Measuring Vulnerability of Complex Networks by Simulating DDoS Attacks

Abstract

— In this paper we assess the vulnerability of different generic complex networks by measuring the throughput for networks with different load in presence of Distributed Denial-of-Service (DDoS) attacks. DDoS attacks are simulated by choosing a number of bot nodes using several measures, such as: random, degree centrality, eigenvector centrality, betweenness centrality and k-medoids clustering algorithm. In order to obtain some information about the vulnerability of the three different complex networks (random, small-world and scale-free) we analyze the useful throughput of these networks in the presence of DDoS attack by the bot nodes.