Evaluation of congestion control protocols for ABR traffic over ATM networks

Abstract

SUMMARYCongestion control is very important for e!ective and stable operation of ATM (Asynchronous Transfer Mode) networks. Owing to the bursty and unpredictable characteristic of data network tra$c, its congestion control is particularly a challenge for network researchers and designers. The ATM Forum has recently adopted rate-based congestion control for ABR (Available Bit-Rate) tra$c which is the service class de"ned for data network applications. However, there is a number of congestion control schemes prevalent. ATM Forum has decided not to specify switch behaviour for ABR tra$c; this has further introduced additional ambiguity. Consequently, an evaluation and comparison of the existing protocols would provide valuable guidance for network designers and engineers; it would also give insight for researchers to explore the essence of di!erent congestion control schemes. In the "rst part of this paper, we investigate the e!ectiveness of ABR congestion control in the presence of bursty source tra$c and the relationship between the burst time scale and the ABR control time scale. Two ABR congestion control schemes, the ABR Explicit Forward Congestion Indication (EFCI) and ABR Congestion Indication (CI) schemes, are compared with Unspeci-"ed Bit Rate (UBR) transport which makes no e!ort to control congestion. Tra$c sources of various burst lengths of 100, 1000, 10 000, and an equal mix of 100 and 10 000 ATM cells are used in simulations. It is found that ABR congestion control schemes e!ectively control low frequency, medium to long-term tra$c load transients. This is further supported by the result of integrating TCP over ATM congestion control schemes included in the paper. ABR control schemes do not control high frequency, short-term load transients well, but ABR control is not necessary in such cases since short-term transients do not require large amount of bu!ering. In the second part of this paper, we evaluate and compare six rate-based congestion control protocols including Scheme I: EFCI, Scheme II: EFCI with separate RM queues, Scheme III: CI, Scheme IV: CI with separate RM queues, Scheme V: the CAPC2 ER (Explicit Rate), and Scheme VI: the EFCI with utilization-based congestion indication. Each scheme is simulated and compared in the LAN, WAN, and GFC (General Fairness Con"guration) environments speci"ed by the ATM Forum. E!ects of varying VC (Virtual Circuits) number and changing endsystem}switch distance has been investigated. Their fairness is also compared using the GFC con"guration. We have found that ER control scheme performs signi"cantly better than the other "ve binary control schemes by its faster response to congestion, smoother regulation of bit-rates, lower queueing delay, shorter bu!er queue length, and fairness. Among the other "ve schemes, the CI scheme performs better than the EFCI scheme. Providing separate RM queues has signi"cantly improved the EFCI scheme in the WAN environment, but has little e!ect on the CI scheme.