Method and system for content distribution network

Abstract

The invention provides a content distribution network system and method. The system includes an edge service layer and an intelligent routing layer. The edge service layer includes a plurality of edge nodes, and the edge nodes can serve as access edge nodes of the user end, or as access parent nodes of the source station. The intelligent routing layer provides a routing scheduling policy to the edge service layer, and the edge service layer performs intelligent routing selection according to the routing scheduling policy.

Description

technical field [0001] The present invention relates to a content distribution network, in particular to a new framework for a content distribution network. Background technique [0002] With the popularization of cloud computing, cloud-based content has led to an even greater dependence on content distribution networks. The traditional content distribution network adopts the "edge + parent" architecture technology. The edge node (hereinafter referred to as "edge") is closer to the user terminal, so that the network resources required by the user can be cached nearby, and the user can obtain the required network resources nearby, reducing the resource acquisition time. The introduction of the parent node (hereinafter referred to as "parent") reduces the access volume of the origin station (hereinafter referred to as "source"). If the edge node does not cache the resources required by the user, the edge node does not need to go directly to the origin station to obtain resour...

AI Technical Summary

Problems solved by technology

The traditional content distribution network architecture does not optimize this uncertain network environment well, such as: optimizing routing, using private protocols, etc., resulting in more uncontrollable resource acquisition time

[0005] Secondly, in the traditional architecture, the distinction between the concepts of "edge" and "parent" reduces resource utilization to a certain extent, and at the same time increases the complexity of resource planning and deployment

The resource waste behavior is mainly manifested in two aspects: one is the waste of hardware resources. The "edge" and "parent" of the traditional architecture do not interoperate in hardware resource planning. Even if the "edge" is idle, it cannot be used as the "parent", unless " "Edge" is changed to "parent" in resource planning; "father" also has a situation similar to "edge", that is, "parent" can not be used as "edge" when it is idle, which is a waste of hardware resources; the second is cache Waste of resources, the "parent" of the traditional architecture cannot provide the nearest acceleration service for the client like the "edge", resulting in a waste of cache resources

Due to the deep-rooted traditional architecture, if you want to transform into a new architecture, you need to re-plan resources and re-adjust the configuration and deployment, which is also the difficulty

Images