Selection method of neighbor nodes related to physical topology in P2P system

Abstract

The invention relates to a selection method of neighbor nodes related to physical topology in a P2P system. The method comprises the following steps: 1) a group of landmark nodes is determined; 2) each landmark node is taken as a center to build a cluster, wherein, the cluster comprises cluster-head nodes and member nodes; the landmark nodes are cluster-head nodes, and delay time from the member nodes in the cluster to the cluster-head nodes is less than a predetermined threshold value; and the cluster-head nodes store an in-cluster delay table, and the in-cluster delay table records the delaytime from the member nodes in the cluster to the cluster-head nodes and sequentially arranges the delay time according to size of each delay time; 3) a certain solicited node requests a neighbor nodelist to the cluster-head nodes in the cluster of the node; and 4) the cluster-head nodes return the neighbor node list to the solicited node according to the in-cluster delay table. The method can help the P2P system improve data exchange efficiency, realize load balance and reduce maintenance expenditure.

Description

technical field [0001] The present invention relates to the field of computer network technology, in particular, the present invention relates to a node selection method related to physical topology in a P2P system. Background technique [0002] In recent years, with the rapid development of broadband communication and multimedia technology, various Internet applications such as online live broadcast, video on demand, and file download have emerged as the times require. New challenges, as the scale of users increases, the traditional client / server (C / S) model service system needs to consume more hardware and software resources, which can no longer meet the needs of large-scale users, so P2P-based services The system develops rapidly and gradually becomes a relatively mature application. [0003] From a functional point of view, a P2P system generally consists of two types of logical layers: 1) Overlay Layer, which mainly describes how nodes are organized in the P2P service ...

AI Technical Summary

Problems solved by technology

[0004] At present, most traditional P2P systems do not use effective methods to select actual adjacent nodes in the physical network when selecting cooperative nodes. Generally, tree or mesh topologies are used to construct the overlay network layer. When a new node joins the system, According to the service capabilities of other nodes in the network, the new node adopts a certain method to join the existing network, and the nodes are loosely connected by service capabilities, so if the selected nodes are far away in the physical network, the data The switching delay is large, and the data packets received by the node per unit time will be reduced. If the playback requirements cannot be met, the node will continue to request data from other nodes in the network, which increases the scheduling burden to a certain extent.

Images