Brocade: Landmark Routing on Overlay Networks

Peer-to-peer networks offer attractive features such as decentralized computing, fault tolerance, sharing and aggregating of resources (e.g., SETI@home). A common (and limiting) assumption on peer-to-peer network is that, in order to achieve fair sharing and fault tolerance, all computers in the network should be considered equal and there should be no server nor explicit structuring in the network. Many theories and algorithms used by existing peer-to-peer infrastructures (such as Tapestry [12], Chord [9], Pastry [7] and CAN [4]) comply with this tenet. However, such systems often show sub-optimal performance due to the asymmetry of nodes in reality and the lack of structure among them. In this paper, we argue that a peer-to-peer system does not have to operate in a pure peer-to-peer fashion to obtain the advantages mentioned above. We believe that such system can benefit from some form of internal organization and differentiation among its nodes, thus enjoying the efficiency of a well-organized system, while still providing users with the P2P features. As a proof of our belief, we proposed and simulated Brocade, a secondary overlay to be layered on top of these systems. Brocade exploits knowledge of underlying network characteristics and recognizes the differences among the nodes. Simulations were run on the GT-ITM [11] transit stub topology and showed encouraging results.