Analytical Study of Adversarial Strategies in Cluster-based Overlays

TL;DR

This paper analyzes adversarial strategies in cluster-based overlay networks, showing how malicious nodes can quickly disrupt DHT overlays or reach a stable state with bounded pollution under certain conditions.

Contribution

It introduces a game-based analysis of adversarial tactics and demonstrates the system's resilience and vulnerabilities in different attack scenarios.

Findings

Malicious nodes can quickly subvert DHT overlays by avoiding leave operations.

A system with limited node lifetime reaches a stable state with bounded pollution.

The ratio of polluted clusters remains bounded regardless of initial corruption.

Abstract

Scheideler has shown that peer-to-peer overlays networks can only survive Byzantine attacks if malicious nodes are not able to predict what is going to be the topology of the network for a given sequence of join and leave operations. In this paper we investigate adversarial strategies by following specific games. Our analysis demonstrates first that an adversary can very quickly subvert DHT-based overlays by simply never triggering leave operations. We then show that when all nodes (honest and malicious ones) are imposed on a limited lifetime, the system eventually reaches a stationary regime where the ratio of polluted clusters is bounded, independently from the initial amount of corruption in the system.