Supernode selection mechanism based on location information
Yongqiong Zhu · Journal of Physics Conference Series · 2020
Abstract In a hierarchical p2p network such as Gnutella0.6, supernodes are selected high-capacity nodes. As the central server of leaf nodes, they process queries more efficiently than a single leaf node. So how to effectively select supernodes and construct a supernode network topology is a key issue in hierarchical P2P networks. Existing research shows that this mechanism of peers randomly selecting logical neighbors without any underlying physical connection information causes a mismatch between the P2P logical overlay network and the underlying physical network. The same message traverses the same physical connection multiple times. A lot of unnecessary network traffic is generated. Therefore, this paper proposes a super node selection mechanism based on distance information. When selecting a super node, the actual physical distance between the nodes must be detected, so that the selected super nodes are logically adjacent, and their physical locations are as close as possible, thereby improving the processing capacity of the network.