On Applications of Cooperative Security in Distributed Networks

Many applications running on the Internet operate in fully or semi-distributed fashion including P2P networks or social networks. Distributed applications exhibit many advantages over classical client-server models regarding scalability, fault tolerance, and cost. Unfortunately, the distributed system operation also brings many security threats along that challenge their performance and reliability. In particular, faulty or misbehaving nodes cannot collude to subvert the system operation. This paper addresses the above threats by applying cooperative security techniques to relevant distributed systems in the Internet. Our goal is to present methods that allow the peers to bootstrap basic trust relationships at the time of joining a distributed network and remove the peers if trust is lost. We consider the specific security caveats of the analyzed systems, investigate the applicability of existing cooperative security-based protocols, and propose general design guidelines for cooperative-security protocol in described distributed systems.