A response method of a content distribution network

Abstract

The present disclosure provides a response method of a content distribution network, aiming at solving the problem that service templates cannot be flexibly adjusted in the related art. The method comprises: obtaining a first file pulling request; in a case where it is determined that a target file pulled by the first file pulling request is not cached, determining a target back intermediate layer strategy of the first file pulling request according to an address analysis result, a performance loss result and a full-link detection result, the back intermediate layer strategy comprising: a near principle strategy, a consistent hash algorithm strategy and a specified intermediate cluster strategy; determining a target intermediate server cluster according to the target back intermediate layer strategy, and sending a second file pulling request to the target intermediate server cluster; and receiving a target file pulled by the target intermediate server cluster from a source station.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of content delivery network, and particularly relates to a response method of content delivery network. BACKGROUND

[0002] There are various sources of customer channels served by a content delivery network (CDN) vendor. Based on the business characteristics of customers, customers can be divided into two categories. One category is provincial customers, and the user and source station of the customers are in the same province. The other category is nationwide customers, and the user of the customers is distributed throughout the province, and the source station is in a certain province. According to the conventional method of each CDN vendor, when a customer of any category is newly connected to the CDN vendor, a service template is used consistently. That is, each province has an edge cluster to respond to the user. If the edge cluster does not have cache, the file is requested from the intermediate cluster in different provinces through a consistent hashing algorithm and is pulled from the source station.

[0003] However, if the same service template is used for different types of customers, the response speed after CDN acceleration may be slower than the response speed of direct access to the source station. When customers complain about the problem, the CDN vendor will adjust the service template of the customer. SUMMARY

[0004] To overcome the problems in the related art, the present disclosure provides a response method of content delivery network. The technical solution of the present disclosure is as follows.

[0005] According to a first aspect of an embodiment of the present disclosure, a response method of content delivery network is provided, applied to an edge server cluster, and includes the following steps.

[0006] A first file pull request is obtained.

[0007] In a case where it is determined that a target file pulled by the first file pull request is not cached, a target intermediate layer strategy of the first file pull request is determined according to an address analysis result, a performance loss result and a full-link detection result. The target intermediate layer strategy includes a near principle strategy, a consistent hashing algorithm strategy and a specified intermediate cluster strategy.

[0008] According to the target intermediate layer strategy, a target intermediate server cluster is determined, and a second file pull request is sent to the target intermediate server cluster.

[0009] The target file pulled from the source station by the target intermediate server cluster is received.

[0010] Optionally, determining the target intermediate layer strategy of the first file pull request according to the address analysis result, the performance loss result and the full-link detection result includes the following steps.

[0011] obtaining an address analysis result from a data analysis platform;

[0012] determining, according to the address analysis result, whether the target service corresponding to the first file pulling request is a first service; the first service representing that a source station and a user of the target service are located in a same region;

[0013] in a case where the target back intermediate layer strategy is a designated intermediate cluster strategy, determining, according to the target back intermediate layer strategy, a target intermediate server cluster, comprising:

[0014] in a case where the target service corresponding to the first file pulling request is the first service, determining, by the designated intermediate cluster strategy, an intermediate server cluster located in the same region as the target service as the target intermediate server cluster.

[0015] Optionally, in a case where the target back intermediate layer strategy is a nearest principle strategy and a target capacity of the target file is less than or equal to a first capacity, determining, according to the target back intermediate layer strategy, a target intermediate server cluster, and sending a second file pulling request to the target intermediate server cluster, comprising:

[0016] determining, by the nearest principle, an intermediate server cluster closest to the edge server cluster as the target intermediate server cluster;

[0017] initiating, to the target intermediate server cluster, a second file pulling request for the target file; and receiving the target file pulled from the source station by the target intermediate server cluster.

[0018] Optionally, in a case where the target back intermediate layer strategy is the nearest principle strategy and the target capacity of the target file is greater than the first capacity, determining, according to the target back intermediate layer strategy, a target intermediate server cluster, and sending a second file pulling request to the target intermediate server cluster, comprising:

[0019] determining, by the nearest principle, an intermediate server cluster closest to the edge server cluster as a first target intermediate server cluster; and determining, by a consistent hashing algorithm, a second target intermediate server cluster corresponding to the first file pulling request;

[0020] determining a first file as a first capacity of the target file, and determining a second file as a remaining capacity of the target file other than the first capacity;

[0021] initiating, to the first target intermediate server cluster, a second file pulling request for the first file, to obtain the first file pulled from the source station by the first target intermediate server cluster;

[0022] initiating a second file pull request for the second file to the second target intermediate server cluster to obtain the second file pulled from the source station by the second target intermediate server cluster.

[0023] Optionally, the full-link probing result is determined by the following steps:

[0024] probing each intermediate server cluster to obtain a first probing result;

[0025] receiving a second probing result transmitted by each intermediate server cluster;

[0026] obtaining a full-link probing result based on the first probing result and the second probing result.

[0027] Optionally, the performance loss result is determined by the following steps:

[0028] collecting domain name service information required by a performance loss model; the domain name service information includes: an average file size requested by each historical first file pull request and a number of each historical first file pull request;

[0029] analyzing the domain name service information based on the performance loss model to obtain an intermediate layer processor loss and a bandwidth consumption;

[0030] determining the intermediate layer processor loss and the bandwidth consumption as the performance loss result.

[0031] According to a second aspect of the embodiments of the present disclosure, a response method of a content distribution network is provided, applied to an intermediate server cluster, and including:

[0032] receiving a second file pull request sent from an edge server cluster and obtaining a target file from a source station according to the second file pull request; the second file pull request is determined by a nearest principle strategy, a consistent hash algorithm strategy or a specified intermediate cluster strategy;

[0033] transmitting the target file to the edge server cluster.

[0034] Optionally, in a case where the second file pull request is determined by the nearest principle strategy and a target capacity of the target file is greater than a first capacity, the intermediate server cluster includes: a first target intermediate server cluster and a first target intermediate server cluster.

[0035] receiving a second file pull request sent from an edge server cluster and obtaining a target file from a source station according to the second file pull request, including

[0036] The first target intermediate server cluster receives a second file pull request for the first file, the first file being a file corresponding to a first capacity of the target file;

[0037] According to the second file pull request, the first file is obtained from a source station, and the first file is transmitted to the second target intermediate server cluster, so that the second target intermediate server cluster caches the first file;

[0038] The second target intermediate server cluster receives a second file pull request for the second file, the second file being a file corresponding to a remaining capacity of the target file except the first capacity;

[0039] According to the second file pull request, the second file is obtained from a source station, and the second file is cached.

[0040] Optionally, further comprising:

[0041] Each source station is probed to obtain a second detection result;

[0042] The second detection result is transmitted to each edge server cluster.

[0043] According to a third aspect of the embodiments of the present disclosure, a response method of a content distribution network is provided, applied to a data analysis platform, comprising:

[0044] Each historical first file pull request is parsed, and according to user address corresponding to each historical first file pull request, user area information corresponding to each historical first file pull request is determined;

[0045] According to source station address corresponding to each historical first file pull request, source station area information corresponding to each historical first file pull request is determined;

[0046] The user area information and the source station area information corresponding to each historical first file pull request are matched to determine a target overlap rate of a target service corresponding to a first file pull request received by an edge server cluster;

[0047] According to the size relationship between the target overlap rate and a first threshold value, an address analysis result is obtained, the address analysis result at least including that the target service is a first service, the first service representing that the source station and the user of the target service are located in the same area;

[0048] The address analysis result is sent to the edge server cluster.

[0049] According to a fourth aspect of the embodiments of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, when the computer program is executed by the processor, the steps of the response method of the content distribution network according to the first aspect, or the steps of the response method of the content distribution network according to the second aspect, or the steps of the response method of the content distribution network according to the third aspect are implemented.

[0050] According to a fifth aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, the computer readable storage medium stores a computer program, when the computer program is executed by a processor, the steps of the response method of the content distribution network according to the first aspect, or the steps of the response method of the content distribution network according to the second aspect, or the steps of the response method of the content distribution network according to the third aspect are implemented.

[0051] According to a sixth aspect of the embodiments of the present disclosure, a computer program product is provided, comprising a computer program, when the computer program is executed by a processor, the steps of the response method of the content distribution network according to the first aspect, or the steps of the response method of the content distribution network according to the second aspect, or the steps of the response method of the content distribution network according to the third aspect are implemented.

[0052] The present disclosure selects the most suitable back-to-intermediate layer strategy through the specific conditions of the file pulling request, including address analysis results, performance loss results, and full-link probing results, which can significantly optimize the response speed of CDN acceleration and avoid the problem that the response speed is slower than direct access to the source station. The present disclosure supports multiple back-to-intermediate layer strategies, and can flexibly select and adjust the back-to-intermediate layer strategy according to actual business needs, which can easily extend more back-to-intermediate layer strategies, thereby optimizing the response speed of CDN acceleration. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed in the description of the embodiments of the present disclosure will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0054] Figure 1 is a step schematic diagram of a response method of a content distribution network according to an embodiment of the present disclosure;

[0055] Figure 2 is a schematic diagram of determining performance loss results according to an embodiment of the present disclosure;

[0056] Figure 3Is a machine learning SVM related code demo and classification result comparison chart shown by an embodiment of the present disclosure.

[0057] Figure 4 Is a schematic diagram of determining a target back intermediate layer strategy shown by an embodiment of the present disclosure.

[0058] Figure 5 Is a step schematic diagram of a response method of a content distribution network shown by an embodiment of the present disclosure.

[0059] Figure 6 Is a schematic diagram of a target file acquisition method of a source station by an intermediate server cluster shown by an embodiment of the present disclosure.

[0060] Figure 7 Is a step schematic diagram of a response method of a content distribution network shown by an embodiment of the present disclosure.

[0061] Figure 8 Is a schematic diagram of determining an address analysis result shown by an embodiment of the present disclosure.

[0062] Figure 9 Is a schematic diagram of an overall access architecture diagram shown by an embodiment of the present disclosure.

[0063] Figure 10 Is a schematic diagram of an electronic device shown by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present disclosure.

[0065] The terms "first", "second", and the like in the specification and claims of the present disclosure are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually a class, not limited to the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the front and rear associated objects are in an "or" relationship.

[0066] For the convenience of understanding, first, the professional terms involved in the present disclosure are explained.

[0067] SVM (support vector machine): a binary classification model, its basic model is the largest linear classifier defined in the feature space, the largest interval makes it different from the perception machine; SVM also includes kernel skills, which makes it a substantive nonlinear classifier. The learning strategy of SVM is to maximize the interval, which can be formalized as a convex quadratic programming problem, which is equivalent to the minimization problem of the regularized hinge loss function. The learning algorithm of SVM is the optimization algorithm for solving convex quadratic programming.

[0068] CDN: a distributed content distribution network built on a data network. The role of CDN is to use streaming server cluster technology to overcome the shortcomings of insufficient output bandwidth and concurrent capacity of single machine system, which can greatly improve the number of concurrent streams supported by the system and reduce or avoid the adverse effects of single point failure.

[0069] There are many sources of customer channels for CDN provider services, and the business characteristics are different. In order to improve the efficiency of customer service, each manufacturer will pay more attention to customers with large bandwidth, which brings the disadvantage that small and medium-sized customers cannot be well compatible and serviced. The main concern of customers is the response speed of CDN acceleration. Based on the business characteristics of customers, customers can be divided into two categories. One is the provincial customer, that is, the user and the source station of the customer are in the same province; the other is the national customer, that is, the user is distributed throughout the province, and the source station is in a province.

[0070] According to the conventional method in the industry, no matter which type of customer is new to the CDN manufacturer, the service template used is consistent, that is, each province has an edge cluster to respond to users. If the edge cluster does not have cache, it will request the intermediate cluster of different provinces through consistent hashing algorithm and pull the file from the source. The advantage of this scheme is that a file is pulled from the source only once, and the disadvantage is that the user may be in the northwest province, the parent layer is in the northeast province, and the source station is in the northwest province, resulting in that the user is slower than directly accessing the source station through CDN acceleration. When customers complain about this, the service template will be adjusted, and only the parent layer in the same province will be left to provide service.

[0071] In addition, the current CDN manufacturer basically has mature dynamic acceleration algorithm, but for static files that need to be cached, there is no mature algorithm. The main difference between static files and dynamic files is that dynamic files are usually small files, high qps, not cached, and high CPU consumption. Static files have small files and large files, need to be cached, and involve disk. If static files and dynamic files are accelerated in the same way, the demand for caching static files cannot be met, and bandwidth will be wasted. Machine performance may not be able to meet the demand.

[0072] To solve the above technical problems, the present disclosure proposes a response method of a content distribution network, which is applied to a response system of the content distribution network, and can determine a back-to-intermediate-layer strategy according to a specific file pull request, so as to avoid the problem that the acceleration through the CDN is slower than direct access to the source station.

[0073] The response system of the content distribution network includes an edge server cluster, an intermediate server cluster, and a data analysis platform. The response system of the content distribution network is used to, in the case where there is no cache of a target file, determine the most suitable intermediate server cluster according to a first file pull request of a user, and pull the target file corresponding to the first file pull request through the intermediate server cluster, so that the user obtains the target file. The response system of the content distribution network determines the most suitable intermediate server cluster corresponding to the first file pull request using a back-to-intermediate-layer strategy, wherein the back-to-intermediate-layer strategy includes a nearest principle strategy, a consistent hash algorithm strategy, and a specified intermediate cluster strategy.

[0074] The edge server cluster is used to receive the first file pull request of the user, determine the back-to-intermediate-layer strategy through the first file pull request, determine the target intermediate server cluster corresponding to the first file pull request through the back-to-intermediate-layer strategy, and initiate a second pull request for the target file to the target intermediate server cluster.

[0075] The intermediate server cluster is used to receive the second pull request, pull the target file through the source according to the second pull request, transmit the target file to the edge server cluster, and cache the target file.

[0076] The data analysis platform is used to determine what type of service provides the target file corresponding to the first file pull request, wherein the service includes a provincial customer and a nationwide customer.

[0077] Figure 1 FIG. 1 is a schematic diagram of steps of a response method of a content distribution network according to an embodiment of the present disclosure. As shown in FIG. 1, the method is applied to an edge server cluster, and can specifically include the following steps: Figure 1

[0078] Step S11: Obtain a first file pull request.

[0079] The edge server cluster, i.e., the edge cluster, refers to a group of servers or computing resources deployed at the edge of the network, which is mainly used to process user requests and provide content delivery services.

[0080] ​A first file pull request sent by a user is received by a server in the edge server cluster. The first file pull request can be a URL (Uniform Resource Locator), and the user can access the corresponding data resource by sending the URL to the edge server cluster, i.e., the target file in the present disclosure.

[0081] Step S12: In a case where it is determined that the target file pulled by the first file pull request is not cached, a target back-to-intermediate layer strategy of the first file pull request is determined according to the address analysis result, the performance consumption result, and the full-link probe result. The back-to-intermediate layer strategy includes a near principle strategy, a consistent hashing algorithm strategy, and a specified intermediate cluster strategy.

[0082] It is determined whether the target file corresponding to the URL is cached. The target file can be cached in each edge server cluster and / or each intermediate server cluster. If the target file is not cached in the edge server cluster receiving the URL and in each intermediate server cluster, it is determined that the target file is not cached in the system and needs to be pulled from the source station.

[0083] In a case where the target file corresponding to the URL needs to be pulled from the source station, a back-to-intermediate layer strategy from the edge server cluster to the intermediate layer is determined by an algorithm; the back-to-intermediate layer strategy is used to determine a target intermediate server cluster from which the target file is pulled from the source station.

[0084] The judgment basis includes an address analysis result, a performance consumption result, and a full-link probe result. The address analysis result is used to indicate what type of service the target file corresponding to the URL is provided by. The performance consumption result is used to indicate the performance consumed by each intermediate server cluster in the intermediate layer when performing file forwarding. The full-link probe result is used to indicate the network delay and network performance of each node in the system.

[0085] The near principle considers the geographical space, and resources on places closer to the center point are preferentially selected. In the present disclosure, the edge server cluster is taken as the center point, and the intermediate server cluster closest to the geographical position of the edge server cluster is determined as the target intermediate server cluster.

[0086] Step S13: The target intermediate server cluster is determined according to the target back-to-intermediate layer strategy, and a second file pull request is sent to the target intermediate server cluster.

[0087] The target intermediate server cluster corresponding to the URL is determined according to the target back-to-intermediate layer strategy. The edge server cluster initiates a second file pull request for the target file to the target intermediate server cluster, so that the target intermediate server cluster pulls the target file from the source station.

[0088] Step S14: receiving the target file pulled by the target intermediate server cluster from the source station.

[0089] After the target intermediate server cluster pulls the target file from the source station, the target intermediate server cluster transmits the target file to the edge server cluster, so that the edge server cluster can obtain the target file, and the user can obtain the target file corresponding to the URL by initiating the URL.

[0090] By using the embodiments of the present disclosure, in the case that the target file is not cached, the method can intelligently select the optimal back-to-intermediate layer strategy according to the address analysis result, the performance loss result and the full-link probe result, which helps to optimize network traffic and improve the efficiency of file transmission. By selecting a suitable intermediate server cluster for file pulling, load balancing can be achieved, and the situation that some server clusters are overloaded while other clusters are idle can be avoided, which not only improves the resource utilization rate, but also prolongs the service life of the server cluster. The present disclosure can quickly respond to file pulling requests and provide high-quality content distribution services, which can significantly improve the user experience.

[0091] In an optional embodiment, the target back-to-intermediate layer strategy of the first file pulling request is determined according to the address analysis result, the performance loss result and the full-link probe result, including: obtaining the address analysis result from the data analysis platform; determining whether the target service corresponding to the first file pulling request is a first service according to the address analysis result; the first service represents that the source station of the target service and the user are located in the same region; in the case that the target back-to-intermediate layer strategy is a specified intermediate cluster strategy, determining the target intermediate server cluster according to the target back-to-intermediate layer strategy, including: in the case that the target service corresponding to the first file pulling request is the first service, determining the intermediate server cluster located in the same region as the target service as the target intermediate server cluster through the specified intermediate cluster strategy.

[0092] Different URLs correspond to different data resources, and users can access different services through URLs, including provincial customers and national customers, to provide files. The first service can be a provincial customer, and the source station storing the files of the provincial customer and the user of the provincial customer service are located in the same region, and the region can be the level of a province.

[0093] According to the address analysis result, it can be determined which type of service the URL initiated by the user to the edge server cluster corresponds to. In the case that it is determined through the address analysis result that the URL corresponds to the files of the provincial customer, i.e., the files of the first service, it is determined that the source station storing the target file corresponding to the URL and the user initiating the URL are located in the same region.

[0094] In a case where the target service corresponding to the URL is the first service, in order to improve the response speed, an intermediate server cluster located in the same region as the source station needs to be selected, and therefore the specified intermediate cluster strategy is determined as the target intermediate layer strategy, and the intermediate server cluster located in the same region as the target service is determined as the target intermediate server cluster.

[0095] With the embodiments of the present disclosure, by obtaining the address analysis result from the data analysis platform, the method can accurately determine whether the target service corresponding to the first file pulling request is the first service. In a case where the target service is determined as the first service, the target intermediate server cluster is closer to the user in geographical position, significantly reducing the delay of data transmission, thereby improving the response speed. Due to the improvement of the response speed, the user can obtain the required data resources more quickly, thereby improving the user experience.

[0096] In an optional embodiment, the full-link probe result is determined by the following steps: probing each intermediate server cluster to obtain a first probe result; receiving a second probe result transmitted by each intermediate server cluster; and obtaining the full-link probe result through the first probe result and the second probe result.

[0097] RTT (Round Trip Time, round trip time) time is used to evaluate network performance, and directly affects the response speed of user requests and the efficiency of data transmission. RTT time can be used to judge and select the optimal source return strategy to improve the user's access experience.

[0098] The edge server cluster includes a probe module.

[0099] Through the probe module, each edge server cluster probes each intermediate server cluster to determine the RTT time of each edge server cluster to each intermediate server cluster; and the first probe result is obtained through the RTT time of each edge server cluster to each intermediate server cluster.

[0100] The second probe result transmitted by each intermediate server cluster is received; the second probe result includes the RTT time of each intermediate server cluster probing the domain name to each source station.

[0101] The first probe result and the second probe result are summarized to obtain a full-link probe result. The full-link probe result can evaluate and analyze the connection performance between each node in the network.

[0102] By adopting the embodiments of the present disclosure, the best path for data transmission can be accurately determined through RTT time determination of full-link probe results, so as to select the optimal source return strategy and improve data transmission efficiency. Based on real-time full-link probe results, the system can automatically adjust service templates and strategies to adapt to changes in network conditions and ensure good service quality under different loads and network conditions. Through analysis of full-link probe results, the computing and bandwidth resources of the intermediate cluster can be better allocated and utilized, resource waste can be avoided, and the overall system efficiency can be improved.

[0103] In an optional embodiment, the performance loss result is determined by the following steps: collecting domain name service information required by the performance loss model; the domain name service information includes: the average requested file size of each historical first file pull request and the number of each historical first file pull request; analyzing the domain name service information based on the performance loss model to obtain the intermediate layer processor loss and bandwidth consumption; and determining the intermediate layer processor loss and the bandwidth consumption as the performance loss result.

[0104] The edge server cluster includes a data acquisition module.

[0105] Through the data acquisition module, the domain name service information required by the performance loss model is collected; the domain name service information includes: the average requested file size of each historical first file pull request and the number of each historical first file pull request.

[0106] The data acquisition module inputs the domain name service information into the performance loss model for analysis, and the performance loss model outputs the intermediate layer processor loss and the bandwidth consumption. The intermediate layer processor loss represents the CPU consumed when the non-uniformity hash intermediate server cluster forwards files. The bandwidth consumption represents the bandwidth consumed when the non-uniformity hash intermediate server cluster forwards files.

[0107] Based on historical data analysis, the smaller the requested file size and the higher the number of requests, the greater the intermediate layer processor loss; and the bandwidth consumption mainly depends on the product of the average requested file size and the number of requests.

[0108] The following formula is the performance loss model formula 1 proposed by the present disclosure:

[0109] Maximum QPS = (RT / CPU Time * CPU core number * CPU utilization) * (1 / RT)

[0110] Wherein, RT represents response time; CPU Time represents the time spent by CPU in executing a certain task; CPU core number represents the core number of server CPU, which determines how many threads or tasks can be executed simultaneously; and CPU utilization represents the percentage of time that CPU is busy processing tasks in a period of time.

[0111] The performance loss model formula 1 can be deduced reversely, so as to roughly obtain the CPU utilization according to the average number of qps, and the obtained CPU utilization can be used to represent the CPU loss.

[0112] The following formula is a performance loss model formula 2 proposed by the present disclosure:

[0113] Bandwidth consumption = average request file size * number of requests

[0114] The bandwidth consumption is determined through the performance loss model formula 2.

[0115] Figure 2 Fig. 1 is a schematic diagram of determining a performance loss result according to an embodiment of the present disclosure. As shown in Fig. 1, the input of the performance loss model is domain name service information, the domain name service information includes the average request file size of each historical first file pull request and the number of each historical first file pull request, the performance loss model receives the domain name service information, analyzes the domain name service information, and obtains a performance loss result, the performance loss result includes an intermediate layer processor loss and a bandwidth consumption. Figure 2

[0116] In the case of determining the back-to-intermediate layer strategy, the bandwidth consumption needs to be controllable, and the CPU of the intermediate layer needs to be used as much as possible without overloading the CPU.

[0117] By collecting detailed data of historical first file pull requests, the performance requirements in actual business scenarios can be more accurately reflected through the embodiments of the present disclosure. Through the analysis of the performance loss model, potential performance bottlenecks can be predicted and avoided, and system configurations can be adjusted in advance according to the prediction results to ensure that the system selects a suitable back-to-intermediate layer strategy.

[0118] For response speed, the main resources of web clients are small files such as dynamic files and pictures, and the focus is on the speed of file loading completion; the main resources of video clients are medium and large files, and the focus is on the first screen time. The common point of web clients and video clients is that they both focus on whether the first few hundred KB (kilobytes) of files can be quickly loaded, so the loading speed of the first few hundred KB (kilobytes) of files can be mainly considered.

[0119] ​The first capacity can be determined by the file size that needs to be quickly loaded by the customer's attention. The first capacity can be changed according to the change of the user's needs. The capacity represents the size of the space occupied by the file in the storage medium, which represents how much data the file contains. Kilobyte (KB), megabyte (MB), gigabyte (GB), terabyte (TB), etc. can be used to represent the capacity of the file.

[0120] The edge server cluster includes an SVM algorithm module, which uses SVM classification. The principle of SVM classification is to calculate a hyperplane that can distinguish multiple data. It is a virtual concept that can divide the input data into two groups, i.e. positive examples and negative examples. The interval of the hyperplane is measured by examples, so that the bias of the hyperplane is as small as possible, so as to obtain the optimal classification result. Figure 3 is a machine learning SVM related code demo and classification result comparison diagram shown by an embodiment of the present disclosure.

[0121] The input of the SVM algorithm module is the address analysis result, the performance loss result and the full link probe result. Through the machine learning-classification algorithm, the target back intermediate layer strategy is output to achieve the purpose of improving the first capacity of the target file.

[0122] The SVM algorithm module can determine the target SVM algorithm module corresponding to the URL under the current system environment according to actual needs. For example, when the intermediate layer processor loss reaches 5% or the bandwidth consumption exceeds 100MB / s, the SVM algorithm module determines that the target back intermediate layer strategy corresponding to the URL at this time adopts the regular strategy, i.e. the consistent hash algorithm strategy. In the case where the intermediate layer processor loss is less than 1% and the bandwidth consumption is less than 10MB / s, and the full link probe result indicates that the RTT time of each intermediate layer is basically consistent, the SVM algorithm module determines that the target back intermediate layer strategy corresponding to the URL at this time adopts the nearest principle strategy.

[0123] In an optional embodiment, in order to ensure that the first capacity of the target file can be quickly loaded, the present disclosure proposes that the target intermediate server cluster can be determined by the nearest principle strategy, and the first capacity of the target file is pulled from the source station through the target intermediate server cluster. The target intermediate server cluster is the nearest intermediate server cluster to the edge server cluster. If the nearest intermediate server cluster to the edge server cluster is used to obtain the target file from the source station, the intermediate server cluster will not cache the target file, but will transmit the target file to another intermediate server cluster for caching. However, transmitting the target file to another intermediate server cluster for caching will consume additional system performance.

[0124] The disclosure needs to consider the consumption of system performance on the basis of improving the response speed of the front first capacity of the target file. Therefore, in the case of using the nearest principle strategy as the back intermediate layer strategy, based on the size relationship between the target capacity of the target file and the first capacity, there are two cases.

[0125] The edge server cluster further comprises a back intermediate layer module, which is configured to execute a target back intermediate layer strategy and send a second file pulling request to a target intermediate server cluster corresponding to the target back intermediate layer strategy. If the target back intermediate layer strategy is the nearest principle strategy, the back intermediate layer module further needs to determine the target capacity of the target file pulled by the first file pulling request, determine the size relationship between the target capacity of the target file and the first capacity, and according to the size relationship, initiate the second file pulling request to the corresponding target intermediate server cluster. If the target back intermediate layer strategy is the consistent hashing algorithm strategy or the specified intermediate cluster strategy, the size relationship between the target capacity of the target file and the first capacity does not need to be determined.

[0126] The following is the case where the target capacity of the target file is less than or equal to the first capacity.

[0127] In an optional embodiment, in the case where the target back intermediate layer strategy is the nearest principle strategy and the target capacity of the target file is less than or equal to the first capacity, according to the target back intermediate layer strategy, the target intermediate server cluster is determined, and a second file pulling request is sent to the target intermediate server cluster, comprising: determining the intermediate server cluster closest to the edge server cluster as the target intermediate server cluster by the nearest principle; initiating a second file pulling request for the target file to the target intermediate server cluster; and receiving the target file pulled by the target intermediate server cluster from the source station.

[0128] In the case where the target back intermediate layer strategy is the nearest principle strategy and the back intermediate layer module determines that the target capacity of the target file is less than or equal to the first capacity, it means that the entire target file of the current URL is the focus of the user's attention to the loading part. The target file is the loading part that the customer focuses on, and if the target file cannot be quickly loaded, it will cause the user of the customer service to have a poor experience when using the file of the customer.

[0129] Therefore, the intermediate server cluster closest to the edge server cluster is directly determined as the target intermediate server cluster by the nearest principle strategy, and the back intermediate layer module in the edge server cluster initiates a second file pulling request for the entire target file to the target intermediate server cluster. The target intermediate server cluster pulls the entire target file from the source station after receiving the second file pulling request.

[0130] Since the target intermediate server cluster is the intermediate server cluster closest to the edge server, using the target intermediate server cluster to pull the file from the source station improves the response speed of the URL.

[0131] With the embodiments of the present disclosure, by adopting the nearest principle strategy, the intermediate server cluster closest to the edge server is determined as the target intermediate server cluster, thereby ensuring that the path of the target intermediate server cluster to pull the target file from the source station is the shortest, and reducing the time and delay of data transmission. For web page type clients and video type clients, since attention is paid to the loading speed of the first capacity of the file, in the case that the target file is less than or equal to the first capacity, by optimizing the fast loading of the entire target file, the user experience can be significantly improved.

[0132] The following is the case where the target capacity of the target file is greater than the first capacity.

[0133] In an optional embodiment, in the case that the target back intermediate layer strategy is the nearest principle strategy, and the target capacity of the target file is greater than the first capacity, according to the target back intermediate layer strategy, the target intermediate server cluster is determined, and a second file pulling request is sent to the target intermediate server cluster, comprising: determining the intermediate server cluster closest to the edge server cluster as the first target intermediate server cluster by the nearest principle; determining the second target intermediate server cluster corresponding to the first file pulling request by a consistent hashing algorithm; determining the first capacity of the target file as the first file, and determining the remaining capacity of the target file other than the first capacity as the second file; initiating a second file pulling request for the first file to the first target intermediate server cluster to obtain the first file pulled by the first target intermediate server cluster from the source station; initiating a second file pulling request for the second file to the second target intermediate server cluster to obtain the second file pulled by the second target intermediate server cluster from the source station.

[0134] In the case that the target back intermediate layer strategy is the nearest principle strategy, and the back intermediate layer module determines that the target capacity of the target file is greater than the first capacity, the target file can be regarded as two parts, one part is the first capacity that needs to be loaded quickly which is focused on by the client, and the second part is the second capacity other than the first capacity, whether the second capacity is loaded quickly or not, the user cannot perceive well when using the resources provided by the client, therefore, if the second part of the target file cannot be loaded quickly, the user experience will not be reduced.

[0135] Therefore, the target file is divided into two parts based on the first capacity, the first capacity is taken as the first file, and the remaining part is taken as the second file.

[0136] In the case that the target intermediate layer policy is the nearest principle policy and the target capacity of the target file is greater than the first capacity, the target intermediate server cluster includes a first target intermediate server cluster and a second target intermediate server cluster.

[0137] The intermediate server cluster closest to the edge server can be determined as the first target intermediate server cluster by the nearest principle, and the second target intermediate server cluster corresponding to the URL can be determined by the consistent hashing algorithm.

[0138] The back intermediate layer module initiates a second file pulling request for the first file to the first target intermediate server cluster, so that the first target intermediate server cluster pulls the first file of the target file from the source station; and the edge server cluster acquires the first file of the target file through the first target intermediate server cluster.

[0139] Since the first target intermediate server cluster is the intermediate server cluster closest to the edge server cluster, the path for the first target intermediate server cluster to pull the target file from the source station is the shortest, the time and delay of data transmission are reduced, and therefore the speed of the edge server cluster receiving the first file is also improved, and the loading speed of the first capacity of the target file is improved.

[0140] The back intermediate layer module initiates a second file pulling request for the second file to the second target intermediate server cluster, so that the second target intermediate server cluster pulls the second file of the target file from the source station; and the edge server cluster acquires the second file of the target file through the second target intermediate server cluster.

[0141] The second part of the target file that does not need to be accelerated for response is acquired by the second target intermediate server cluster instead of the first target intermediate server cluster, which reduces the amount of data transmitted by the first target intermediate server cluster to the second target intermediate server cluster, thereby avoiding waste of system resources and preventing excessive internal bandwidth from being wasted.

[0142] When the size of the target file exceeds the preset first capacity, the target file is divided into two parts for processing, and different intermediate server clusters are selected for pulling in the embodiments of the present disclosure, which not only improves the response speed of the first capacity of the file, but also realizes reasonable allocation of resources. Since the second target intermediate server cluster is responsible for pulling the second part of the file that is not so urgent, the transmission burden of the first target intermediate server cluster is reduced, thereby improving the overall performance of the system. For the first capacity that users pay more attention to and need to load quickly, the intermediate server cluster closest to the edge server is selected by the nearest principle to pull, which improves the loading speed of this part of content and optimizes the user experience.

[0143] Figure 4 is a schematic diagram of determining a target back middle layer strategy, which is illustrated by an embodiment of the present disclosure. According to Figure 4 As shown in the figure, the edge server cluster includes a data collection module, a detection module, an SVM algorithm module, and a back middle layer module. First, the full link detection result is obtained through the detection module, and the full link detection result includes a first detection result and a second detection result; the first detection result is obtained by detecting the intermediate server cluster through the edge server cluster; and the second detection result is obtained by receiving the transmission of the intermediate server cluster. The domain name service information is collected by the data collection module, and the domain name service information is input into the performance loss model to obtain the performance loss result through the performance loss model. The SVM algorithm module receives the full link detection result, the performance loss result, and the address analysis result provided by the data analysis platform, and outputs one of the nearest principle strategy, the consistent hash algorithm strategy, and the specified intermediate cluster strategy as the target back middle layer strategy. The back middle layer module determines the target intermediate server cluster corresponding to the target back middle layer strategy according to the target back middle layer strategy, and sends a second file pulling request to the target intermediate server cluster to make the target intermediate server cluster pull the target file from the source station.

[0144] Based on the same idea, the present disclosure proposes a response method of a content distribution network, which is applied to an intermediate server cluster and can solve the problems in the related art.

[0145] Figure 5 is a schematic diagram of steps of a response method of a content distribution network, which is illustrated by an embodiment of the present disclosure. According to Figure 5 As shown in the figure, the method is applied to an intermediate server cluster, and can specifically include the following steps:

[0146] Step S51: receiving a second file pulling request sent from an edge server cluster, and obtaining a target file from a source station according to the second file pulling request; the second file pulling request is determined through a nearest principle strategy, a consistent hash algorithm strategy, or a specified intermediate cluster strategy.

[0147] Step S52: transmitting the target file to the edge server cluster.

[0148] The intermediate server cluster includes a cache module, a detection module, and a data transmission module.

[0149] After receiving the second file pulling request, the intermediate server cluster performs corresponding operations to pull the target file corresponding to the second file pulling request from the source station according to the target back middle layer strategy determined by the second file pulling request. The back middle layer strategy includes the nearest principle strategy, the consistent hash algorithm strategy, and the specified intermediate cluster strategy, and one of the nearest principle strategy, the consistent hash algorithm strategy, and the specified intermediate cluster strategy is determined as the target back middle layer strategy.

[0150] After the target file is obtained by the intermediate server cluster, the target file is transmitted to the edge server cluster, so that the user can obtain the target file through the URL.

[0151] By receiving and processing the file pull request from the edge server cluster through the intermediate server cluster, the content distribution process can be more efficiently managed by adopting the embodiments of the present disclosure. The application of the nearest principle strategy, the consistent hashing algorithm strategy and the designated intermediate cluster strategy helps to optimize resource allocation and load balancing. By transmitting the target file from the source station to the intermediate server cluster and then from the intermediate server cluster to the edge server cluster, the distance between the user and the required content can be shortened, which helps to reduce the delay and improve the user experience.

[0152] After the target file is pulled from the source station, the intermediate server cluster will cache the target file. In the case where the second file pull request is determined by the designated intermediate cluster strategy and the consistent hashing algorithm strategy, the intermediate server cluster stores the pulled target file in its own storage medium; in the case where the second file pull request is determined by the nearest principle strategy, the intermediate server cluster stores the pulled target file in another intermediate server cluster.

[0153] The following is the operation performed by the intermediate server cluster after receiving the second file pull request when the second file pull request is determined by the nearest principle strategy.

[0154] In an optional embodiment, when the second file pull request is determined by the nearest principle strategy and the target capacity of the target file is greater than the first capacity, the intermediate server cluster includes: a first target intermediate server cluster and a first target intermediate server cluster; receiving a second file pull request sent from an edge server cluster and obtaining a target file from a source station according to the second file pull request, including the first target intermediate server cluster receiving a second file pull request for the first file, the first file being a file corresponding to the first capacity of the target file; obtaining the first file from the source station according to the second file pull request, and transmitting the first file to the second target intermediate server cluster to make the second target intermediate server cluster cache the first file; the second target intermediate server cluster receives a second file pull request for the second file, the second file being a file corresponding to the remaining capacity of the target file except the first capacity; obtaining the second file from the source station according to the second file pull request and caching the second file.

[0155] In the case that the target capacity of the target file is greater than the first capacity, the edge server cluster initiates a second file pulling request to two target intermediate server clusters, including the first target intermediate server cluster and the second target intermediate server cluster.

[0156] The first target intermediate server cluster is determined by the principle of proximity, and the first target intermediate server cluster is the intermediate server cluster closest to the edge server cluster initiating the second file pulling request.

[0157] The second target intermediate server cluster is determined by a consistent hashing algorithm, and the intermediate server cluster corresponding to the URL sent by the user is determined as the second target intermediate server cluster by the consistent hashing algorithm.

[0158] First, the edge server cluster initiates a second file pulling request for the first file to the first target intermediate server cluster. The first file is the file corresponding to the first capacity of the target file. By obtaining the first file through the first target intermediate server cluster, the transmission distance can be effectively reduced, so that the first capacity of the target file can be loaded at the fastest speed, improving the user experience.

[0159] The first target intermediate server cluster receives the second file pulling request for the first file through the data transmission module included therein. After the data transmission module receives the second file pulling request for the first file, it pulls the first file from the source station and transmits the first file to the edge server cluster, so that the user obtains the first file.

[0160] After obtaining the first file, the data transmission module will also determine whether the intermediate server cluster in which it is located is determined by the consistent hashing algorithm for the URL calculation. If it is determined that it is not, it will determine which intermediate server cluster determined by the consistent hashing algorithm for the URL is, that is, the second target intermediate server cluster in this embodiment. The data transmission module of the first target intermediate server cluster transmits the first file to the data transmission module of the second target intermediate server cluster, and transmits the first file to the cache module of the second target intermediate server cluster through the data transmission module of the second target intermediate server cluster to store the first file.

[0161] Among them, the data transmission module of the first target intermediate server cluster initiates an http3 post request to the data transmission module of the second target intermediate server cluster to complete the transmission of the first file. This is because the intermediate server cluster is an internal cluster and is deployed near the backbone network, and the network condition is good, the use of http3 can improve the transmission speed and single link multiplexing, and less occupy the connection number of the intermediate cluster, prevent affecting the normal business.

[0162] For the rest of the capacity of the target file except the first capacity corresponding to the second file, since this part of the content of the target file is not the content that the customer focuses on and needs to be loaded quickly. Therefore, the edge server cluster will send a second file pull request to the second target intermediate server cluster to pull the second file, instead of the first target intermediate server cluster. This is because the first target intermediate server cluster is not the consistent hash intermediate server cluster corresponding to the URL determined by the consistent hash algorithm, so the first target intermediate server cluster needs to transmit the file obtained by itself to the consistent hash intermediate server cluster corresponding to the URL, that is, the second target intermediate server cluster in the embodiment; the first target intermediate server cluster transmits the file obtained by itself to the second target intermediate server cluster, which will consume additional transmission resources of the system, so for the content in the target file that is not the focus and needs to be accelerated, the consistent hash intermediate server cluster, that is, the second target intermediate server cluster, is used to pull from the source station.

[0163] The second target intermediate server cluster receives the second file pull request for the second file through the cache module included in the second target intermediate server cluster. After the cache module receives the second file pull request for the second file, the second file is pulled from the source station and transmitted to the edge server cluster, so that the user can receive the second file. For the second file, the cache module of the second target intermediate server cluster is directly used to cache the second file.

[0164] By pulling the first file from the source station through the first target intermediate server cluster, and pulling the second file from the source station through the second target intermediate server cluster, the intermediate server cluster completes the acquisition of the target file from the source station, and transmits the first file and the second file to the edge server cluster, so that the user obtains the target file corresponding to the URL.

[0165] When the target file is large, by dividing it into two parts, the first target intermediate server cluster determined by the nearest principle strategy is used to preferentially process and transmit the first file, which can significantly reduce the transmission delay and improve the user experience. According to the importance of the file and the user attention, the resources of the intermediate server cluster are reasonably allocated, the first target intermediate server cluster focuses on quickly transmitting the first file part that the user focuses on, and the second target intermediate server cluster processes the remaining second file part. By segmenting the target file and caching and transmitting it by different intermediate server clusters, distributed storage and access of file content are realized, which helps to reduce the storage pressure of a single server and improve the scalability and fault tolerance of the entire system.

[0166] Figure 6 is a schematic diagram of an intermediate server cluster acquiring a target file from a source station according to an embodiment of the present disclosure. According to the embodiment, the target file is divided into two parts, and the first target intermediate server cluster is used to process and transmit the first part of the target file, and the second target intermediate server cluster is used to process and transmit the second part of the target file. Figure 6As shown, in the case where the target capacity of the target file corresponding to the URL is greater than the first capacity, the acquisition of the target file by the intermediate server cluster includes steps 1-6, and in the case where the target capacity of the target file corresponding to the URL is less than or equal to the first capacity, the acquisition of the target file by the intermediate server cluster includes steps 1-4.

[0167] The first-stage acquisition process of the first file of the first capacity of the target file.

[0168] Step 1: The edge server node initiates a second file pull request for the first file to the first target intermediate server cluster.

[0169] Step 2: The data transmission module of the first target intermediate server cluster receives the second file pull request and pulls the first file from the source station.

[0170] Step 3: After pulling the first file, the data transmission module of the first target intermediate server cluster transmits the first file to the edge server cluster and uses the post request of http3 to transmit the first file to the data transmission module of the second target intermediate server cluster.

[0171] Step 4: After receiving the first file, the data transmission module of the second target intermediate server cluster transmits the first file to the cache module of the second target intermediate server cluster, and the cache of the first file is completed.

[0172] The second-stage acquisition process of the second file corresponding to the remaining capacity of the target file other than the first capacity.

[0173] Step 5: The edge server cluster initiates a second file pull request for the second file to the second target intermediate server cluster.

[0174] Step 6: The cache module of the second target intermediate server cluster receives the second file pull request for the second file, pulls the second file from the source station, transmits the second file to the edge server cluster after the cache module pulls the second file, and caches the second file.

[0175] If the target capacity of the target file corresponding to the URL is less than or equal to the first capacity, the target file is the first file, and only the steps of the first stage are executed to complete the acquisition of the target file by the edge server cluster; if the target capacity of the target file corresponding to the URL is greater than the first capacity, the complete steps of the first stage and the second stage need to be executed to complete the acquisition of the target file by the edge server cluster.

[0176] In an optional embodiment, the method further comprises: detecting each source station to obtain a second detection result; and transmitting the second detection result to each edge server cluster.

[0177] Each intermediate server cluster detects the RTT time of a domain name to each source station. By measuring the RTT time, each intermediate cluster can evaluate the network delay between different source stations.

[0178] Each RTT time obtained is summarized to obtain a second detection result. The intermediate server cluster transmits the second detection result to each edge server cluster, so that the edge server cluster can obtain the full-link detection result of the system, thereby determining the most appropriate back-to-intermediate layer strategy.

[0179] By using the embodiments of the present disclosure, the network conditions between each source station and the intermediate server cluster can be accurately mastered through comprehensive detection. After receiving this information, the edge server cluster can more accurately evaluate the network delay between different source stations and itself, thereby determining the most appropriate back-to-intermediate layer strategy, which helps to optimize the routing path of data and reduce unnecessary network delay and bandwidth consumption.

[0180] The response system of the content distribution network further comprises a data analysis platform, and the data analysis platform is configured to obtain an address analysis result.

[0181] Figure 7 is a step schematic diagram of a content distribution network response method according to an embodiment of the present disclosure. The method is applied to a data analysis platform, and the data analysis platform comprises a data analysis module and a service template module. The data analysis module is mainly used to collect URL granularity user address region information of an edge server cluster and to collect source station address region information provided by a configuration platform; and the service template module is used to adjust a service template.

[0182] According to the method shown in Figure 7 , the method can specifically comprise the following steps:

[0183] Step 71: Each historical first file pulling request is analyzed, and user region information corresponding to each historical first file pulling request is determined according to a user address corresponding to each historical first file pulling request.

[0184] The data analysis platform can analyze each historical first file pulling request.

[0185] The data analysis platform in the data analysis platform can analyze each historical first file pull request, extract the user address corresponding to each historical first file pull request respectively, and the user address can represent the user area information of each user, such as the region name, so as to determine the user area information corresponding to each historical first file pull request respectively.

[0186] Step 72: According to the source station address corresponding to each historical first file pull request respectively, determine the source station area information corresponding to each historical first file pull request respectively.

[0187] The source station address of each source station can be obtained from the configuration platform. Therefore, the data analysis platform in the data analysis platform can determine the source station address corresponding to each historical first file pull request respectively by obtaining information from the configuration platform. The source station address can represent the source station area information of each source station, such as the region name, so that the data analysis platform can determine the source station area information corresponding to each historical first file pull request respectively.

[0188] Step 73: Match the user area information and the source station area information corresponding to each historical first file pull request respectively, and determine the target coincidence rate of the target service corresponding to the first file pull request received by the edge server cluster.

[0189] After obtaining the user area information and the source station area information corresponding to each historical first file pull request respectively, the data analysis platform can match the user area information and the source station area information of each historical first file pull request.

[0190] In the case where the user area information and the source station area information of a certain historical first file pull request represent the same region name, it is determined that the address information of the historical first file pull request coincides.

[0191] Group each historical first file pull request according to different services to determine the historical first file pull request included in each service. The services include provincial customers and national customers.

[0192] Determine which service the target file corresponding to the URL is provided by. Determine the historical first file pull request included in the target service. Respectively determine whether the address information of each historical first file pull request coincides, so as to determine the target coincidence rate corresponding to the target service. For example, the target service includes 100 URLs, respectively determine the user area information and the source station area information corresponding to each URL, and determine whether the address information of each URL coincides; if the address information of 98 URLs coincides, then the target coincidence rate of the target service is determined to be 98%.

[0193] Step 74: obtaining an address analysis result according to the size relationship between the target coincidence rate and the first threshold, the address analysis result at least including that the target service is a first service, the first service representing that the source station of the target service and the user are located in the same region.

[0194] The first threshold is obtained. After determining the target coincidence rate of the target service, the service template module compares the target coincidence rate with the first threshold, and the service template can be used to determine the back-middle layer strategy of the edge server cluster. The address analysis result includes a first service representing a provincial customer and a second service representing a nationwide customer. The source station of the provincial customer and all users of the provincial customer are located in the same region, and the source station of the nationwide customer and all users of the nationwide customer are not located in the same region.

[0195] Specifically, in the case that the target coincidence rate is greater than the first threshold, the target service is determined as the first service; in the case that the target coincidence rate is less than or equal to the first threshold, the target service is determined as the second service.

[0196] Step 75: sending the address analysis result to the edge server cluster.

[0197] After obtaining the address analysis result, the address analysis result is sent to the edge server cluster, so that the edge server cluster can determine the most suitable back-middle layer strategy of the current URL based on the address analysis result.

[0198] By analyzing the historical first file pulling request, the user address and the source station address are obtained, and the user region information and the source station region information are determined accordingly. By matching the user region information and the source station region information, the target coincidence rate of the target service is calculated. According to the size relationship between the target coincidence rate and the first threshold, the target service is determined as the first service or the second service. The system can formulate a more suitable back-middle layer strategy according to the service type, and improve the efficiency and accuracy of content distribution.

[0199] Figure 8 is a schematic diagram of determining an address analysis result according to an embodiment of the present disclosure. According to the embodiment of the present disclosure, the address analysis result is determined by analyzing the historical first file pulling request, obtaining the user address and the source station address, and determining the user region information and the source station region information according to the user address and the source station address. Figure 8As shown, the data analysis module of the data analysis platform can collect that the edge server cluster receives each historical URL, the data analysis module parses each historical URL to obtain the user area information corresponding to each historical URL respectively; the configuration platform can provide each source station address, based on which the data analysis module can determine the source station area information corresponding to each historical URL respectively; by matching the user area information and the source station area information of each historical URL, the target coincidence rate of the target service is determined; and the size relationship between the target coincidence rate and the first threshold value is judged. The service template module adjusts the service template, and sends the address analysis result corresponding to the service template to the edge server cluster.

[0200] Figure 9 is a schematic diagram of an overall access architecture diagram according to an embodiment of the present disclosure.

[0201] In one case, according to Figure 9 As shown, the edge server cluster 2 receives the URL request initiated by the user 2; after determining that there is no cache of the target file corresponding to the URL in the current system, the edge server cluster 2 determines the target back intermediate layer strategy corresponding to the URL; in the case where the target back intermediate layer strategy is the nearest principle strategy, it is judged whether the target capacity of the target file corresponding to the URL is greater than the first capacity; in the case where the judgment is yes, the nearest principle is used to determine that the intermediate server cluster 3 is the first target intermediate server cluster, and the intermediate server cluster 2 is determined as the second target intermediate server cluster through the consistent hashing algorithm.

[0202] Subsequently, the pulling step of the target file corresponding to the URL is performed:

[0203] Step 1: the edge server cluster 2 initiates a file pulling request of the first file to the intermediate server cluster 3;

[0204] Step: the intermediate server cluster 3 obtains the first file of the target file from the source station through the intermediate server cluster 3;

[0205] Step 3: after obtaining the first file, the intermediate server cluster 3 transmits the first file to the intermediate server cluster 2 by initiating an http3+post request, and the intermediate server cluster 2 caches the first file;

[0206] Step 4: the edge server cluster 2 initiates a file pulling request of the second file to the intermediate server cluster 2;

[0207] Step 5: the second file is obtained from the source station through the intermediate server cluster 2; the intermediate server cluster 2 stores the second file.

[0208] The intermediate server cluster 2 completes the cache of the target file corresponding to the URL by caching the first file and the second file.

[0209] If the target file corresponding to the URL is cached in the intermediate server cluster 2, and a user from province A sends the same URL to the edge server cluster 1, the user can directly obtain the target file corresponding to the URL from the intermediate server cluster 2 without having to determine the intermediate layer strategy again.

[0210] In another case, according to Figure 9 As shown, edge server cluster 1 receives the user's URL. After determining that there is no cached target file corresponding to the URL in the current system, edge server cluster 1 determines the back-to-middle-layer strategy corresponding to the URL. If it is learned through the data analysis platform that the target file corresponding to the URL is provided by the first service, that is, a customer within the province, the specified middle cluster strategy is determined as the target back-to-middle-layer strategy. The middle server cluster 1 located in the same region as user 1 and the origin station, that is, province A, is determined as the target middle server cluster, and the target file corresponding to the URL is pulled from the origin station through middle server cluster 1.

[0211] This disclosure also provides an electronic device, with reference to... Figure 10 , Figure 10 This is a schematic diagram of an electronic device according to an embodiment of this disclosure. For example... Figure 10 As shown, the electronic device 1000 includes a memory 1010 and a processor 1020. The memory 1010 and the processor 1020 are connected via a bus. The memory 1010 stores a computer program that can run on the processor 1020 to implement the steps in the response method of the content distribution network disclosed in this embodiment.

[0212] This disclosure also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the response method of the content delivery network disclosed in this disclosure.

[0213] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the response method of the content delivery network disclosed in this disclosure.

[0214] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0215] Those skilled in the art will appreciate that embodiments of the disclosure can be supplied as a method, an apparatus, or a computer program product. Thus, embodiments of the disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the disclosure can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, and the like) embodying computer readable program code.

[0216] Embodiments of the disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, electronic devices, and computer program products according to embodiments of the disclosure. It will be understood that each flow and / or block of the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flow or flows and / or blocks. Figure 1 one or more functions specified in the flow or flows and / or blocks.

[0217] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing terminal apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow or flows and / or blocks. Figure 1 one or more functions specified in the flow or flows and / or blocks. Figure 1 one or more functions specified in the flow or flows and / or blocks.

[0218] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal apparatus to cause a series of operational steps to be performed on the computer or other programmable terminal apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal apparatus provide steps for implementing the flow or flows and / or blocks. Figure 1 one or more functions specified in the flow or flows and / or blocks. Figure 1 one or more functions specified in the flow or flows and / or blocks.

[0219] While certain embodiments of the disclosure have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the disclosure. Embodiments of the disclosure can include other variations as will become apparent to those skilled in the art, once given the benefit of this disclosure. Accordingly, the attached claims are intended to embrace all such variations as fall within the scope of the disclosure.

[0220] The response method of the content distribution network provided by the present disclosure is described in detail above, and the principles and implementation modes of the present disclosure are described by applying specific examples in this paper. The above description of the embodiments is only used to help understand the method of the present disclosure and its core idea; at the same time, for those skilled in the art, according to the idea of the present disclosure, the specific implementation modes and application ranges will be changed, and the above description should not be understood as a limitation of the present disclosure.

Claims

1. A response method for a content delivery network, characterized in that, Applied to edge server clusters, including: Get the first file fetch request; If it is determined that the target file to be pulled by the first file pull request is not cached, the target return-to-intermediate-layer strategy for the first file pull request is determined based on the address analysis results, performance loss results, and end-to-end probing results. The return-to-intermediate-layer strategy includes: the nearest principle strategy, the consistent hashing algorithm strategy, and the designated intermediate cluster strategy. Based on the target back-to-middle layer strategy, the target middle server cluster is determined, and a second file pull request is sent to the target middle server cluster. Receive the target file pulled from the source site by the target intermediate server cluster; When the target return-to-intermediate-layer strategy is a proximity-based strategy and the target file size is greater than the first size, the target intermediate server cluster is determined according to the target return-to-intermediate-layer strategy, and a second file pull request is sent to the target intermediate server cluster, including: Based on the principle of proximity, the intermediate server cluster closest to the edge server cluster is determined as the first target intermediate server cluster; the second target intermediate server cluster corresponding to the first file retrieval request is determined using a consistent hashing algorithm. The first portion of the target file is defined as the first file, and the remaining portion of the target file excluding the first portion is defined as the second file. Initiate a second file pull request for the first file to the first target intermediate server cluster, so as to obtain the first file pulled by the first target intermediate server cluster from the source site; Initiate a second file pull request to the second target intermediate server cluster to obtain the second file pulled from the source site by the second target intermediate server cluster.

2. The method according to claim 1, characterized in that, Based on the address analysis results, performance loss results, and end-to-end probing results, the target back-to-intermediate-layer strategy for the first file fetch request is determined, including: Obtain address analysis results from the data analysis platform; Based on the address analysis results, it is determined whether the target service corresponding to the first file retrieval request is the first service; the first service indicates that the origin server and user of the target service are located in the same region; When the target return-to-intermediate-layer strategy is a specified intermediate cluster strategy, the target intermediate server cluster is determined according to the target return-to-intermediate-layer strategy, including: If the target service corresponding to the first file retrieval request is the first service, the intermediate server cluster located in the same region as the target service is determined as the target intermediate server cluster through the specified intermediate cluster strategy.

3. The method according to claim 1, characterized in that, When the target return-to-intermediate-layer strategy is a proximity-based strategy and the target file's target size is less than or equal to the first size, the target intermediate server cluster is determined according to the target return-to-intermediate-layer strategy, and a second file retrieval request is sent to the target intermediate server cluster, including: Based on the principle of proximity, the intermediate server cluster closest to the edge server cluster is determined as the target intermediate server cluster; Initiate a second file retrieval request for the target file to the target intermediate server cluster; to receive the target file retrieved by the target intermediate server cluster from the source site.

4. The method according to claim 1, characterized in that, The end-to-end detection results are determined through the following steps: Probe each intermediate server cluster to obtain the first probe results; Receive the second probe results transmitted from each intermediate server cluster; The full-link detection results are obtained by combining the first detection result and the second detection result.

5. The method according to claim 1, characterized in that, The performance loss result is determined through the following steps: Collect the domain name service information required for the performance loss model; the domain name service information includes: the average file size of each historical first file fetch request and the number of each historical first file fetch request; Based on the performance loss model, the domain name service information is analyzed to obtain the intermediate layer processor loss and bandwidth consumption. The intermediate layer processor loss and the bandwidth consumption are determined as the performance loss result.

6. A response method for a content delivery network, characterized in that, Applications include intermediate server clusters, including: Receive a second file pull request from the edge server cluster, and retrieve the target file from the source server according to the second file pull request; the second file pull request is determined by the proximity principle strategy, the consistent hashing algorithm strategy, or the specified intermediate cluster strategy; The target file is transferred to the edge server cluster; When the second file retrieval request is determined by the proximity principle strategy and the target size of the target file is greater than the first size, the intermediate server cluster includes: a first target intermediate server cluster and a second target intermediate server cluster. The first target intermediate server cluster is determined by the proximity principle, and the second target intermediate server cluster is determined by the consistent hashing algorithm. Receive the second file pull request sent from the edge server cluster, and obtain the target file from the source server according to the second file pull request, including: The first target intermediate server cluster receives a second file pull request for the first file, where the first file is the file corresponding to the first first capacity of the target file; According to the second file retrieval request, the first file is obtained from the source server and transmitted to the second target intermediate server cluster so that the second target intermediate server cluster caches the first file. The second target intermediate server cluster receives a second file pull request for the second file, wherein the second file is the file corresponding to the remaining capacity of the target file excluding the first capacity; The second file is retrieved from the origin server according to the second file retrieval request, and the second file is cached.

7. The method according to claim 6, characterized in that, Also includes: Each source station was probed, and the second probe result was obtained; The second detection result is transmitted to each of the edge server clusters.

8. A response method for a content delivery network, characterized in that, Applications in data analytics platforms include: Each historical first file retrieval request is parsed, and the user region information corresponding to each historical first file retrieval request is determined based on the user address corresponding to each historical first file retrieval request. Based on the source server address corresponding to each historical first file retrieval request, determine the source server region information corresponding to each historical first file retrieval request; The user region information and the source site region information corresponding to each of the historical first file retrieval requests are matched to determine the target overlap rate of the target service corresponding to the first file retrieval request received by the edge server cluster. Based on the relationship between the target overlap rate and the first threshold, an address analysis result is obtained. The address analysis result includes at least the target service being a first service, where the first service indicates that the origin station and user of the target service are located in the same area. The address analysis result includes a first service representing customers within a province and a second service representing customers nationwide. If the target overlap rate is greater than the first threshold, the target service is determined as the first service. If the target overlap rate is less than or equal to the first threshold, the target service is determined as the second service. The address analysis results are sent to the edge server cluster.

Images