A CDN edge injection distribution method and system

Abstract

The invention discloses a CDN edge injection and distribution method and system, and relates to the technical field of CDN services. The method includes an injection process and a distribution process. The injection process includes the following steps: sending content information to be uploaded to the nearest CDN through a CDN content push interface node; assign a content number to the content information to be uploaded, and inject the content to be uploaded according to the upload mode of the content information to be uploaded; the distribution process includes the following steps: injecting the distribution agent component to distribute the content to be distributed according to the corresponding distribution strategy, injecting the distribution agent component The distribution result is recorded in the corresponding distribution content node; the content number of the content to be distributed is recorded, and the distribution strategy and distribution feedback status of the corresponding content to be distributed are obtained. The present invention processes the distribution process by injecting a decentralized injection process and a distributed coordination service, thereby sharing the work load of the central node in the prior art and improving work efficiency.

Description

technical field [0001] The invention relates to the technical field of CDN services, in particular to a CDN edge injection distribution method and system. Background technique [0002] ZooKeeper is an indispensable part of the distributed file system HDFS and the distributed database HBASE in the big data platform Hadoop. It is also an open source, simple, and lightweight thin file system that provides high availability and high performance distributed Coordination service. It is often used to provide configuration services, distributed queues, distributed locks, node elections, etc. [0003] Generally, CDN injection refers to uploading the user's content to the CDN's central content repository, and distribution refers to pre-pushing the content injected into the center to regional or edge nodes. This center-region-edge injection distribution model is prone to the following problems: 1. Centralization is more prone to single-point problems. If there is a problem with the i...

AI Technical Summary

Problems solved by technology

This center-region-edge injection distribution model is prone to the following problems: 1. Centralization is more prone to single-point problems. If there is a problem with the injection of the central content library, it will cause all users to be unable to upload content and affect user experience. Especially for live broadcast services with relatively high operational requirements, more hardware equipment and network resources must be invested in building backup nodes to solve single points

2. In centralization, the network conditions between users and the center are not optimal. If the content uploaded by users is ultra-high-definition, 4K or even VR and other super-large content, it will greatly occupy and consume the bandwidth resources of the center and affect the efficiency of injection.

3. The distribution efficiency in this mode is not high. The injected content must be written to the database first, and then the signaling is organized and distributed to each node in a queue. The coupling degree is relatively high and the efficiency is low.

4. Due to the limitation of the bandwidth of the central node, generally speaking, there will be distribution flow control, which also indirectly affects the distribution efficiency

Images