Effects of variations of load distribution on network performance

TL;DR

This study examines how different network topologies, from random to scale-free, affect traffic performance, revealing that random networks outperform scale-free ones in transmission efficiency due to load distribution issues.

Contribution

It introduces a novel traffic generator and analyzes the impact of network topology on performance, highlighting the disadvantages of scale-free networks in traffic handling.

Findings

Random networks have higher transmission rates and packet delivery.

Scale-free networks exhibit worse performance despite shorter path lengths.

Load distribution and congestion are key factors affecting network performance.

Abstract

This paper is concerned with the characterization of the relationship between topology and traffic dynamics. We use a model of network generation that allows the transition from random to scale free networks. Specifically, we consider three different topological types of network: random, scale-free with \gamma = 3, scale-free with \gamma = 2. By using a novel LRD traffic generator, we observe best performance, in terms of transmission rates and delivered packets, in the case of random networks. We show that, even if scale-free networks are characterized by shorter characteristic-path- length (the lower the exponent, the lower the path-length), they show worst performances in terms of communication. We conjecture this could be explained in terms of changes in the load distribution, defined here as the number of shortest paths going through a given vertex. In fact, that distribu- tion is characterized by (i) a decreasing mean (ii) an increas- ing standard deviation, as the networks becomes scale-free (especially scale-free networks with low exponents). The use of a degree-independent server also discriminates against a scale-free structure. As a result, since the model is un- controlled, most packets will go through the same vertices, favoring the onset of congestion.