Generating Realistic Network Graph Models for Fault-Tolerant Algorithm Evaluation

Roman Messmer

Keywords

Network modeling, graph, similarity, fault-tolerance

Abstract

While networks like the Internet lack a central control and develop into a hub-based scale-free topology, current work demonstrates that most wide-area networks, which are under a central administration develop into small-world like topologies. The testing of fault-tolerant algorithms requires most realistic network models. In this work we show methods that produce such network graphs by working out adequate parameters for graph generators and introduce a new functional similarity index, which allows the selection of graphs that possess topologies with similar influence on algorithms as real existing graphs.

Important Links:



Go Back