A request processing method and device, and a storage medium

Abstract

The application discloses a request processing method and device and a storage medium. The method comprises the following steps: in the case that a target edge server in an edge server cluster receives a user request and executes an action indicated by continuity action information for completing the user request, the running state of the target edge server is listened to; in the case that the running state is an abnormal state, the characteristic information of the user request is acquired from a preset shared database, and a standby edge server is determined from the edge server cluster based on the characteristic information; the continuity action information stored in the preset shared database is synchronized to the standby edge server, and the action not executed by the target edge server in the action indicated by the continuity action information is continued to be executed by using the standby edge server. Through the technical scheme, the request processing capability is improved.

Description

Technical Field

[0001] This application relates to the field of mobile communication technology, and in particular to a request processing method, apparatus and storage medium. Background Technology

[0002] With the rapid development of Internet technology, end users have increasingly higher requirements for the response time and processing capabilities of access servers, which has driven the development of mobile terminals and "edge computing". Edge servers have also emerged. Currently, end users can directly interact with edge servers to process their access requests, which can effectively alleviate the computing pressure on the terminal.

[0003] Currently, for user requests generated during the interaction between end users and edge servers, the edge mesh service, while ensuring load balancing, distributes a series of user requests to the same edge server for processing. However, if the edge server executing the user request encounters an error, the unexecuted user requests will still be forwarded to the malfunctioning edge server. Only after the edge server administrator monitors the edge server's error on the console and restores the edge server can the edge server continue to process the user request. This results in interruptions in the edge server's user request processing and a low capacity for handling user requests. Summary of the Invention

[0004] To address the aforementioned technical problems, embodiments of the present invention aim to provide a request processing method, apparatus, and storage medium that, in the event of a malfunction in the target edge server processing a user request, directly selects a suitable backup edge server from the edge server cluster to continue processing the user request, ensuring uninterrupted processing of user requests and improving request processing capabilities.

[0005] The technical solution of this invention is implemented as follows:

[0006] This invention provides a request processing method, the method comprising:

[0007] When a target edge server in an edge server cluster receives a user request and executes an action indicated by continuous action information to complete the user request, the operating status of the target edge server is monitored.

[0008] In the event that the operating state is abnormal, the feature information of the user request is obtained from the preset shared database, and a backup edge server is determined from the edge server cluster based on the feature information.

[0009] Synchronize the continuity action information stored in the preset shared database to the backup edge server, and continue to execute the action indicated by the continuity action information which is not executed by the target edge server.

[0010] In the above method, before the target edge server in the edge server cluster receives the user request, the method further comprises:

[0011] Forwarding the user request to the edge grid service based on routing rules by using the edge gateway;

[0012] Sending the user request to the target edge server based on a preset load balancing strategy by using the edge grid service.

[0013] In the above method, the feature information includes: the type of the user request, the public key and signature information carried, the parameter information, the domain name information, the source information, the heat, and the user type.

[0014] In the above method, determining the backup edge server from the edge server cluster based on the feature information comprises:

[0015] Determining the edge server type matched with the feature information as a target type by using a preset classification model.

[0016] Selecting an edge server of the target type from the edge server cluster as the backup edge server.

[0017] In the above method, before determining the edge server type matched with the feature information as a target type by using a preset classification model, the method further comprises:

[0018] Obtaining sample features corresponding to a sample request, and performing feature weighting on the sample features by using a preset feature weight to obtain weighted sample features;

[0019] Determining the server type corresponding to the sample features based on the weighted sample features by using a to-be-trained classification model;

[0020] Calculating loss information between the server type and a target server type preset for the sample request;

[0021] Model training is performed on the to-be-trained classification model based on the loss information to obtain the preset classification model.

[0022] The application provides a request processing device, comprising:

[0023] The listening module is configured to listen to a running state of the target edge server in a case where the target edge server receives a user request and performs an action indicated by continuity action information for completing the user request;

[0024] The determining module is configured to acquire feature information of the user request from a preset shared database and determine a backup edge server from the edge server cluster based on the feature information in a case where the running state is an abnormal state.

[0025] The executing module is configured to synchronize the continuity action information stored in the preset shared database to the backup edge server and continue to perform an action that is not performed by the target edge server in the action indicated by the continuity action information by using the backup edge server.

[0026] In the above device, the forwarding module is further configured to forward the user request to an edge grid service based on a routing rule by using an edge gateway and send the user request to the target edge server based on a preset load balancing strategy by using the edge grid service.

[0027] In the above device, the feature information includes a type of the user request, public key and signature information carried by the user request, parameter information, domain name information, source information, heat, and a user type.

[0028] In the above device, the determining module is specifically configured to determine an edge server type matched with the feature information by using a preset classification model, determine the edge server type as a target type, and select an edge server of the target type from the edge server cluster as the backup edge server.

[0029] In the above device, the training module is further configured to acquire sample feature corresponding to a sample request, perform feature weighting on the sample feature by using a preset feature weight to obtain weighted sample feature, determine a server type corresponding to the sample information based on the weighted sample feature by using a to-be-trained classification model, calculate loss information between the server type and a target server type preset for the sample request, and perform model training on the to-be-trained classification model based on the loss information to obtain the preset classification model.

[0030] The application provides a request processing device, which comprises a processor, a memory and a communication bus.

[0031] The communication bus is configured to realize communication connection between the processor and the memory.

[0032] The processor is configured to execute a request processing program stored in the memory to implement the request processing method.

[0033] The application provides a computer readable storage medium storing one or more programs, which can be executed by one or more processors to implement the request processing method.

[0034] The application provides a request processing method, device and storage medium, the method comprising: in the case that a target edge server in an edge server cluster receives a user request and performs an action indicated by continuity action information for completing the user request, monitoring the running state of the target edge server; in the case that the running state is an abnormal state, obtaining feature information of the user request from a preset shared database, and determining a backup edge server from the edge server cluster based on the feature information; synchronizing the continuity action information stored in the preset shared database to the backup edge server, and continuing to perform the action not performed by the target edge server in the action indicated by the continuity action information by using the backup edge server. The technical scheme provided by the application directly selects a suitable backup edge server from the edge server cluster to continue to process the user request in the case that the target edge server processing the user request is abnormal, ensures uninterrupted processing of the user request, and improves the request processing capability. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 A flowchart of a request processing method provided by an embodiment of the application is shown in the figure;

[0036] Figure 2 A flowchart of an exemplary user request forwarding provided by an embodiment of the application is shown in the figure;

[0037] Figure 3 A flowchart of an exemplary request processing provided by an embodiment of the application is shown in the figure;

[0038] Figure 4 A structure diagram of a request processing device provided by an embodiment of the application is shown in the figure Figure 1 ;

[0039] Figure 5 A structure diagram of a request processing device provided by an embodiment of the application is shown in the figure Figure 2 . DETAILED DESCRIPTION

[0040] The technical solutions in the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings.

[0041] The present application provides a request processing method, applied to a request processing device, Figure 1 A flowchart of a request processing method provided by an embodiment of the present application is shown in FIG. 1. As shown in the figure, the method mainly includes the following steps: Figure 1

[0042] S101, in the case that the target edge server in the edge server cluster receives a user request and performs an action indicated by the continuity action information for completing the user request, the running state of the target edge server is monitored.

[0043] In the embodiment of the present application, the request processing device monitors the running state of the target edge server in the case that the target edge server in the edge server cluster receives a user request and performs an action indicated by the continuity action information for completing the user request.

[0044] It should be noted that, in the embodiment of the present application, the request processing device can directly receive a user request from the target edge server in the edge server cluster, wherein the user request can be sent by a third-party user, and the third-party user can be a mobile phone, a computer, a tablet, or other devices with communication function.

[0045] Specifically, in the embodiment of the present application, before the request processing device receives the user request from the target edge server in the edge server cluster, the request processing device can further perform the following steps: forwarding the user request to the edge grid service based on a routing rule by using the edge gateway; and sending the user request to the target edge server based on a preset load balancing strategy by using the edge grid service.

[0046] It should be noted that, in the embodiment of the present application, in the case that the request processing device receives the user request sent by the third-party user, the request processing device forwards the user request to the edge grid service based on a routing rule by using the edge gateway, and then sends the user request to the target edge server based on a preset load balancing strategy by using the edge grid service; wherein the edge gateway is used to manage the edge server and open a unified gateway interface for the third-party user; the edge grid service is used as an access interface of the edge server to the outside, and can realize mutual access between all edge servers in the same edge server cluster; and the preset load balancing strategy can be a traffic strategy.

[0047] Figure 2 ​This is a schematic diagram illustrating an exemplary user request forwarding process provided in an embodiment of the present invention. Figure 2 As shown, the request processing device uses the edge gateway to forward user requests to the edge mesh service based on routing rules. Then, using the edge mesh service, the target edge server is selected from the edge server cluster based on traffic policies. Finally, the user request is sent to the target edge server. The edge server in the edge server cluster can be a Pod, which is the smallest basic unit for deploying applications or services in the Container Orchestration Engine (Kubernetes). An edge server can encapsulate multiple application containers or a single application container.

[0048] It should be noted that, in the embodiments of the present invention, while the request processing device receives a user request from the target edge server in the edge server cluster and executes the action indicated by the continuous action information to complete the user request, it also monitors the running status of the target edge server.

[0049] It should be noted that, in the embodiments of the present invention, if the target edge server is in a normal operating state, the request processing device will continuously use the target edge server to execute the action indicated by the continuous action information to complete the user request until the action indicated by the continuous action information is completed.

[0050] S102. In the case of an abnormal running state, obtain the feature information of the user request from the preset shared database, and determine the backup edge server from the edge server cluster based on the feature information.

[0051] In an embodiment of the present invention, when the request processing device is in an abnormal operating state, it obtains the feature information of the user request from a preset shared database, and determines a backup edge server from the edge server cluster based on the feature information.

[0052] It should be noted that, in the embodiments of the present invention, when the request processing device detects that the target edge server is in an abnormal operating state, it will obtain the feature information of the user request from a preset shared database, wherein the preset shared database stores the feature information of the user request.

[0053] Specifically, in embodiments of the present invention, the feature information includes: the type of user request, the public key and signature information carried, parameter information, domain name information, source information, popularity, and user type.

[0054] It should be noted that, in the embodiments of the present invention, when the request processing device detects that the target edge server is in an abnormal running state, a large number of user requests may be made on the target edge server in an abnormal running state. However, these user requests are of different types and access different container applications in the target edge server. For example, the request type can be GET: requesting to obtain a specific resource from the target edge server, POST: creating a new resource on the target edge server, or PATCH: updating a resource on the target edge server. The container application accessed in the target edge server can be a database container application, an e-commerce platform container application, or a billing system container application.

[0055] It should be noted that, in the embodiments of the present invention, when the request processing device detects that the target edge server is in an abnormal operating state, it needs to forward the user requests executed on the target edge server to other suitable edge servers. In addition, the request processing device needs to shut down the target edge server in an abnormal operating state and perform abnormal cause diagnosis and repair.

[0056] It should be noted that, in the embodiments of the present invention, after obtaining the feature information of the user request, the request processing device will determine the backup edge server from the edge server cluster based on the feature information.

[0057] Specifically, in an embodiment of the present invention, the request processing device determines a backup edge server from the edge server cluster based on feature information, including: using a preset classification model to determine the type of edge server that matches the feature information and determine it as the target type; and selecting an edge server of the target type from the edge server cluster and determining it as the backup edge server.

[0058] It should be noted that, in the embodiments of the present invention, the request processing device will use a preset classification model to determine the type of edge server that matches the feature information and determine it as the target type. The target type can be large capacity, high processing efficiency, or good security, or other types. The specific target type can be set according to the actual situation and application scenario.

[0059] It should be noted that, in the embodiments of the present invention, after the request processing device inputs the feature information into the preset classification model, it can directly obtain the edge server type that matches the feature information and determine it as the target type. Then, based on the target type, an edge server of the target type is selected from the edge server cluster as a backup edge server. In other words, the selection of the backup edge server takes into account the feature information of the user request, and then selects the backup edge server most suitable for forwarding the user request from the edge server cluster based on the feature information carried by the user request, thus avoiding the problem that the selected backup edge server cannot handle the user request instantly.

[0060] Specifically, in embodiments of the present invention, the request processing device uses a preset classification model to determine the edge server type that matches the feature information. Before determining it as the target type, it may also perform the following steps: obtain the sample features corresponding to the sample request, and use preset feature weights to weight the sample features to obtain weighted sample features; use the classification model to be trained to determine the server type corresponding to the sample features based on the weighted sample features; calculate the loss information between the server type and the preset target server type for the sample request; and train the classification model to be trained based on the loss information to obtain the preset classification model.

[0061] It should be noted that, in the embodiments of the present invention, the request processing device obtains the sample feature x corresponding to the sample request, where x = (x 1 ,x 2 ,…,x n ), representing an n-dimensional feature vector composed of the sample features corresponding to the request.

[0062] It should be noted that, in the embodiments of the present invention, the specific sample characteristics are the type of sample request (GET, POST, PATCH, etc.), the public key and signature information carried, parameter information, domain name information, source information, popularity, and user type. Among them, the user type may include Internet users, group government and enterprise users, group approval users, and test users. The source information is whether it comes from outside the cluster or inside the cluster.

[0063] It should be noted that, in the embodiments of the present invention, after obtaining the sample features, the request processing device selects k nearest neighbor samples of a certain sample x from the sample features, and presets a target server type y for each nearest neighbor sample. Specifically, the target server type can be characterized by large capacity, high processing efficiency, and good security, and is specifically represented as follows: Among them, y i The target server type is preset; then, the maximum interval of these 3k samples on the f-th dimension feature is calculated, and the specific calculation formula is shown in formula (1):

[0064]

[0065] Among them, v f Let v represent the maximum margin of the 3k nearest neighbor samples of sample x with different target server types on the f-th dimension feature, where k and i represent the number of nearest neighbors of sample x and the i-th nearest neighbor, respectively. A larger margin indicates that the target server type is more likely to change as the f-th dimension feature value changes, meaning the f-th dimension feature is more important to the target server type. Therefore, the maximum margin across all dimensions of the sample can be expressed as v = (v... 1 ,v 2 ,…,v n Then, the normalized interval vector is used as the preset feature weight w = (w 1 ,w 2 ,…,w n Then the weighted sample features can be represented as (w T x).

[0066] It should be noted that, in the embodiments of the present invention, after obtaining the weighted sample features, the request processing device inputs the weighted sample features into the classification model to be trained to obtain the server type corresponding to the sample features; then, it calculates the loss information between the server type and the preset target server type y for the sample request; based on the loss information, it trains the classification model to be trained to obtain the preset classification model; wherein, the classification model to be trained can be a support vector machine (SVM) model.

[0067] S103. Synchronize the continuous action information stored in the preset shared database to the backup edge server, and use the backup edge server to continue to execute the actions that the target edge server has not executed in the actions indicated by the continuous action information.

[0068] In an embodiment of the present invention, the request processing device synchronizes the continuous action information stored in the preset shared database to the backup edge server, and uses the backup edge server to continue to execute the actions that the target edge server has not executed in the actions indicated by the continuous action information.

[0069] It should be noted that, in the embodiments of the present invention, after the request processing device selects a backup edge server from the edge server cluster, it first synchronizes the preset stored continuous action information to the backup edge server. The preset shared database stores real-time continuous action information. In this way, when the request processing device uses the backup edge server to execute the action indicated by the continuous action information, it will not repeat the action already executed by the target edge server. Furthermore, since the user request feature information and continuous action information are stored in the preset shared database in real time, when using the backup edge server to continue executing the action indicated by the continuous action information that the target edge server has not executed, the real-time continuous action information can be directly obtained from the preset shared database without re-identification, thereby ensuring uninterrupted access for third-party users.

[0070] Figure 3 This is a schematic diagram illustrating an exemplary request processing flow according to an embodiment of the present invention. For example... Figure 3 As shown, when a request processing device receives a user request from a third-party user, it forwards the request to the edge mesh service via the edge gateway. Then, according to the load balancing strategy, it forwards the request to the target edge server to execute the actions indicated by the continuous action information to complete the user request. If the target edge server remains in a normal operating state while the request processing device is executing the actions indicated by the continuous action information, then all actions indicated by the continuous action information will be completed by the target edge server. If, during the execution of the five actions indicated by the continuous action information, the target edge server becomes abnormal while executing the third action, the request processing device retrieves feature information from a preset shared database. Based on this feature information, it determines a backup edge server from the edge server cluster and synchronizes the continuous action information stored in the preset shared database to the backup edge server. Using the backup edge server, execution continues from the third action indicated by the continuous action information until all actions indicated by the continuous action information have been completed.

[0071] This invention provides a request processing method, comprising: when a target edge server in an edge server cluster receives a user request and executes actions indicated by continuous action information to complete the user request, monitoring the operating status of the target edge server; if the operating status is abnormal, retrieving feature information of the user request from a preset shared database, and determining a backup edge server from the edge server cluster based on the feature information; synchronizing the continuous action information stored in the preset shared database to the backup edge server, and using the backup edge server to continue executing the actions not executed by the target edge server in the continuous action information indication. The request processing method provided by this invention, when the target edge server forwarding the user request malfunctions, directly selects a suitable backup edge server from the edge server cluster to continue executing the actions not executed by the target edge server, ensuring uninterrupted processing of user requests and improving request processing capabilities.

[0072] This invention provides a request processing apparatus. Figure 4 A schematic diagram of the structure of a request processing device provided in an embodiment of the present invention. Figure 1 .like Figure 4 As shown, it includes:

[0073] The monitoring module 401 is used to monitor the running status of the target edge server when the target edge server in the edge server cluster receives a user request and performs an action indicated by the continuous action information to complete the user request.

[0074] The determination module 402 is used to obtain the feature information of the user request from a preset shared database when the running state is abnormal, and determine a backup edge server from the edge server cluster based on the feature information.

[0075] The execution module 403 is used to synchronize the continuous action information stored in the preset shared database to the backup edge server, and use the backup edge server to continue to execute the actions that the target edge server has not executed in the actions indicated by the continuous action information.

[0076] Optionally, the request processing device further includes a forwarding module (not shown in the figure), used to forward the user request to the edge mesh service based on routing rules using the edge gateway; and to send the user request to the target edge server using the edge mesh service based on a preset load balancing strategy.

[0077] Optionally, the determining module 402 is specifically used to use a preset classification model to determine the type of edge server that matches the feature information and determine it as the target type; and to select an edge server of the target type from the edge server cluster and determine it as the backup edge server.

[0078] Optionally, the feature information includes: the type of the user request, the public key and signature information carried, parameter information, domain name information, source information, popularity, and user type.

[0079] Optionally, the request processing device further includes a training module (not shown in the figure), used to acquire sample features corresponding to the sample request, and to perform feature weighting on the sample features using preset feature weights to obtain weighted sample features; to determine the server type corresponding to the sample features based on the weighted sample features using a classification model to be trained; to calculate the loss information between the server type and the preset target server type for the sample request; and to train the classification model to be trained based on the loss information to obtain the preset classification model.

[0080] This invention provides a request processing apparatus. Figure 5 A schematic diagram of the structure of a request processing device provided in an embodiment of the present invention. Figure 2 .like Figure 5 As shown, the request processing device includes: a processor 501, a memory 502, and a communication bus 503;

[0081] The communication bus 503 is used to realize the communication connection between the processor 501 and the memory 502;

[0082] The processor 501 is used to execute the request processing program stored in the memory 502 to implement the above-described request processing method.

[0083] This invention provides a request processing apparatus. When a target edge server in an edge server cluster receives a user request and executes actions indicated by continuous action information to complete the request, the apparatus monitors the operating status of the target edge server. If the operating status is abnormal, the apparatus retrieves feature information of the user request from a preset shared database and, based on this feature information, determines a backup edge server from the edge server cluster. The continuous action information stored in the preset shared database is synchronized to the backup edge server, and the backup edge server is used to continue executing the actions indicated by the continuous action information that the target edge server failed to perform. This request processing apparatus, when the target edge server processing the user request experiences an operational anomaly, directly selects a suitable backup edge server from the edge server cluster to continue processing the user request, ensuring uninterrupted processing of user requests and improving request processing capabilities.

[0084] This invention provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the aforementioned request processing method. The computer-readable storage medium can be volatile memory, such as random-access memory (RAM); or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or it can be a device comprising one or any combination of the above-mentioned memories, such as a mobile phone, computer, tablet device, personal digital assistant, etc.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0086] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of 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 apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0089] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this utility application should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A request processing method, characterized in that, The method includes: When a target edge server in an edge server cluster receives a user request and executes an action indicated by continuous action information to complete the user request, the operating status of the target edge server is monitored. In the event that the operating state is abnormal, the feature information of the user request is obtained from the preset shared database, and a backup edge server is determined from the edge server cluster based on the feature information. The continuous action information stored in the preset shared database is synchronized to the backup edge server, and the backup edge server is used to continue executing the actions that the target edge server has not executed in the actions indicated by the continuous action information; the preset shared database stores real-time continuous action information.

2. The method according to claim 1, characterized in that, Before the target edge server in the edge server cluster receives the user request, the method further includes: Using an edge gateway, based on routing rules, the user request is forwarded to the edge mesh service; Using edge mesh services and based on a preset load balancing strategy, the user request is sent to the target edge server.

3. The method according to claim 1, characterized in that, The feature information includes: the type of the user request, the public key and signature information carried, parameter information, domain name information, source information, popularity, and user type.

4. The method according to claim 1, characterized in that, The step of determining a backup edge server from the edge server cluster based on the feature information includes: Using a preset classification model, the type of edge server that matches the feature information is determined and identified as the target type; From the edge server cluster, one edge server of the target type is selected and designated as the backup edge server.

5. The method according to claim 4, characterized in that, Before determining the type of edge server matching the feature information using a preset classification model and identifying it as the target type, the method further includes: Obtain the sample features corresponding to the sample request, and use preset feature weights to perform feature weighting on the sample features to obtain the weighted sample features; Using the classification model to be trained, the server type corresponding to the sample features is determined based on the weighted sample features; Calculate the loss information between the server type and the preset target server type for the sample request; Based on the loss information, the classification model to be trained is trained to obtain the preset classification model.

6. A request processing apparatus, characterized in that, include: The monitoring module is used to monitor the running status of the target edge server when the target edge server in the edge server cluster receives a user request and executes an action indicated by the continuous action information to complete the user request. The determination module is used to obtain the feature information of the user request from a preset shared database when the running state is abnormal, and determine a backup edge server from the edge server cluster based on the feature information. The execution module is used to synchronize the continuous action information stored in the preset shared database to the backup edge server, and use the backup edge server to continue to execute the actions that the target edge server has not executed in the actions indicated by the continuous action information; the preset shared database stores real-time continuous action information.

7. The apparatus according to claim 6, characterized in that, The determining module is specifically used to use a preset classification model to determine the type of edge server that matches the feature information and determine it as the target type; and to select an edge server of the target type from the edge server cluster and determine it as the backup edge server.

8. The apparatus according to claim 7, characterized in that, It also includes a training module. The training module is used to acquire sample features corresponding to the sample request, and to perform feature weighting on the sample features using preset feature weights to obtain weighted sample features; to use the classification model to be trained to determine the server type corresponding to the sample features based on the weighted sample features; and to calculate the loss information between the server type and the preset target server type for the sample request. Based on the loss information, the classification model to be trained is trained to obtain the preset classification model.

9. A request processing apparatus, characterized in that, include: Processor, memory, and communication bus; The communication bus is used to realize the communication connection between the processor and the memory; The processor is configured to execute a request processing program stored in the memory to implement the request processing method according to any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the request processing method according to any one of claims 1-5.

Images