Two-layer P2P system based on location

Shen Xin-peng, Leijie Zeng, Zhao Xiao-nan · 2010

Structured peer-to-peer network organize nodes into an overlay network. The distributed hash table is used to lookup the resource. In these systems, data is associated with keys and each node is responsible for a subset of the keys. The distance between two adjacent peers on the overlay network may be very far on IP network so that the network access speed is greatly reduced. To solve the problem, a two-layer P2P system is proposed. In the system, each node has a peer identifier and an area identifier. All nodes, who have the same area identifier, gather to an area P2P subnet. In the way, these nodes, which exist in the same subnet, are geographical neighbor nodes. When a node wants to get a resource, it executes the query in its subnet at first. The network delay of a hop is less than that of neighbor nodes in the layout network. So the performance is improved.

Read the paper · More papers on PaperTik