CN115567591A

CN115567591A - Content resource distribution method, content distribution network, cluster and medium

Info

Publication number: CN115567591A
Application number: CN202110744984.4A
Authority: CN
Inventors: 敬宜; 周岩; 周勇志; 王友飞; 韩宁
Original assignee: Huawei Cloud Computing Technologies Co Ltd
Current assignee: Huawei Cloud Computing Technologies Co Ltd
Priority date: 2021-06-30
Filing date: 2021-06-30
Publication date: 2023-01-03
Also published as: WO2023274267A1

Abstract

The application provides a content resource distribution method, which comprises the following steps: a first edge node in the CDN receives a content resource access request from a client, when the heat degree of a requested target content resource meets a preset condition, the first edge node sends a source returning request to a second edge node, the source returning request is used for requesting the target content resource, then the first edge node receives the target content resource sent by the second edge node, stores the target content resource, and returns a content resource access response carrying the target content resource to the client. The method returns the source based on the heat of the target content resource, optimizes the distribution of the target content resource in the CDN, avoids the content resource with lower heat from occupying a large amount of storage space of the edge node, improves the utilization rate of the storage resource in the edge node, and reduces the cost of the CDN for distributing the content resource.

Description

Content resource distribution method, content distribution network, cluster and medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a content resource distribution method, a content distribution network, a computer cluster, a computer-readable storage medium, and a computer program product.

Background

With the increasing scale of the Internet, the situation of network congestion also occurs. In order to avoid network congestion from affecting user experience, acceleration may be performed by a Content Delivery Network (CDN). Specifically, the CDN is configured to distribute content resources from the source station to an acceleration node close to the user, so that the user can obtain required content resources nearby, and response speed and success rate of user access are improved.

The CDN comprises a central node and edge nodes distributed in different areas (i.e. the acceleration nodes described above). When the edge node receives an access request of a content resource, and the edge node and its upper nodes (including direct and indirect upper nodes, such as a central node) do not store the corresponding content resource, the access request may be returned to the source. The source return means that the access request (for example, a source return request) is forwarded to the source station server, and the source station server responds to the source return request. The central node generally needs to aggregate back-source requests from each operator network, and therefore, the CDN generally adopts a multi-line access manner, which has high bandwidth overhead.

In order to reduce the bandwidth overhead caused by the back-sourcing of the edge node, the industry proposes to back-source a full-connection-peer network (full mesh network). Specifically, when an edge node receives an access request of a content resource, and the edge node does not store the corresponding content resource, the edge node returns to the source of the edge node (also referred to as a peer node) in the same hierarchy, thereby reducing the bandwidth overhead of the central node.

However, many content resources are accessed less often, i.e. the reuse rate of many content resources after downloading to the edge node is low. Therefore, the utilization rate of the storage resources of the edge node is low, and the cost of delivering the content resources by the CDN is increased.

Disclosure of Invention

The application provides a content resource distribution method. The method returns the source based on the heat degree of the target content resource, reduces the flow returning to the source of the central node on one hand, thereby reducing the bandwidth overhead of the central node, optimizes the distribution of the target content resource in the CDN, avoids the content resource with lower heat degree from occupying a large amount of the storage space of the edge node, improves the utilization rate of the storage resource in the edge node, and reduces the cost of distributing the content resource by the CDN. The application also provides a content distribution network, a computer cluster, a computer readable storage medium and a computer program product corresponding to the method.

In a first aspect, the present application provides a content resource distribution method. The method can be applied to the CDN. The CDN comprises a central node and a plurality of edge nodes. Wherein the plurality of edge nodes includes a first edge node and a second edge node. The first edge node and the second edge node may specifically be cache servers that cache content resources.

Specifically, the first edge node receives a content resource access request from a client, the content resource access request requesting a target content resource. And when the heat degree of the target content resource meets a preset condition, the first edge node sends a source returning request to the second edge node. The back source request is used to request the target content resource. And then the first edge node receives the target content resource sent by the second edge node, stores the target content resource and returns a content resource access response to the client. The content resource access response carries the target content resource.

In the method, the edge node can return to the source based on the heat degree of the target content resource, when the heat degree of the target content resource meets the preset condition, the source is returned to the peer node, on one hand, the flow returning to the source of the central node is reduced, and therefore the bandwidth overhead of the central node is reduced, on the other hand, when the heat degree meets the preset condition, the target content resource is stored in the edge node, the distribution of the target content resource in the CDN is optimized, the situation that a large amount of content resources with low heat degree occupy the storage space of the edge node is avoided, the utilization rate of the storage resources in the edge node is improved, and the cost of the CDN for distributing the content resources is reduced.

In some possible implementation manners, the condition that the heat degree of the target content resource satisfies the preset condition includes that the heat degree of the target content resource is not included in the local heat degree record, and the heat degree of the target content resource in the global heat degree record stored by the central node is greater than or equal to a first preset threshold. The heat of the target content resource in the global heat record stored by the central node may also be referred to as a central heat or a global heat. The global heat can be determined according to the number of back-source requests counted by the central node.

Therefore, the local heat record does not include the heat of the target content resource, and when the global heat is greater than or equal to a first preset threshold, the return to the source can be triggered. In particular, the central node may determine location information of a second edge node storing the target content resource. The first edge node may receive the location information of the second edge node sent by the central node before sending the back-to-source request to the second edge node, and then send the back-to-source request to the second edge node according to the location information. And the first edge node receives the target content resource sent by the second edge node, stores the target content resource and returns a content resource access response to the client.

The local heat records do not include the heat of the target content resource, the global heat is higher, the fact that the flow of other edge nodes such as the second edge node is larger is indicated, the copies of the target content resource are stored in the first edge node and the second edge node, the flow can be balanced, and the response efficiency is improved.

In some possible implementation manners, the condition that the heat degree of the target content resource meets the preset condition includes that the heat degree of the target content resource is included in the local heat degree record, and the heat degree of the target content resource is greater than or equal to a second preset threshold. The heat of the target content resource included in the local heat record stored by the edge node is also referred to as local heat.

Therefore, the local heat record comprises the heat of the target content resource, and when the local heat is greater than or equal to a second preset threshold, the return to the source can be triggered. Specifically, the local heat record further includes location information of the second edge node storing the target content resource, and the first edge node may send a request to return to the source to the second edge node according to the location information of the second edge node stored in the local heat record. And the first edge node receives the target content resource sent by the second edge node, stores the target content resource and returns a content resource access response to the client.

The local heat records comprise the heat of the target content resource, the local heat is high, the fact that the flow of the first edge node is large is indicated, the copy of the target content resource is stored in the first edge node, the content resource access request can be directly responded based on the copy, and response efficiency is improved.

In some possible implementations, the local heat record of the first edge node does not include the heat of the target content resource. The first edge node may send a back source request to the central node. And the central node determines the node storing the target memory resource according to the global heat record. The nodes storing the target memory resource include a second edge node. The first edge node may return location information of the second edge node to the client, so that the client acquires the target content resource from the second edge node.

The first edge node may return location information of the second edge node to the client when the global heat is less than a first preset threshold, so that the client acquires the target content resource from the second edge node. Therefore, the content resources with lower heat degree are prevented from additionally occupying the storage space of the edge node, and the utilization rate of the storage resources is improved.

In some possible implementation manners, the local heat record of the first edge node includes the heat of the target content resource and the location information of the node storing the target content resource, and the node storing the target content resource includes the second edge node. In this way, the first edge node may return the location information of the second edge node to the client, so that the client obtains the target content resource from the second edge node.

The first edge node may return location information of the second edge node to the client when the local heat is less than a second preset threshold, so that the client acquires the target content resource from the second edge node. Therefore, the content resources with lower heat degree are prevented from additionally occupying the storage space of the edge node, and the utilization rate of the storage resources is improved.

In some possible implementations, the local heat record of the first edge node does not include the heat of the target content resource. The first edge node may send a source return request to the central node, and the central node determines that the first edge node is the first source return according to the global heat record, and then returns the target content resource to the first edge node. The first edge node returns a content resource access response to the client. Therefore, the content resources with lower heat degree are prevented from additionally occupying the storage space of the edge node, and the utilization rate of the storage resources is improved.

In some possible implementations, the CDN comprises a central node and at least one edge node mutual group. The CDN may perform the mutual help back to the source or provide the mutual help service in units of mutual help groups. The mutual-aid return source is a mode different from the central return source, and is used for sending a return source request to the peer node so as to obtain a target content resource from the peer node and then responding according to the target content resource. The mutual aid service is to return the position information of the peer node to the client, so that the client can obtain the target content resource from the peer node. That is, the server node jumps to the peer node where the content resource is located and responds by the peer node.

Specifically, the edge node mutual group includes the first edge node and the second edge node. The first edge node may also synchronize the local heat record of the second edge node. In this way, the first edge node may determine whether to trigger the mutual aid return source or mutual aid service according to the synchronized local heat record (the local heat record of the edge node mutual aid group).

By taking the edge node mutual-aid group as a unit, the mutual-aid return source or mutual-aid service can be realized in a small range, so that the bandwidth overhead can be reduced, and the cost of the mutual-aid return source or mutual-aid service can be reduced.

In some possible implementations, the edge nodes in the edge node mutual group may have the same or related domain name, or belong to the same operator. Wherein nodes having the same or related domain names may be edge nodes of the same organization as the same company. The client has a high probability of accessing the same domain name or related domain names multiple times within a time period. Thus, grouping in the manner described above may improve the efficiency of a mutual-aid return source or mutual-aid service. In addition, the edge nodes between different operators have higher interaction cost, and the edge nodes belonging to different operators are divided into different groups, so that the interaction cost can be reduced, and the cost of mutual-help return source or mutual-help service is reduced.

In a second aspect, the present application provides a content distribution network. The network includes a central node and a plurality of edge nodes including a first edge node and a second edge node.

The first edge node is used for receiving a content resource access request from a client, wherein the content resource access request is used for requesting a target content resource; when the heat degree of the target content resource meets a preset condition, sending a source returning request to the second edge node, wherein the source returning request is used for requesting the target content resource;

the second edge node is used for sending the target content resource to the first edge node;

the first edge node is further configured to store the target content resource, and return a content resource access response to the client, where the content resource access response carries the target content resource.

In some possible implementation manners, the condition that the heat degree of the target content resource meets the preset condition includes that the heat degree of the target content resource is not included in a local heat degree record, and the heat degree of the target content resource in a global heat degree record stored by the central node is greater than or equal to a first preset threshold;

the first edge node is further configured to receive, before the first edge node sends a source back request to the second edge node, location information of the second edge node sent by the central node.

In some possible implementation manners, the fact that the heat degree of the target content resource meets the preset condition includes that the heat degree of the target content resource is included in a local heat degree record, and the heat degree of the target content resource is greater than or equal to a second preset threshold;

the local heat record also comprises the position information of the second edge node storing the target content resource.

In some possible implementations, the local heat record of the first edge node does not include the heat of the target content resource;

the first edge node is used for sending a source return request to the central node;

the central node is configured to determine, according to a global heat record, a node in which the target memory resource is stored, where the node in which the target memory resource is stored includes the second edge node;

the first edge node is further configured to return location information of the second edge node to the client, so that the client acquires the target content resource from the second edge node.

In some possible implementations, the local heat record of the first edge node includes the heat of the target content resource and the location information of the node storing the target content resource, where the node storing the target content resource includes the second edge node;

the first edge node is further configured to return location information of the second edge node to the client, so that the client obtains the target content resource from the second edge node.

the central node is used for determining the first edge node as a first return source according to the global heat record and returning the target content resource to the first edge node;

and the first edge node is also used for returning a content resource access response to the client.

In some possible implementations, the content distribution network includes a central node and at least one edge node mutual group, the edge node mutual group including the first edge node and the second edge node;

the first edge node is further configured to synchronize the local heat records of the second edge node.

In some possible implementations, the edge nodes in the edge node mutual group have the same or related domain name, or belong to the same operator.

In a third aspect, the present application provides a computer cluster. The computer cluster comprises a plurality of computers, each computer comprising a processor and a memory, the memory having stored therein computer readable instructions, the processor of a first computer of the plurality of computers being configured to execute the computer readable instructions stored in the memory of the first computer to perform the method steps performed by the first edge node as described in any implementation of the first aspect or the first aspect of the present application, and the processor of a second computer of the plurality of computers being configured to execute the computer readable instructions stored in the memory of the second computer to perform the method steps performed by the second edge node as described in any implementation of the first aspect or the first aspect of the present application.

In a fourth aspect, the present application provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and the instructions instruct a computer to execute the content resource distribution method according to the first aspect or any implementation manner of the first aspect.

In a fifth aspect, the present application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the content resource distribution method of the first aspect or any implementation manner of the first aspect.

The present application can further combine to provide more implementations on the basis of the implementations provided by the above aspects.

Drawings

In order to more clearly illustrate the technical method of the embodiments of the present application, the drawings used in the embodiments will be briefly described below.

Fig. 1 is a system architecture diagram of a content distribution network according to an embodiment of the present application;

fig. 2 is a flowchart of a content resource distribution method provided in an embodiment of the present application;

fig. 3 is a flowchart of a content resource distribution method according to an embodiment of the present application;

fig. 4 is a flowchart of a content resource distribution method according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of a computer cluster according to an embodiment of the present application.

Detailed Description

The terms "first", "second" in the embodiments of the present application are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or as implying any indication of the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

Some technical terms referred to in the embodiments of the present application will be first described.

Content resource means a resource that provides content. The content asset is typically stored and transmitted in the form of a digital file and may be presented to the user by a player or reader corresponding to the content asset. For example, a content asset may be a media asset such as audio, video, etc., where the audio may be played by an audio player and the video may be played by a video player. As another example, the content resource may be an electronic book, an electronic publication, which may be opened by the reader and presented to the user.

The CDN is used for distributing the content resources from the source station to the accelerating nodes close to the user, so that the user can obtain the required content resources nearby, and the response speed and success rate of the user access are improved. Where a source site refers to a site of origin, which may be a computer that provides the original content resource. The computer may be, for example, a server, and thus the source station may also be referred to as a source station server.

The back-to-source means that the access request (for example, a back-to-source request) is forwarded to the source station server by the source station, and the back-to-source request is responded by the source station server. It should be noted that the back-source is generally triggered by the acceleration node of the CDN when the acceleration node of the CDN does not cache the content resource requested to be accessed, or the cached content resource is expired.

At present, a widely used source returning scheme in the industry is to return the source through a full-connection network (full mesh network). Specifically, when an edge node receives an access request of a content resource and the edge node does not store the corresponding content resource, the edge node returns to the source of the edge node (also called peer node) in the same hierarchy, thereby reducing the bandwidth overhead of the central node. However, many content resources are accessed less frequently, i.e., the reuse rate of many content resources after being downloaded to the edge node is low. Therefore, the utilization rate of the storage resources of the edge node is low, and the cost of delivering the content resources by the CDN is increased.

In order to solve the above technical problem, an embodiment of the present application provides a content resource distribution method. The method can be applied to the CDN. Specifically, an edge node in the CDN receives a content resource access request from a client, where the content resource access request is used to request a target content resource, and when the heat of the target content resource satisfies a preset condition, the edge node sends a back-source request to another edge node (peer node), where the back-source request is used to obtain the target content resource from the peer node, and then the edge node receives the target content resource sent by the peer node, stores the target content resource, and returns a content resource access response to the client. Wherein the content resource access response carries the target content resource.

In the embodiment of the application, the edge node can return to the source based on the heat of the target content resource, when the heat of the target content resource meets the preset condition, the source is returned to the peer node, on one hand, the flow returning to the central node is reduced, so that the bandwidth overhead of the central node is reduced, on the other hand, when the heat meets the preset condition, the target content resource is stored at the edge node, the distribution of the target content resource in the CDN is optimized, the situation that a large amount of content resources with low heat occupy the storage space of the edge node is avoided, the utilization rate of the storage resources in the edge node is improved, and the cost of the CDN for distributing the content resources is reduced.

The embodiments of the present application will be described in detail below with reference to the general concepts described above.

Referring to fig. 1, fig. 1 is a schematic diagram of a system architecture of a CDN according to an embodiment of the present disclosure.

As shown in fig. 1, the content distribution network 100 includes a central node 102 and a plurality of edge nodes 104. The central node 102 may be one or more, and fig. 1 illustrates the content distribution network 100 as including one central node 102. The central node 102 establishes a communication connection with the edge node 104.

The central node 102 is responsible for the distribution of content resources and the management of the edge nodes 104. In particular, the central node 102 may determine the edge node 104 to use to respond to the content resource access request, distribute the content resource through the edge node 104, and the central node 102 may monitor the traffic and health of the edge node 104, such that forwarding the access request to an overloaded or failed edge node 104 may be avoided. The edge node 104, also referred to as an acceleration node, a cache node, is responsible for caching the content resource and responding to a content resource access request from the client based on the content resource.

For convenience of description, in the embodiment of the present application, one edge node 104 of the plurality of edge nodes 104 is referred to as a first edge node, and another edge node 104 of the plurality of edge nodes 104 is referred to as a second edge node. The first edge node and the second edge node may establish a communication connection.

The method comprises the steps that a first edge node receives a content resource access request from a client, the content resource access request is used for requesting a target content resource, when the heat degree of the target content resource meets a preset condition, the first edge node can send a source returning request to a second edge node, the source returning request is used for obtaining the target content resource from the second edge node, then the first edge node receives the target content resource sent by the second edge node, stores the target content resource, and returns a content resource access response to the client, and the content resource access response carries the target content resource.

Next, a content resource distribution method provided by an embodiment of the present application will be described in detail from the perspective of the content distribution network 100 with reference to the drawings.

Referring to fig. 2, a flow chart of a content resource distribution method is shown, where the method includes:

s202: the first edge node receives a content resource access request from a client. When the degree of heat of the target content resource requested to be accessed satisfies the first preset condition, S204 is executed, and when the degree of heat of the target content resource requested to be accessed satisfies the second preset condition, S206 is executed.

Specifically, the content resource access request may carry an identifier of the target content resource, such as a unique identifier of the target content resource, such as a name and a hash value. The first edge node can obtain the heat degree of the target content resource, and when the heat degree meets a preset condition, a source return request can be sent to the peer node.

The popularity of the target content resource may include the popularity of the target content resource at the edge node 104, such as the first edge node, and the popularity of the target content resource at the central node 102. The degree of popularity of a target content resource at an edge node 104 is also referred to as the local degree of popularity, and the degree of popularity of a target content resource at a central node 102 is also referred to as the central degree of popularity or the global degree of popularity. Based on this, the preset condition may include a first preset condition and a second preset condition.

The first preset condition may be that the local heat record does not include the heat of the target content resource, and the heat of the target content resource in the global heat record stored by the central node 102 is greater than or equal to a first preset threshold. The heat of the target content resource recorded in the global heat record may be a global heat of the target content resource, and the global heat may be determined according to the number of times of source return requests of the target content resource counted by the central node 102. When the target content resource meets a first preset condition, the mutual help return source can be triggered. Unlike a central back source (which sends back source requests to the central node 102), a mutual-aid back source is a request to send back source requests to a peer node, such as a second edge node.

It should be noted that, the first edge node may first determine whether the local heat record includes the heat of the target content resource, and when the local heat record does not include the heat of the target content resource, the first edge node may send a source return request to the central node 102. The central node 102 receives the back-source request, and updates the heat of the target content resource in the global heat record, for example, when the heat of the target content resource is represented by the back-source request times, an operation may be performed on the back-source request times to update the heat of the target content resource. The central node 102 then compares the heat (e.g., global heat) of the target content resource with a first preset threshold, thereby determining whether the heat of the target content resource satisfies a first preset condition.

The second preset condition may be that the local heat record includes the heat of the target content resource, and the heat of the target content resource is greater than or equal to a second preset threshold. Wherein, the hot degree of the target content resource recorded in the local hot degree record may be the local hot degree of the target content resource. The local heat may be determined according to the number of access requests of the target content resource counted by the edge node 104 (e.g., the first edge node). When the target content resource meets a second preset condition, the mutual help return source can be triggered.

S204: the first edge node receives the location information of the second edge node sent by the central node 102. Then S206 is executed.

Specifically, the central node 102 receives the back-source request, updates the global heat of the target content resource according to the back-source request, and queries the edge node 104 storing the target content resource, for example, the second edge node, when the global heat is greater than or equal to the first threshold, to obtain the location information of the second edge node.

Wherein the location information is used to characterize the location of the second edge node. The location information may comprise an IP address of the second edge node, and further the location information may comprise a port number. The location information may be used to establish a communication connection. For example, the first edge node may establish a communication connection with the second edge node based on the location information of the second edge node.

It should be noted that S204 may not be executed when the content resource distribution method according to the embodiment of the present application is executed, for example, S206 may be executed after S202 is executed when the heat degree of the target content resource satisfies the second preset condition.

S206: the first edge node sends a back-source request to the second edge node.

The back source request is for requesting a target content resource. The first edge node may send the source return request to the second edge node according to the location information of the second edge node. The location information of the second edge node may come from the central node 102 or may come locally from the first edge node.

Specifically, when the heat degree of the target content resource satisfies the first preset condition, the first edge node may send a source return request to the second edge node according to the location information after obtaining the location information of the second edge node through S204; when the heat degree of the target content resource meets a second preset condition, the first edge node may obtain the location information of the second edge node from the local, and send a source return request to the second edge node according to the location information. The second preset condition may be that the local heat record includes the heat of the target content resource, and the heat of the target content resource is greater than or equal to a second preset threshold.

S208: and the first edge node receives the target content resource sent by the second edge node.

The second edge node stores a target content resource. The first edge node may obtain the target content resource from the second edge node. It should be noted that the target content resource obtained by the first edge node is a copy.

S210: the first edge node stores the target content resource and returns a content resource access response to the client.

The first edge node may store the target content resource for subsequent response to a new content resource access request. And, the first edge node may generate a content resource access response according to the target content resource acquired from the second edge node. The content resource access response includes the target content resource.

Based on the above description, an embodiment of the present application provides a content resource distribution method. In the method, the edge node can return to the source based on the heat degree of the target content resource, when the heat degree of the target content resource meets the preset condition, the source is returned to the peer node, on one hand, the flow returning to the central node is reduced, and therefore the bandwidth overhead of the central node is reduced, on the other hand, when the heat degree meets the preset condition, the target content resource is stored in the edge node, the distribution of the target content resource in the CDN is optimized, the fact that the content resource with low heat degree occupies a large amount of the storage space of the edge node is avoided, the utilization rate of the storage resource in the edge node is improved, and the cost of the CDN for distributing the content resource is reduced.

While the embodiment shown in fig. 2 details the content resource distribution in a mutual-aid source-back manner, in some possible implementations, the edge node 104 may also implement the content resource distribution in a mutual-aid service manner. The following situations may be included in the edge node 104 for realizing content resource distribution in a mutual service manner, and each situation is illustrated below.

In the first case, the target content resource is accessed for the first time. Specifically, the local heat record of the first edge node does not include the heat of the target content resource. The target content resource misses at the first edge node. Based on this, the first edge node may send a source return request to the central node 102, and the central node 102 determines that the first edge node is the first source return according to the global heat record, and then returns the target content resource to the first edge node. And the first edge node returns a content resource access response to the client. Therefore, the content resources with lower heat degree are prevented from additionally occupying the storage space of the edge node, and the utilization rate of the storage resources is improved.

In the second case, the target content resource has been accessed at the second edge node, where it is accessed for the first time. Specifically, the local heat record of the first edge node does not include the heat of the target content resource. The target content resource misses at the first edge node, upon which the first edge node may send a back source request to the central node 102. The central node 102 determines the node storing the target memory resource according to the global heat record. The node storing the target memory resource comprises a second edge node. The first edge node may return location information of the second edge node to the client, so that the client acquires the target content resource from the second edge node.

It should be noted that the first edge node may redirect the packet by constructing a redirection packet, for example, a 302 packet. Therein 302 represents a status code in the hypertext transfer protocol. The redirection message is used for redirecting the content resource access request to the second edge node. Based on this, the redirection message may carry location information of the second edge node. And the first edge node sends a redirection message to the client, so that the position information of the second edge node is returned to the client.

In a third case, the target content resource has been accessed at the second edge node and is accessed again at the first edge node. The local heat record of the first edge node comprises the heat of the target content resource and the position information of the node storing the target content resource. Wherein the node storing the target content resource comprises the second edge node. In this way, the first edge node may return the location information of the second edge node to the client, so that the client obtains the target content resource from the second edge node.

Similar to the second case, the first edge node may construct a redirection packet, for example, a 302 packet, and then send the redirection packet to return the location information of the second edge node to the client, so that the client acquires the target content resource from the second edge node.

In some possible implementations, the CDN may also perform mutual-aid return source or mutual-aid service in units of edge node mutual-aid groups. The edge nodes 104 in the CDN may be grouped by domain name and/or operator. Based on this, the CDN includes a central node 102 and at least one edge node mutual group. The edge node mutual aid group comprises a first edge node and a second edge node. The edge nodes 104 (e.g., first edge node, second edge node) in the edge node mutual aid group have the same or related domain name or are owned by the same operator. The grouping information of the edge nodes 104 may also be synchronized to the various edge nodes 104 so that the various edge nodes 104 may obtain location information of other nodes within the group based on the grouping information.

Wherein nodes having the same or related domain names may be edge nodes of the same organization as the same company. The client has a high probability of accessing the same domain name or related domain names multiple times within a time period. Thus, grouping in the manner described above may improve the efficiency of the mutual aid back source or mutual aid service. In addition, the edge nodes between different operators have higher interaction cost, and the edge nodes belonging to different operators are divided into different groups, so that the interaction cost can be reduced, and the cost of mutual-help return source or mutual-help service is reduced.

In some possible implementations, the first edge node may also synchronize the local heat record of the second edge node. In this way, the first edge node may determine whether to trigger the mutual aid return source or mutual aid service according to the synchronized local heat record (the local heat record of the edge node mutual aid group). Because the edge node mutual-aid group is taken as a unit, the mutual-aid return source or mutual-aid service can be realized in a small range, so that the bandwidth overhead can be reduced, and the cost of the mutual-aid return source or mutual-aid service can be reduced. In addition, the resources are shared among the edge nodes, and the traffic contribution rate of the content resources with lower heat can be improved.

For ease of understanding, the following description of the process of enabling the distribution of content resources via a mutual aid service and the distribution of content resources via a mutual aid source is provided in connection with a specific example.

Referring to the flowchart of the content resource distribution method shown in fig. 3, where an edge node a, an edge node B, and an edge node C belong to the same edge node mutual group, and the node C is not shown in fig. 3, the method includes the following steps:

s302: the client A sends a content resource access request to the edge node A.

Wherein the content resource access request is for requesting a target content resource, which may be, for example, C1. The content resource access request may be through a get method in a hypertext transfer protocol (HTTP). Based on this, the content resource access request can be represented as Http Get C1 as shown in fig. 3.

S304: the target content resource C1 is locally missing at the edge node a, and the edge node a queries the heat of the target content resource C1 in the local heat record. When the degree of heat of the target content resource C1 is not included in the local degree of heat record, S306 is executed.

S306: edge node a sends a back-source request to central node 102.

S308: the central node 102 queries the global heat record and determines that the target content resource C1 is downloaded for the first time.

Based on the determination, the central node 102 may query the heat of the target content resource C1 in the global heat record to determine whether the target content resource C1 is downloaded for the first time.

S310: the central node 102 returns the target content resource C1 to the edge node a.

S312: the central node 102 updates the copy location of the target content resource in the global heat record.

Wherein the copy location of the target content asset C1 may be a location of an edge node comprising the target content asset. For example, the copy location may be the location of the second edge node. The central node 2 may update the copy location of the target content resource C1 in the global heat record to include the location of the second edge node.

It should be noted that S310 and S312 may be executed in parallel, or may be executed sequentially according to a set sequence, for example, S310 and then S312 may be executed first, or S312 and then S310 may be executed first.

S314: the edge node A returns a content resource access response to the client A. The content resource access response carries the target content resource C1.

The above-mentioned S302 to S314 describe the content resource distribution method when the target content resource C1 is first accessed in the edge node mutual aid group, and then, the content resource distribution method when the target content resource C1 is first accessed in the edge node B is introduced. The method specifically comprises the following steps:

the client B1 sends a content resource access request to the edge node B S316.

S318, the target content resource C1 is not hit in the local of the edge node B, and the edge node B inquires the heat of the target content resource C1 in the local heat record. When the heat of the target content resource C1 is not included in the local heat record, S320 is performed.

S320: the edge node B sends a back-source request to the central node 102.

S322: the central node 102 queries the global heat record, and determines that the target content resource C1 is stored in the edge node a in the edge node mutual aid group.

S324: the central node 102 returns the location information of edge node a to edge node B.

Specifically, the central node 102 may construct a return packet, where the return packet carries a special header field, and the special header field may point to the edge node a where the target content resource C1 is located.

S326: the edge node B sends a redirect message to the client B1.

The redirection message is used for instructing to redirect the content resource access request to the edge node a.

S328: the edge node B updates the copy location of the target content resource C1 in the local heat record.

Specifically, the edge node B updates an edge node including the target content resource C1, for example, the edge node a, in the local heat record.

It should be noted that, the foregoing S326 and S328 may be executed in parallel, or may be executed sequentially according to a set sequence, which is not limited in this embodiment of the application.

S330: the client B1 sends a content resource access request to the edge node a.

S332: the target content resource C1 locally hits in the edge node A, and the edge node A returns a content resource access response to the client B1.

The above-mentioned S316 to S314 describe the content resource distribution method of the target content resource C1 when the edge node B is accessed for the first time. Next, a content resource distribution method when the target content resource C1 is accessed again at the edge node B is described, which specifically includes the following steps:

s334: the client B2 sends a content resource access request to the edge node B.

S336: the target content resource C1 is locally missed in the edge node B, the local heat record of the edge node B includes copy location information of the target content resource C1, and the edge node B returns location information of the edge node a to the client B2.

Specifically, the edge node B may construct a redirection packet according to the copy location information of the target content resource C1, for example, the location information of the edge node a including the target content resource C1, and then send the redirection packet to the client B2. The redirection packet carries the location information of the edge node a.

S338: the client B2 sends a content resource access request to the edge node a.

S340: the target content resource C1 locally hits in the edge node A, and the edge node A returns a content resource access response to the client B2.

Based on the content description, the embodiment of the application provides a policy of mutual service between edge nodes, optimizes the back source processing of the content resource with low heat, changes the central back source request into 302 scheduling service, specifically skips the content resource access request to the node where the content resource is located, and responds by the node where the content resource is located, so that the flow of the central back source request can be greatly reduced, and the invalid back source download flow can be reduced.

The embodiment shown in fig. 3 illustrates a mutual aid service, followed by an example of a mutual aid return source.

Referring to the flowchart of the content resource distribution method shown in fig. 4, where an edge node a, an edge node B, and an edge node C belong to the same edge node mutual group, the method includes the following steps:

s402: the client C sends a content resource access request to the edge node C.

S404: the target content resource C1 misses locally at the edge node C, which queries the local heat record. When the local heat record does not include the heat of the target content resource C1, S406 is performed.

S406: edge node C sends a back-source request to the central node 102.

S408: the central node 102 queries the global heat record, and executes S410 and S412 when the global heat of the target content resource C1 is greater than a first preset threshold.

S410: the central node 102 returns the location information of edge node a to edge node B.

Specifically, the central node 102 may construct a return packet, where the return packet carries a special header field, and the special header field may point to the edge node a where the target content resource C1 is located. It should be noted that the return packet may also carry a mutual-aid return source identifier, which is used to indicate that the target content resource C1 is stored at the edge node C, so as to perform mutual-aid return source.

S412: the central node 102 updates the copy location of the target content resource C1 in the global heat record.

Specifically, the central node 102 determines that the edge nodes storing the target content resource C1 further include the edge node C, and may update the location information of the edge node C storing the target content resource C1 in the global heat record.

S410 and S412 may be executed in parallel, or may be executed sequentially according to a set sequence, which is not limited in this embodiment of the present application.

S414: edge node C sends a back source request to edge node a.

S416: the target content resource C1 is hit locally at edge node a, which returns the target content resource C1 to edge node C.

S418: and the edge node C returns a content resource access response, and the content resource access response carries the target content resource C1.

S420: the edge node C stores a target content asset C1.

The foregoing S402 to S420 describe a specific implementation of triggering the mutual aid feedback source based on the global heat, and then a specific implementation of triggering the feedback source based on the local heat is described, which specifically includes the following steps:

s422: the client Bn sends a content resource access request to the edge node B.

S424: the target content resource C1 misses locally at the edge node B, which queries the local heat record. When the target content resource C1 exists at the edge node a in the local heat record, and the local heat of the target content resource C1 is greater than or equal to the second preset threshold, S426 is executed.

S426: edge node B sends a back-source request to edge node a.

S428: the target content resource C1 hits at edge node a, which returns the target content resource C1 to edge node B.

S430: and the edge node B returns a content resource access response, and the content resource access response carries the target content resource C1.

S432: the edge node B stores a target content asset C1.

Based on the content description, the cold resources and the hot resources can be identified based on the heat degree of the target content resources in the local heat degree record or the global heat degree record, and the distribution of the target content resources in the CDN is quickly adjusted after the resources are changed from cold to hot, so that the utilization rate of a storage space is improved, the storage capacity requirement is reduced, and further, the hardware investment cost is reduced.

The content resource distribution method provided by the embodiment of the present application is described in detail above with reference to fig. 1 to 4, and the content distribution network provided by the embodiment of the present application is described below with reference to the accompanying drawings.

Referring to the schematic structural diagram of the content distribution network shown in fig. 1, the content distribution network 10 includes a central node 102 and a plurality of edge nodes 104, where the plurality of edge nodes 104 includes a first edge node and a second edge node.

In some possible implementation manners, the fact that the heat degree of the target content resource meets the preset condition includes that the heat degree of the target content resource is not included in a local heat degree record, and the heat degree of the target content resource in a global heat degree record stored by the central node 102 is greater than or equal to a first preset threshold;

the first edge node is further configured to receive, before the first edge node sends a source return request to the second edge node, location information of the second edge node sent by the central node 102.

In some possible implementation manners, the condition that the heat degree of the target content resource meets the preset condition includes that the heat degree of the target content resource is included in a local heat degree record, and the heat degree of the target content resource is greater than or equal to a second preset threshold;

the first edge node is configured to send a source return request to the central node 102;

the central node 102 is configured to determine, according to a global heat record, a node in which the target memory resource is stored, where the node in which the target memory resource is stored includes the second edge node;

the central node 102 is configured to determine that the first edge node is a first return source according to the global heat record, and return the target content resource to the first edge node;

In some possible implementations, the content distribution network includes a central node 102 and at least one edge node mutual group, the edge node mutual group including the first edge node and the second edge node;

The content distribution network 100 according to the embodiment of the present application may correspond to perform the method described in the embodiment of the present application, and the above and other operations and/or functions of each part in the content distribution network are respectively for implementing corresponding flows of each method in the embodiments shown in fig. 2 to 4, and are not described herein again for brevity.

The embodiment of the application further provides a computer cluster. The computer cluster may be a computer cluster formed by multiple computers in a cloud environment, edge environment, or end device. The computer cluster is specifically used to implement the functions of the content distribution network 100 in the embodiment shown in fig. 1.

Fig. 5 provides a schematic diagram of a computer cluster, and as shown in fig. 5, the computer cluster 50 includes a plurality of computers 500, and each computer 500 includes a bus 501, a processor 502, a communication interface 503, and a memory 504. The processor 502, memory 504, and communication interface 503 communicate with each other via a bus 501.

The bus 501 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.

The processor 502 may be any one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Micro Processor (MP), a Digital Signal Processor (DSP), and the like.

The communication interface 503 is used for communication with the outside. For example, the communication interface 503 of one computer 500 may be used to receive content resource access requests from a client, send back source requests to another computer 500, receive target content resources sent by another computer 500, and so on.

The memory 504 may include volatile memory (volatile memory), such as Random Access Memory (RAM). The memory 504 may also include non-volatile memory (non-volatile memory), such as read-only memory (ROM), flash memory, a Hard Disk Drive (HDD), or a Solid State Drive (SSD).

The memory 504 has stored therein executable code that the processor 502 executes to perform the aforementioned content asset distribution method.

In particular, in case of implementing the embodiment shown in fig. 1, and in case that the functions of the parts of the content distribution network 100 described in the embodiment of fig. 1, such as the first edge node, the second edge node, are implemented by software, the software or program code required for executing the functions in fig. 1 may be stored in the memory 504. The processor 502 executes the program code stored in the memory 504 to perform the content asset distribution method described previously.

The embodiment of the application also provides a computer readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store or a data storage device, such as a data center, that contains one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk), among others. The computer-readable storage medium includes instructions that instruct a computing device to perform the content asset distribution method described above.

The embodiment of the application also provides a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computing device, cause the processes or functions described in accordance with embodiments of the application to occur, in whole or in part.

The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, or data center to another website site, computer, or data center by wire (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wirelessly (e.g., infrared, wireless, microwave, etc.).

The computer program product may be a software installation package which may be downloaded and executed on a computing device in the event that any of the aforementioned content resource distribution methods are required to be used.

The description of the flow or structure corresponding to each of the above drawings has emphasis, and a part not described in detail in a certain flow or structure may refer to the related description of other flows or structures.

Claims

1. A content resource distribution method is applied to a Content Distribution Network (CDN), the CDN comprises a central node and a plurality of edge nodes, the plurality of edge nodes comprise a first edge node and a second edge node, and the method comprises the following steps:

the first edge node receives a content resource access request from a client, wherein the content resource access request is used for requesting a target content resource;

when the heat degree of the target content resource meets a preset condition, the first edge node sends a source returning request to the second edge node, wherein the source returning request is used for requesting the target content resource;

and the first edge node receives the target content resource sent by the second edge node, stores the target content resource, and returns a content resource access response to the client, wherein the content resource access response carries the target content resource.

2. The method according to claim 1, wherein the fact that the target content resource has a heat degree satisfying a preset condition includes the fact that the target content resource is not included in a local heat degree record, and the heat degree of the target content resource in a global heat degree record stored by the central node is greater than or equal to a first preset threshold;

before the first edge node sends a back-to-source request to the second edge node, the method further comprises:

and the first edge node receives the position information of the second edge node sent by the central node.

3. The method according to claim 1, wherein the target content resource's degree of heat meeting the preset condition comprises the target content resource's degree of heat being included in a local degree of heat record, and the target content resource's degree of heat being greater than or equal to a second preset threshold;

the local heat record further includes location information of the second edge node storing the target content resource.

4. The method according to any of claims 1 to 3, wherein the local heat record of the first edge node does not include the heat of the target content resource, the method further comprising:

the first edge node sends a source returning request to the central node;

the central node determines a node storing the target memory resource according to a global heat record, wherein the node storing the target memory resource comprises the second edge node;

and the first edge node returns the position information of the second edge node to the client so that the client acquires the target content resource from the second edge node.

5. The method according to any one of claims 1 to 3, wherein the local hot record of the first edge node includes the hot degree of the target content resource and the location information of the node storing the target content resource, and the node storing the target content resource includes the second edge node;

the method further comprises the following steps:

6. The method of any of claims 1 to 3, wherein the local heat record of the first edge node does not include the heat of the target content resource, the method further comprising:

the first edge node sends a source returning request to the central node;

the central node determines the first edge node as a first return source according to the global heat record, and returns the target content resource to the first edge node;

and the first edge node returns a content resource access response to the client.

7. The method of any of claims 1-6, wherein the CDN comprises a hub node and at least one edge node mutual group, the edge node mutual group comprising the first edge node and the second edge node; the method further comprises the following steps:

the first edge node synchronizes the local heat record of the second edge node.

8. The method of claim 7, wherein the edge nodes in the edge node mutual aid group have the same or related domain name or belong to the same operator.

9. A content distribution network, the network comprising a central node and a plurality of edge nodes, the plurality of edge nodes comprising a first edge node and a second edge node;

the second edge node is configured to send the target content resource to the first edge node;

10. The network according to claim 9, wherein the fact that the degree of heat of the target content resource satisfies the preset condition includes the fact that the degree of heat of the target content resource is not included in a local degree of heat record, and the degree of heat of the target content resource in a global degree of heat record stored by the central node is greater than or equal to a first preset threshold;

11. The network according to claim 9, wherein the fact that the degree of heat of the target content resource satisfies the preset condition includes that the degree of heat of the target content resource is included in a local degree of heat record, and the degree of heat of the target content resource is greater than or equal to a second preset threshold;

12. The network according to any one of claims 9 to 11, wherein the local heat record of the first edge node does not include the heat of the target content resource;

13. The network according to any one of claims 9 to 11, wherein the local hot record of the first edge node includes hot degree of the target content resource and location information of a node storing the target content resource, and the node storing the target content resource includes the second edge node;

14. The network according to any one of claims 9 to 11, wherein the local heat record of the first edge node does not include the heat of the target content resource;

the first edge node is used for sending a source returning request to the central node;

15. The network according to any of claims 9 to 14, wherein the content distribution network comprises a central node and at least one edge node mutual group, the edge node mutual group comprising the first edge node and the second edge node;

16. The network of claim 15, wherein the edge nodes in the edge node mutual aid group have the same or related domain name or belong to the same operator.

17. A computer cluster comprising a plurality of computers, each computer comprising a processor and a memory having stored therein computer readable instructions, the processor of a first computer of the plurality of computers being configured to execute the computer readable instructions stored in the memory of the first computer to perform the method steps performed by a first edge node of any of claims 1 to 8, the processor of a second computer of the plurality of computers being configured to execute the computer readable instructions stored in the memory of the second computer to perform the method steps performed by a second edge node of any of claims 1 to 8.

18. A computer-readable storage medium, comprising computer-readable instructions which, when run on a computer, cause the computer to perform the content resource distribution method of any one of claims 1 to 8.