Designing P2P networks tolerant to attacks and faults based on bimodal degree distribution

Recently, in contrast with the centralized networks (e.g., traditional client/server systems), the distributed networks such as Peer-to-Peer (P2P) networks and grid networks have attracted much attention due to their scalability. While the distributed networks have the advantage of allowing the node(s) to join or leave the network easily, the issue of lack of resiliency to both attacks and faults still remains. In this paper, we classify the existing distributed networks based on their degree distributions. Then, we demonstrate that they are not resilient to attacks and/or faults. For example, unstructured P2P networks, which have a power-law degree distribution, are vulnerable to attacks such as DOS. To address and resolve this issue, we propose a method to construct a network following bimodal degree distribution, which is robust to deal with both attacks and faults. Performance evaluation is conducted through computer simulations, which show that the proposed method can achieve higher resilience compared with other existing networking approaches.