A method of peer-to-peer content distribution

Abstract

The invention discloses a method for point-to-point content distribution, which comprises the steps of sending a file download request from a download terminal to an ETE-TRACKER server; as an OSPF server, the ETE-TRACKER server collects TE-LSAs on the network to obtain network topology and traffic matrix; The problem is assigned to the ETE-TRACKER server, and the detachable sub-problem is assigned to the distributed solution of the end system; the ETE-TRACKER server solves the remaining capacity problem, and the end system solves the detachable sub-problem, and merges and sends it to the ETE-TRACKER server; ETE-TRACKER The server transmits the calculation result to the SDN controller, and the SDN controller deploys the Send-Packet message to the openflow switch according to the link weight, and the openflow switch returns a confirmation message; the end system downloads the file and transmits the file to the download terminal. Adoption of the technical scheme of the present invention enables flexible configuration of the operation management of the P2P file sharing system, quickly provides routing and traffic distribution schemes in the face of large-scale problems, and takes into account both network energy consumption and link robustness.

Description

technical field [0001] The present invention relates to the technical field of content sharing, and in particular, to a method for point-to-point content distribution. Background technique [0002] In recent years, content sharing and distribution services have developed rapidly, including P2P download services based on file sharing and dissemination and CDN services for efficient video distribution, or a combination of the two. Its essential features are to break the simple traditional CS architecture, but take the overlay topology as the object of communication and control, and have a strong distributed feature for the server that provides download services. Due to the large proportion of traffic of this kind of business, more and more green network research literatures have focused on this kind of business in recent years. However, these studies generally have some problems. For example, the optimization method is single, only the energy consumption index is optimized, a...

AI Technical Summary

Problems solved by technology

However, these studies generally have some problems, such as a single optimization method, only optimizing energy consumption indicators, and using other indicators such as performance as constraints, ignoring other aspects such as the reliability of the existing network

Secondly, it is only optimized for the overlay layer, and less attention is paid to the network layer and other services, resulting in limited practicability

Third, in terms of business energy consumption optimization, more attention has been paid to end system PCs or PADs, and relatively less attention has been paid to energy consumption optimization of network equipment, and the degree of optimization is not thorough enough, etc.

[0003] Due to the conflict between energy saving and traffic engineering on whether the direction of traffic optimization should be converged as much as possible or evenly distributed, when the network topology, download requirements, seed nodes, etc. are all the same, peer selection and traffic distribution can significantly affect the reliability and performance of the network. consumption

Images