Exposing and Eliminating Vulnerabilities to Denial of Service Attacks in Secure Gossip-Based Multicast

Abstract

We propose a framework and methodology for quantifying the effect of denial of service (DoS) attacks on a distributed system. We present a systematic study of the resistance of gossip-based multicast protocols to DoS attacks. We show that even distributed and randomized gossip-based protocols, which eliminate single points of failure, do not necessarily eliminate vulnerabilities to DoS attacks. We propose Drum-a simple gossip-based multicast protocol that eliminates such vulnerabilities. Drum was implemented in Java and tested on a large cluster. We show, using closed-form mathematical analysis, simulations, and empirical tests, that Drum survives severe DoS attacks.

A computational intelligence method for traversing dynamically constructed networks of knowledge

“…Group based multicast systems such as described in [12] attempt to address the problems that face mobile devices that are intermittently attached to a network by building a hierarchy as receivers join the multicast group. Servers are connected to other servers higher in the hierarchy, and eventually back to the source with delivery from servers to children through unicast reliable protocols.…”

Efficient security mechanisms for overlay multicast based content delivery

“…Badishi et al [4] proposed a gossip-based multicast protocol called Drum, which combines multiple techniques such as push, pull, random port selections, and resource bounds, for mitigating DoS attacks in secure gossip-based multicast. Wright et al [26] presented k-redundant depender graphs for distributing public-key certificate revocation lists (CRLs), which provides every node in the graph with k disjoint paths to the root of the graph, thus guaranteeing delivery even when up to kÀ1 paths between them have failed.…”

Localized Techniques for Broadcasting in Wireless Sensor Networks

“…To estimate the probability that at least two choices fail, note that there are less than (a log(n)) 2 2 pairs of times during the procedure at which the choices can fail, and for each such pair the probability of failure in both is at most ( 2a log(n) n/2k 2 ) 2 (assuming as before the event in Lemma 3.1). The probability that at least two of the 2a log(n) choices are failures is therefore at most (a log(n)) 2 2…”

“…Let A denote the event "G n r,c contains at least two distinct 1 8k 2 -giant components". Note that in view of Proposition 2 with the estimate s = 1 8k 2 (11) that comes out of its proof, the only thing that can cause (2) to go wrong is if there exists an ε > 0 such that…”

“…2 We denote as visibility graph G r the graph in which there is a vertex for each sensor node, and an edge between any two neighboring nodes (i.e., between the nodes within each other transmission radius). 3 In the following we will make the approximation that energy is consumed only when receiving a packet addressed to the node.…”