Fast-fair handling of flows. (2017)

• Record Type: Journal Article
• Authors: Tsioliaridou, Ageliki; Zhang, Chi; Liaskos, Christos; Tsaoussidis, Vassilios
• Is Part Of: International journal of communication networks and distributed systems. Volume 18:Number 1(2017)
• Page Start: 32
• Page End: 57
• Publication Date: 2017
• Subjects: fair handling -- fast resource allocation -- network congestion control -- fairness -- convergence speed -- bandwidth allocation -- network traffic flow -- simulation; Computer networks -- Periodicals; Telecommunication systems -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; 004.6
• Languages: English
• ISSNs: 1754-3916

Abstract

Allocating fairly the maximum available bandwidth among competing flows is an important problem as it allows maximising system throughput and fairness. Given the dynamic nature of traffic load in current internet and the short duration of most modern applications, fast convergence to fairness is necessitated. In this paper, we use fairness as the major criterion to adjust traffic and present a novel congestion control scheme which allows each flow independently to estimate efficiently: 1) its deviation from the fair-share; 2) adjust its window to its fair-share. In our model, no flow is aware of the number of competitors in the channel, no flow is aware of the bandwidth and buffer capacity; however, the system reaches to equilibrium state fast, based on throughput measurements. Simulation results confirm that the proposed congestion scheme promotes fast system fairness without damaging system efficiency and responsiveness.