obtained by approximating a network b
y a simpler one, containing only the bottleneck switch. We show that the ma
ximum queue lengths under the approximating approach may largely underestim
ate the ones obtained in the real network.