SARED: A Self-Adaptive Active Queue Management Scheme for Improving Quality of Service in Network Systems

Aminu Adamu

doi:10.7494/csci.2021.22.2.4020

Abstract

Considering the phenomenal growth of network systems, congestion remains a threat to the quality of service provided in such systems, hence, research on congestion control is still relevant. Internet research community regards Active Queue Management (AQM) as an effective approach to address congestion in network systems. Most of the existing AQM schemes possess static drop patterns and lack self-adaptation mechanism, as such don’t work well for networks where traffic load fluctuates. This paper proposes Self-Adaptive Random Early Detection (SARED) scheme which smartly adapts its drop pattern based on current network’s traffic load in order to maintain better and stable performance. In light to moderate load conditions, SARED operates in nonlinear modes in order to maximize utilization and throughput, while in high load condition, it switches to linear mode in order to avoid forced drops and congestion. Experiments conducted have revealed that regardless of traffic load’s condition, SARED provides optimal performance.

Highlights

The Internet has experienced rapid growth in recent years; Cisco has forecasted that several billion devices will be connected to the Internet in the near future [6]
This paper proposes a selfa adaptive random early detection (SARED) scheme that smartly adapts its drop pattern based on a current network’s traffic load in order to maintain improved
Congestion is a key factor that affects the quality of service (QoS) that is provided in network systems [5, 12, 13]

Summary

Introduction

The Internet has experienced rapid growth in recent years; Cisco has forecasted that several billion devices will be connected to the Internet in the near future [6]. Karmeshu et al [14] believed that the use of avg to calculate dropping probability in terms of minth and maxth (as done in RED and some of its enhanced versions) is ineffective during heavy-load situations, since the maxth threshold will frequently be crossed (resulting in frequent packet dropping) As such, they proposed another AQM scheme called adaptive queue management with random dropping (AQMRD). Most AQM schemes use the computed average queue length (avg) as a congestion indicator; based on this, the packet-dropping probability is defined while ignoring the network’s traffic load, which directly affects the observed average queue length. The self-adaptive random early detection (SARED) algorithm is proposed, which considers the computed average queue length as a congestion indicator and the current traffic load; based on this information, the packet-dropping probability is defined. B Note that, with SARED, the load scenarios can be extended to more than three if desired

Withininlocraedase in load

Ear Packet drop probability maxp minth maxth avg

Fixed parameters

Load Scenario RED TRED SARED