Improving Unresponsive Flow Control by Active Queue Management Algorithm

Abstract

Active queue management (AQM) algorithm is important to manage buffers and transfer packets for wired/wireless networks. However, the unresponsive flows to the network congestion control are dangerous to the network equilibrium and quality of service (QoS), especially from the point of network resource allocation. In some certain scenarios, the limited network resource (buffers) can be occupied by the unresponsive flows easily, which results in reduction of quality (RoQ). In this paper, a new AQM algorithm - Clue (compare and limited unresponsive flows) is proposed in order to strengthen the robustness of Internet against unresponsive flows. As a sort of scheduling algorithm, Clue relies on the detection and punishment of unresponsive flows and gets the elastics control of unresponsive flows, which benefit the buffer queue with the high performance. Via the comparison and evaluation experiments, it has been proved that Clue can detect and restrain unresponsive flows more accurately compared to other AQM algorithms.