Performance improvement of active queue management with per-flow scheduling

Abstract

Active queue management (AQM) has two main objectives. One is to achieve high throughput and low average queueing delay simultaneously. The other is to achieve fair bandwidth allocation among competing flows. Although many algorithms have been proposed and investigated in an effort to fulfil the goal of AQM, there are few studies on AQM with per-flow scheduling. This is because longest queue drop (LQD), which is commonly used as a queue management discipline with per-flow scheduling, has good performance in terms of throughput and fairness. However, LQD suffers from long queueing delays. In addition, it cannot support explicit congestion notification (ECN). Accordingly, for AQM with per-flow scheduling, the following requirements can be specified: it achieves simultaneously lower average queueing delays than LQD and the same throughput as LQD, it achieves the same degree of fairness as LQD and it achieves lower packet loss ratios than LQD when flows are ECN-capable. A new AQM is proposed that fulfils these requirements. In the mechanism, packets are dropped probabilistically before the buffer is full. The probability is maintained for each active flow. The probability of the flow with the longest queue length is increased when the network is congested. Simulations confirm that the mechanism achieves the requirements set for AQM with per-flow scheduling.