Modeling the interactions of congestion control and switch scheduling

Abstract

Abstract-In this paper, we study the interactions of user-based congestion control algorithms and router-based switch scheduling algorithms. We show that switch scheduling algorithms that were designed without taking into account these interactions can exhibit a completely different behavior when interacting with feedbackbased Internet traffic. Previous papers neglected or mitigated these interactions, and typically found that flow rates reach a fair equilibrium. On the contrary, we show that these interactions can lead to extreme unfairness with temporary flow starvation, as well as to large rate oscillations. For instance, we prove that this is the case for the MWM switch scheduling algorithm, even with a single router output and basic TCP flows. We also show that the iSLIP switch scheduling algorithm achieves fairness among ports, instead of fairness among flows. Finally, we fully characterize the network dynamics for both these switch scheduling algorithms.

Performance of gradient-based OFDMA subcarrier schedulers with TCP traffic

“…Both non-iterative and iterative QBMW rules show extremely unfair allocation between groups. This confirms the observation in [1], [3] that TCP sources in Group 1 which start earlier than those in Group 2 tend to have longer queue lengths due to larger congestion windows and hence are assigned more subcarriers, making them more advantageous than those sources in Group 2 who are starved of bandwidth.…”

“…Such fluctu ations are expected to be further amplified in OFDMA systems, which are usually composed of many subcarriers. Many OFDMA subcarrier scheduling algorithms have been proposed (e.g., [4], [5], [6]) and analyzed but few were evaluated with TCP traffic sources, while the studies in [1], [3] involve both schedulers and TCP but the sched ulers are considered as non-iterative schedulers which lead to overassignment of subcarriers. In this work, we extend the concept of iterative schedulers presented in [6] and evaluate their performance in the system involving TCP traffic sources.…”

“…Both TCP congestion control and a wireless channel scheduler at a base station are designed to give high performance when they are considered independently [1]. They share the same common goals as regulating the traffic in order to maximize channel utilization, minimize packet loss, and provide QoS which is basically con sidered as fairness among users.…”