next up previous contents
Next: Packet-Level Traffic Generation Up: diss Previous: Overview   Contents

Related Work

$\textstyle \parbox{0.2in}{\mbox{}}$$\textstyle \parbox{5.25in}{\begin{singlespace}\textit{A scientific theory should be as simple as possible, but no simpler.}\end{singlespace}}$$\textstyle \parbox{0.2in}{\mbox{}}$
$\textstyle \parbox{5.25in}{\begin{flushright}--- \textsc{{Albert Einstein (1879--1955)}}\end{flushright}}$

$\textstyle \parbox{0.2in}{\mbox{}}$$\textstyle \parbox{5.25in}{\begin{singlespace}\textit{The greatest challenge to...
... is stating the problem in a way that will allow a solution.}\end{singlespace}}$$\textstyle \parbox{0.2in}{\mbox{}}$
$\textstyle \parbox{5.25in}{\begin{flushright}--- \textsc{{Bertrand Russell (1872--1970)}}\end{flushright}}$

This chapter presents an overview of the research literature relevant for realistic traffic generation. We consider two types of works. First, we discuss the body of literature that developed the concepts and techniques currently in use for generating synthetic traffic in simulations and testbed experiments. Second, we examine the Internet measurement literature that informs the discussion of what is meant by ``realistic'' traffic generation. Intuitively, synthetic traffic resembling Internet traffic can only be realistic if derived from measurements conducted from real network links. We could argue that any Internet measurement paper helps to gain a better understanding of the nature of the Internet and its traffic, being therefore relevant for realistic traffic generation. However, the sheer size of the Internet measurement literature makes a complete overview impractical, so we will restrict ourselves to the main works that had a direct impact on Internet traffic generation. It is also interesting to note that the most recent trend in the field of traffic generation is precisely to combine traffic measurement and generation into a single, coherent approach [HCJS$^+$01,LH02,SB04,HCSJ04].

Traffic generation for experimental networking research was identified as one of the key challenges in Internet modeling and simulation by Paxson and Floyd [PF95] in 1995. Interestingly, Floyd and Kohler [FK03] made a similar point in 2003, and argued that it was still difficult to conduct experiments with representative, validated synthetic traffic. While traffic measurement and Internet measurement in general have become increasingly popular in recent years, most studies are exploratory and provide little foundation to build traffic generators. This chapter provides an overview of the major works in the field of Internet traffic generation, considering first packet-level traffic generation and then source-level traffic generation. Other aspects of traffic generation, such as load scaling, incorporating network-dependencies and implementation issues are discussed at the end of the chapter.

next up previous contents
Next: Packet-Level Traffic Generation Up: diss Previous: Overview   Contents

Doctoral Dissertation: Generation and Validation of Empirically-Derived TCP Application Workloads
© 2006 Félix Hernández-Campos