Recent measurements of local‐area and wide‐area traffic have shown that network traffic exhibits variability at a wide range of scales. What is striking is the ubiquitousness of the phenomenon, which has been observed in diverse networking contexts, from Ethernet to ATM, LAN and WAN, compressed video, and HTTP‐based WWW traffic. Such scale‐invariant variability is in strong contrast to traditional models of network traffic, which show burstiness at short time scales but are essentially smooth at large time scales; that is, they lack long‐range dependence. Since scale‐invariant burstiness can exert a significant impact on network performance, understanding the causes and effects of traffic self‐similarity is an important problem.
Still in its infancy, is the problem of controlling self‐similar network traffic. By the control of self‐similar traffic, we mean the problem of modulating traffic flow such that network performance including throughput is optimized.
In this chapter, we show the feasibility of “predicting the future” under self‐similar traffic conditions with sufficient reliability such that the information can be effectively utilized for congestion control purposes. First, we show that long‐range dependence can be on‐line detected to predict future traffic levels and contention at time scales above and beyond the time scale of the feedback congestion control. Second, we present a traffic modulation mechanism based on multiple time scale congestion control framework (MTSC) and show that it is able to effectively exploit this information to improve network performance, in particular, throughput. The congestion control mechanism works by selectively applying aggressiveness using the predicted future when it is warranted, throttling the data rate upward if the predicted future contention level is low, being more aggressive the lower the predicted contention level. We show that the selective aggressiveness mechanism is of benefit even for short‐range‐dependent traffic; however, being significantly more effective for long‐range dependent traffic, leading to comparatively large performance gains. We also show that as the number of connections engaging in selective aggressiveness control (SAC) increases, both fairness and efficiency are preserved. The latter refers to the total throughput achieved across all SAC‐controlled connections.
The chapter is organized as follows. We give a brief overview of self‐similar network traffic and the specific setup employed in this chapter. We describe the predictability mechanism and its efficacy at extracting the correlation structure present in long‐range dependent traffic. This is followed by a description of the SAC protocol and a refinement of the predictability mechanism for on‐line, per‐connection estimation. We show performance results of SAC and show its efficacy under different long‐range dependence conditions and when the number of SAC connections is varied. We conclude with a discussion of current results and future work.
Congestion Control for Self‐Similar Network Traffic
2000-08-21
34 pages
Article/Chapter (Book)
Electronic Resource
English
Adaptive Traffic Congestion Control of Traffic Network
British Library Conference Proceedings | 1993
|Adaptive traffic congestion control of traffic network
Tema Archive | 1993
|European Patent Office | 2022
|Managing traffic congestion by using traffic control
IET Digital Library Archive | 1995
|