A network content diffusion method based on distributed pagerank of network nodes

Abstract

The present invention provides a network content diffusion method based on the distributed Pagerank of network nodes, comprising: step 1) using each node in the network server as a central node to construct a subgraph of the connection relationship between the central node and the adjacent nodes of the network; Step 2) Calculate the Pagerank value between any two nodes in the sub-graph; Step 3) Calculate and update all Pagerank values ​​in the sub-graph according to the interactive operation between the two nodes in the same sub-graph; Step 4) Pass According to the set node diffusion rules, two nodes corresponding to the Pagerank values ​​that meet the rules are selected in each updated subgraph to diffuse the network content. The network content diffusion method of the present invention sets a node representing all other global information in the neighbor connection graph of each node, so as to continuously iterate and gradually approach the global information, thereby alleviating the problem of excessive differences between the local model and the actual situation, and improving the content Diffusion performance.

Description

technical field [0001] The invention relates to the technical field of computer networks, in particular to a network content diffusion method based on distributed Pagerank of network nodes. Background technique [0002] Currently, video and content services have become one of the main ways of Internet entertainment and occupy most of the network traffic. In order to ensure service quality, service providers usually use content distribution network, cloud service and other technologies to process large-scale user requests nearby and in a decentralized manner, thereby reducing the pressure on the center and improving processing efficiency. However, there are still some problems in such a structure, for example: the location of the data center is still far away from the user, and it is difficult to truly reflect "nearby"; the user resources in the network are huge, such as PCs, mobile phones, set-top boxes and other devices. These resources are still in an idle state, huge res...

AI Technical Summary

Problems solved by technology

However, there are still some problems in such a structure, for example: the location of the data center is still far away from the user, and it is difficult to truly reflect "nearby"; the user resources in the network are huge, such as PCs, mobile phones, set-top boxes and other devices. These resources are still in an idle state, the huge resources have not been rationally utilized, and the total amount of resources deployed is limited, and service bottlenecks will continue to appear

Existing methods pay more attention to local information, but there is no suitable method for approximating global information

Images