Virtual visualization of distributed network traffic using fractals

Abstract

The fractal behavior on a distributed network is analyzed and simulated. The fractal behavior in a distributed network-traffic is revealed with respect to different time scales. The traffic tends to exhibit multi-fractal, mono-fractal behavior that accounts to its burstiness on the time scales. The burstiness over a wide range of time scales tends to exhibit self-similar behavior that can be represented by fractals which are the appropriate mathematical tools to describe self-recurring aspects of network traffic. These behavioral characteristics have direct impact on the network performance. The objective of this research is to develop a better understanding of fractal behavior on network traffic by representing the network traffic visually over a graph, which will in turn pave a pathway to have a better quality of service on the network in future. The study focuses on fractal analysis — a more fundamental and broader view of fractals compared to the traditional Poisson-based analysis model.