Data caching control method, electronic device and storage medium

Abstract

Embodiments of the present disclosure disclose a data caching control method, an electronic device, and a storage medium. The method includes: obtaining cache system information, and determining an application server, a plurality of first cache server clusters corresponding to the application server, and a plurality of second cache server clusters according to the cache system information; obtaining service type information of a target service corresponding to data caching by the application server and service characteristic information of the target service; obtaining a pre-trained cache configuration model, and using the service type information and the service characteristic information as inputs to input into the cache configuration model to obtain cache configuration information; and sending the cache configuration information to the application server. This technical solution can configure the order of cached data for the target service corresponding to data caching for reading, ensure a high efficiency of reading cached data, and improve the user experience.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data cache control, and in particular, to a data cache control method, an electronic device, and a storage medium. Background Art

[0002] In recent years, with the development of communication technologies, people's dependence on the network has been increasing in all aspects of daily life. The terminal devices used by people can access the network through network devices closer to themselves to use corresponding network applications. With the development of the Internet, the user scale and data scale are getting larger and larger, and higher requirements are put forward for the stability of network applications. Among them, the cache system is generally used to support network applications, and higher requirements are placed on the stability of the cache system.

[0003] In related technologies, a distributed cache system with relatively high stability can be configured for network applications. The distributed cache system usually includes an application server and multiple cache server clusters. When the application server reads cache data, it can sequentially read cache data from multiple cache server clusters according to the order specified by the pre-set cache configuration information.

[0004] Although the stability of the distributed cache system is good in the above solution, since the cache configuration information is pre-set, the order of reading cache data is also pre-set. When the network application executes different types of services, it is impossible to adjust the configuration of the primary cluster and the secondary cluster, that is, the order of reading cache data, for different types of services, resulting in low efficiency of reading cache data and poor user experience. Summary of the Invention

[0005] To solve the problems in related technologies, embodiments of the present disclosure provide a data cache control method, an electronic device, and a storage medium.

[0006] In a first aspect, an embodiment of the present disclosure provides a data cache control method. The method is applied to a data cache control device and includes:

[0007] Obtain cache system information, and determine an application server, multiple first cache server clusters corresponding to the application server, and multiple second cache server clusters according to the cache system information. The first cache server cluster includes multiple first cache servers located in the same computer room as the application server, and the second cache server cluster includes multiple second cache servers located in different computer rooms from the application server. The first cache servers and the second cache servers are used to receive data sent by the application server and cache it;

[0008] Obtain the business type information of the target business corresponding to data caching by the application server and the business characteristic information of the target business. The business type information is used to indicate the business type of the business, and the business types include live broadcast business, streaming media business, online retail business, game business, and chat business. The business characteristic data is used to indicate the write data rate, read data rate, concurrent thread count, number of errors reported per unit time, and number of restarts per unit time of the business;

[0009] Obtain the pre-trained cache configuration model, and use the business type information and business characteristic information as inputs, and input them into the cache configuration model to obtain the cache configuration information. The cache configuration information is used to indicate the order in which the application server switches to read data from the cache server cluster when the cache server cluster for the application server to read data fails;

[0010] Send the cache configuration information to the application server.

[0011] In one implementation manner of the present disclosure, before obtaining the cache system information, the method further includes:

[0012] Obtain the historical cache log of the application server. The historical cache log includes the business type information of the target business corresponding to data caching by the application server, the business characteristic information of the target business, the user satisfaction information of the target business, and the cache server cluster identifier for the application server to perform data caching;

[0013] Obtain the target cache order according to the cache server cluster identifier and the user satisfaction information, and generate the target cache configuration information according to the target cache order;

[0014] Obtain the initial cache configuration model, use the business type information of the target business and the business characteristic information of the target business as inputs, and use the target cache configuration information as the output, and train the initial cache configuration model to obtain the cache configuration model.

[0015] In one implementation manner of the present disclosure, obtaining the historical cache log of the application server includes:

[0016] Send a historical cache log upload instruction;

[0017] Receive the historical cache log sent by the application server in response to the historical cache log upload instruction.

[0018] In a second aspect, an embodiment of the present disclosure provides a data caching control method, and the method is applied to an application server and includes:

[0019] The acquisition application server collects the service type information of the target service corresponding to data caching, the service feature information of the target service, and the cache server cluster identifier for the application server to perform data caching. The service type information is used to indicate the service type of the service, and the service types include live broadcast service, streaming media service, online retail service, game service, and chat service. The service feature data is used to indicate the write data rate, read data rate, concurrent thread count, number of errors reported per unit time, and number of restarts per unit time of the service;

[0020] Determine the target terminal device that matches the target service, and send a user satisfaction upload instruction;

[0021] Receive the user satisfaction information of the target service sent by the target terminal device in response to the user satisfaction upload instruction;

[0022] Send the service type information of the target service, the service feature information of the target service, the user satisfaction information of the target service, and the cache server cluster identifier for the application server to perform data caching.

[0023] Thirdly, an embodiment of the present disclosure provides a data caching control method, which is applied to a terminal device and includes:

[0024] Receive a user satisfaction upload instruction;

[0025] In response to the user satisfaction upload instruction, obtain the human-computer interaction delay information of the target service indicated by the user satisfaction upload instruction, the service error information of the target service, the service repeated operation information of the target service, and the service customer evaluation information of the target service;

[0026] Use the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, and the customer evaluation information of the target service as inputs, and input them into a pre-trained user satisfaction detection model to obtain the user satisfaction information of the target service;

[0027] Send the user satisfaction information of the target service.

[0028] In an implementation manner of the present disclosure, before receiving the user satisfaction upload instruction, the method further includes:

[0029] Receive the updated weight parameter sent by the edge server, and update the private user satisfaction detection model according to the updated weight parameter;

[0030] Real-time collect the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service;

[0031] Taking the human-machine interaction delay information, error message, repeated operation information, and customer evaluation information collected in real time as inputs, and taking the user satisfaction information collected in real time as outputs, training the updated private user satisfaction detection model;

[0032] When the trained private user satisfaction detection model does not converge, obtaining a gradient update vector according to the trained private user satisfaction detection model, and sending the gradient update vector. The edge server is used to aggregate the gradient update vector and update the weight parameters of the common user satisfaction detection model of the edge server according to the aggregated gradient update vector to obtain updated weight parameters;

[0033] When the trained private user satisfaction detection model converges, obtaining a target user satisfaction detection model according to the trained private user satisfaction detection model.

[0034] In an implementation of the present disclosure, before receiving the updated weight parameters sent by the edge server and updating the private user satisfaction detection model according to the updated weight parameters, the method further includes:

[0035] Receiving a private data upload instruction;

[0036] In response to the private data upload instruction, sending the human-machine interaction delay information of the target service, the error message of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance;

[0037] Receiving the initial weight parameters sent by the edge server;

[0038] Updating the initial user satisfaction detection model according to the initial weight parameters to obtain a private user satisfaction detection model.

[0039] In an implementation of the present disclosure, when sending the human-machine interaction delay information of the target service, the error message of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance, the method further includes:

[0040] Adding noise to the pre-collected human-machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information to obtain the human-machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information with noise added;

[0041] Sending the human-machine interaction delay information of the target service, the error message of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance includes:

[0042] Send the human-computer interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information after adding noise.

[0043] In a fourth aspect, an electronic device is provided in an embodiment of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored on the memory. Wherein, the processor executes the computer program to implement the method of any one of the first aspect or any one of the second aspect.

[0044] In a fifth aspect, a computer-readable storage medium is provided in an embodiment of the present disclosure, on which a computer instruction is stored. Wherein, when the computer instruction is executed by a processor, the method of any one of the first aspect or any one of the second aspect is implemented.

[0045] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0046] In the above technical solution, by obtaining cache system information and determining an application server, a plurality of first cache server clusters corresponding to the application server, and a plurality of second cache server clusters based on the cache system information, that is, the first cache server clusters and the second cache server clusters that can be used by the application server for data caching, where the first cache server clusters include a plurality of first cache servers set in the same computer room as the application server, and the second cache server clusters include a plurality of second cache servers set in different computer rooms from the application server, it is ensured that the application server cannot perform data caching due to the failure of a single computer room; then, obtain the service type information of the target service corresponding to the application server for data caching and the service characteristic information of the target service, where the service type information is used to indicate the service type of the service, and the service types include live broadcast services, streaming media services, online retail services, game services, and chat services, and the service characteristic data is used to indicate the write data rate, read data rate, concurrent thread count, number of errors reported per unit time, and number of restarts per unit time of the service. Both the service type information and the service characteristic information are used to indicate information that can affect the cache efficiency when the application server performs data caching; obtain a pre-trained cache configuration model, and use the service type information and the service characteristic information as inputs, and input them into the cache configuration model to obtain cache configuration information. The cache configuration information is used to indicate the order of the cache server clusters that the application server turns to read data when the cache server cluster for the application server to read data fails. The cache configuration model can be understood as a model that has learned the rule between the service type information, the service characteristic information, and the cache configuration information with higher cache efficiency through pre-training. Therefore, the cache configuration information output by the cache configuration model can ensure higher cache efficiency when the application server runs the target service and performs data caching; by sending the cache configuration information to the application server, the application server performs data caching according to the cache configuration information, so that the order of caching data can be configured for the target service corresponding to the data read, ensuring higher efficiency in reading cached data and improving the user experience.

[0047] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In conjunction with the accompanying drawings, through the following detailed description of non-limiting embodiments, other features, objects, and advantages of the present disclosure will become more apparent. In the drawings:

[0049] Figure 1 Show a schematic structural block diagram of a data caching control system according to an embodiment of the present disclosure;

[0050] Figure 2Schematic flowchart showing a data caching control method according to an embodiment of the present disclosure;

[0051] Figure 3 Schematic flowchart showing a data caching control method according to an embodiment of the present disclosure;

[0052] Figure 4 Schematic flowchart showing a data caching control method according to an embodiment of the present disclosure;

[0053] Figure 5 Schematic block diagram showing an electronic device according to an embodiment of the present disclosure;

[0054] Figure 6 Schematic diagram of a structure of an electronic device suitable for implementing a data caching control method according to an embodiment of the present disclosure. Detailed implementation manners

[0055] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. In addition, for clarity, parts irrelevant to the description of the exemplary embodiments are omitted in the drawings.

[0056] In the present disclosure, it should be understood that terms such as "including" or "having" are intended to indicate the presence of features, numbers, steps, actions, components, parts, or combinations thereof disclosed in this specification, and do not exclude the possibility of the presence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0057] It should be further noted that, without conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0058] Details of the embodiments of the present disclosure will be introduced in detail below through specific examples.

[0059] In recent years, with the development of communication technologies, people's dependence on the network has been increasing in all aspects of daily life. The terminal devices used by people can access the network through network devices closer to themselves in order to use corresponding network applications. With the development of the Internet, the user scale and data scale have become larger and larger, and higher requirements have been put forward for the stability of network applications. Among them, the caching system is generally used to support network applications, and higher requirements are placed on the stability of the caching system.

[0060] In the related art, a distributed cache system with relatively high stability can be equipped for a network application. The distributed cache system generally includes an application server and multiple cache server clusters. When the application server reads cache data, it can sequentially read the cache data from multiple cache server clusters according to the order specified by the pre-set cache configuration information.

[0061] Although in the above solution, the stability of the distributed cache system is good, since the cache configuration information is pre-set, the order of reading the cache data is also pre-set. When the network application executes different types of services, it is impossible to adjust the configuration of the master cluster and the slave cluster, that is, the order of reading the cache data, for different types of services, which reduces the efficiency of reading the cache data and damages the user experience.

[0062] In view of the above defects, in the technical solution provided by the present disclosure, by obtaining cache system information and determining an application server and a plurality of first cache server clusters and a plurality of second cache server clusters corresponding to the application server according to the cache system information, that is, the first cache server clusters and the second cache server clusters that can be used by the application server for data caching, where the first cache server cluster includes a plurality of first cache servers set in the same computer room as the application server, and the second cache server cluster includes a plurality of second cache servers set in different computer rooms from the application server, it is ensured that the application server cannot perform data caching due to the failure of a single computer room; then, obtain the service type information of the target service corresponding to the application server for data caching and the service characteristic information of the target service, where the service type information is used to indicate the service type of the service, and the service types include live broadcast service, streaming media service, online retail service, game service, chat service, and the service characteristic data is used to indicate the write data rate, read data rate, concurrent thread number, error reporting times per unit time, and restart times per unit time of the service. Both the service type information and the service characteristic information are used to indicate the information that can affect the cache efficiency when the application server performs data caching; obtain a pre-trained cache configuration model, and use the service type information and the service characteristic information as inputs, and input them into the cache configuration model to obtain cache configuration information. The cache configuration information is used to indicate the order of the cache server clusters that the application server turns to read data when the cache server cluster for the application server to read data fails. The cache configuration model can be understood as a model that has learned the rules between the service type information, the service characteristic information, and the cache configuration information with higher cache efficiency through pre-training. Therefore, the cache configuration information output by the cache configuration model can ensure higher cache efficiency when the application server runs the target service and performs data caching; by sending the cache configuration information to the application server, the application server performs data caching according to the cache configuration information, so that the order of caching data can be configured for the target service corresponding to the read data caching, ensuring higher efficiency in reading cached data and improving the user experience.

[0063] Figure 1 FIG. 4 shows a schematic structural block diagram of a data caching control system according to an embodiment of the present disclosure. The data caching control system includes a data caching control device 101, an application server 102, a first cache server cluster 103, a second cache server cluster 104, and a network 105. The network 105 is used as a medium for providing communication links between the data caching control device 101 and the application server 102, between the application server 102 and the first cache server cluster 103, and between the application server 102 and the second cache server cluster 104. The network 105 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0064] The data cache control device 101 can be various electronic devices with network access functions, including but not limited to mobile communication terminals, desktop computers, tablet computers, laptop computers, wearable devices, etc.

[0065] The application server 102, the first cache server in the first cache server cluster 103, and the second cache server in the second cache server cluster 104 can be cloud servers or servers provided by a data cache control service provider.

[0066] Figure 2 The schematic flowchart showing a data cache control method according to an embodiment of the present disclosure, which is applied to a data cache control device. As Figure 2 shown, the data cache control method includes the following steps:

[0067] In step S101, obtain cache system information, and determine the application server and multiple first cache server clusters and multiple second cache server clusters corresponding to the application server according to the cache system information.

[0068] Among them, the first cache server cluster includes multiple first cache servers set in the same computer room as the application server, the second cache server cluster includes multiple second cache servers set in different computer rooms from the application server, and the first cache server and the second cache server are used to receive and cache the data sent by the application server.

[0069] In an embodiment of the present disclosure, to obtain the cache system information, it can be by reading the cache system information pre-stored in the data cache control device, or by obtaining the cache system information from other devices or systems. For example, the data cache control device can send a cache system information upload instruction to the application server and receive the cache system information sent by the application server in response to the cache system information upload instruction.

[0070] In step S102, obtain the service type information and service characteristic information of the target service corresponding to the data cache of the application server.

[0071] Among them, the service type information is used to indicate the service type of the service, the service type includes live broadcast service, streaming media service, online retail service, game service, chat service, and the service characteristic data is used to indicate the write data rate, read data rate, concurrent thread number, number of errors reported per unit time, and number of restarts per unit time of the service.

[0072] In an embodiment of the present disclosure, obtaining the service type information of the target service and the service feature information of the target service may be to search in the service information database pre-stored in the data cache control device according to the server identifier of the application server to obtain the service type information of the target service and the service feature information of the target service, or to obtain the cache system information from other devices or systems. For example, the data cache control device may send a service information upload instruction to the application server and receive the service type information of the target service and the service feature information of the target service sent by the application server in response to the service information upload instruction.

[0073] In step S103, obtain a pre-trained cache configuration model, and use the service type information and the service feature information as inputs, and input them into the cache configuration model to obtain cache configuration information.

[0074] Among them, the cache configuration information is used to indicate the order of the cache server clusters that the application server turns to read data when the cache server cluster for the application server to read data fails.

[0075] In an embodiment of the present disclosure, the cache configuration model may be a neural network (NN) model, a convolutional neural network (CNN) model, a long short-term memory (LSTM) model, or the like.

[0076] In an embodiment of the present disclosure, the cache configuration model may be pre-stored in the data cache control device, or may be obtained from other devices or systems.

[0077] In step S104, send the cache configuration information to the application server.

[0078] In this embodiment, by obtaining cache system information and determining an application server, a plurality of first cache server clusters corresponding to the application server, and a plurality of second cache server clusters based on the cache system information, that is, the first cache server clusters and the second cache server clusters that can be used by the application server for data caching, where the first cache server clusters include a plurality of first cache servers set in the same computer room as the application server, and the second cache server clusters include a plurality of second cache servers set in different computer rooms from the application server, it is ensured that the application server cannot perform data caching due to a single computer room failure; then, obtain the service type information of the target service corresponding to the data caching of the application server and the service characteristic information of the target service, where the service type information is used to indicate the service type of the service, and the service types include live broadcast services, streaming media services, online retail services, game services, and chat services, and the service characteristic data is used to indicate the write data rate, read data rate, concurrent thread number, number of error reports per unit time, and number of restarts per unit time of the service. Both the service type information and the service characteristic information are used to indicate the information that can affect the cache efficiency when the application server performs data caching; obtain a pre-trained cache configuration model, and use the service type information and the service characteristic information as inputs, and input them into the cache configuration model to obtain cache configuration information. The cache configuration information is used to indicate the order of the cache server clusters to which the application server switches to read data when the cache server cluster for the application server to read data fails. The cache configuration model can be understood as a model that has learned the rules between the service type information, the service characteristic information, and the cache configuration information with higher cache efficiency through pre-training. Therefore, the cache configuration information output by the cache configuration model can ensure higher cache efficiency when the application server runs the target service and performs data caching; by sending the cache configuration information to the application server, the application server performs data caching according to the cache configuration information, so that the order of caching data can be configured for the target service corresponding to the read data caching, ensuring higher efficiency in reading cached data and improving the user experience.

[0079] In one implementation manner of the present disclosure, in step 101, before obtaining the cache system information, the method further includes the following steps:

[0080] Obtain the historical cache log of the application server, where the historical cache log includes the service type information of the target service corresponding to the data caching of the application server, the service characteristic information of the target service, the user satisfaction information of the target service, and the cache server cluster identifier of the application server for data caching;

[0081] Obtain the target cache order according to the cache server cluster identifier and the user satisfaction information, and generate target cache configuration information according to the target cache order;

[0082] Obtain an initial cache configuration model, use the service type information of the target service and the service feature information of the target service as inputs, and use the target cache configuration information as the output to train the initial cache configuration model to obtain a cache configuration model.

[0083] In an embodiment of the present disclosure, the user satisfaction information may be information fed back by a terminal device that executes the target service. The user satisfaction information is used to indicate the user satisfaction when the application server performs data caching and after data caching, and the user executes the target service through the terminal device. This user satisfaction is positively correlated with the user experience.

[0084] In an embodiment of the present disclosure, obtaining the target cache order according to the cache server cluster identifier and the user satisfaction information may be to sort the cache server clusters indicated by the cache server cluster identifier from high to low according to the user satisfaction indicated by the user satisfaction information to obtain the target cache order.

[0085] In an embodiment of the present disclosure, to obtain the initial cache configuration model, it may be possible to read the initial cache configuration model pre-stored in the data cache control device, or obtain the initial cache configuration model from other devices or systems. The initial cache configuration model may be a neural network model, a convolutional neural network model, or a long short-term memory network model, etc.

[0086] In this embodiment, by obtaining the historical cache log of the application server, obtaining the target cache order according to the cache server cluster identifier and the user satisfaction information, and generating the target cache configuration information according to the target cache order, it can be ensured that when the application server performs data caching according to the target cache order indicated by the target cache configuration information, a higher user satisfaction can be obtained; by obtaining the initial cache configuration model, using the service type information of the target service and the service feature information of the target service as inputs, and using the target cache configuration information as the output to train the initial cache configuration model to obtain a cache configuration model, it can be ensured that the cache configuration model learns the rules between the cache configuration information that can obtain a higher user satisfaction and the service type information and service feature information.

[0087] In an implementation manner of the present disclosure, obtaining the historical cache log of the application server includes:

[0088] Send a historical cache log upload instruction;

[0089] Receive the historical cache log sent by the application server in response to the historical cache log upload instruction.

[0090] In this embodiment, by sending a historical cache log upload instruction and receiving the historical cache log sent by the application server in response to the historical cache log upload instruction, it is relatively convenient to obtain a real and effective historical cache log.

[0091] Figure 3 FIG. shows a schematic flowchart of a data cache control method according to an embodiment of the present disclosure. The data cache control method is applied to an application server. As Figure 3 shown, the data cache control method includes the following steps:

[0092] In step S201, collect the service type information of the target service corresponding to the data cache of the application server, the service characteristic information of the target service, and the cache server cluster identifier for the application server to perform data caching.

[0093] Among them, the service type information is used to indicate the service type of the service. The service types include live broadcast services, streaming media services, online retail services, game services, and chat services. The service characteristic data is used to indicate the write data rate, read data rate, concurrent thread count, number of errors reported per unit time, and number of restarts per unit time of the service. The cache configuration information is used to indicate the order of the cache server clusters to which the application server turns to read data when the cache server cluster for the application server to read data fails.

[0094] In step S202, determine the target terminal device that matches the target service, and send a user satisfaction upload instruction.

[0095] In step S203, receive the user satisfaction information of the target service sent by the target terminal device in response to the user satisfaction upload instruction.

[0096] In step S204, send the service type information of the target service, the service characteristic information of the target service, the user satisfaction information of the target service, and the cache server cluster identifier for the application server to perform data caching.

[0097] In this embodiment, by collecting the service type information of the target service corresponding to the data cache of the application server, the service characteristic information of the target service, and the cache server cluster identifier for the application server to perform data caching, determining the target terminal device that matches the target service, sending a user satisfaction upload instruction, receiving the user satisfaction information of the target service sent by the target terminal device in response to the user satisfaction upload instruction, and sending a historical cache log including the service type information of the target service, the service characteristic information of the target service, the user satisfaction information of the target service, and the cache configuration information for the application server to perform data caching, the data cache control device can obtain the information for training the cache configuration model to obtain the cache configuration model, which is convenient for the data cache control device to train the cache configuration model.

[0098] Figure 4 A schematic flowchart showing a data cache control method according to an embodiment of the present disclosure, the data cache control method being applied to a terminal device. As Figure 4 shown, the data cache control method includes the following steps:

[0099] In step S301, a user satisfaction upload instruction is received.

[0100] In step S302, in response to the user satisfaction upload instruction, obtain the human-computer interaction delay information of the target service indicated by the user satisfaction upload instruction, the service error information of the target service, the service repeated operation information of the target service, the service customer evaluation information of the target service, and the user satisfaction detection model.

[0101] In an embodiment of the present disclosure, the human-computer interaction delay information of the target service can be used to indicate the human-computer interaction delay duration, which is the duration from the moment when the terminal device obtains a human-computer interaction instruction through the human-computer interaction device to the moment when the terminal device receives the human-computer interaction feedback information sent by the application server in response to the human-computer interaction instruction uploaded by the terminal device when the terminal device executes the target service.

[0102] The service error information of the target service can be used to indicate an error that occurs when the terminal device executes the target service.

[0103] The service repeated operation information of the target service can be the number of times of adjacent two identical operations on the target service on the terminal device when the terminal device executes the target service, where the time difference between the adjacent two identical operations is less than or equal to a preset time difference threshold.

[0104] The service customer evaluation information of the target service can be understood as an evaluation of the application corresponding to the target service made through the human-computer interaction device of the terminal device, or can be understood as an evaluation of the function corresponding to the target service made through the human-computer interaction device of the terminal device.

[0105] In step S303, use the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, and the customer evaluation information of the target service as inputs, and input them into a pre-trained user satisfaction detection model to obtain the user satisfaction information of the target service.

[0106] In step S304, send the user satisfaction information of the target service.

[0107] In this embodiment, by receiving a user satisfaction upload instruction, in response to the user satisfaction upload instruction, obtaining the human-computer interaction delay information of the target service indicated by the user satisfaction upload instruction, the service error information of the target service, the service repeated operation information of the target service, and the service customer evaluation information of the target service, and using the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, and the customer evaluation information of the target service as inputs, and inputting them into a pre-trained user satisfaction detection model to obtain the user satisfaction information of the target service, and then sending the user satisfaction information of the target service, it can be ensured that the application server can obtain user satisfaction information that can accurately indicate the user's satisfaction when the user executes the target service through the end user.

[0108] In an implementation manner of the present disclosure, before receiving the user satisfaction upload instruction in step S301, the method further includes the following steps:

[0109] Receiving the updated weight parameters sent by the edge server, and updating the private user satisfaction detection model according to the updated weight parameters.

[0110] Real-time collecting the human-computer interaction delay information of the target service, the service error information of the target service, the service repeated operation information of the target service, the service customer evaluation information of the target service, and the user satisfaction information of the target service; using the real-time collected human-computer interaction delay information, error information, repeated operation information, and customer evaluation information as inputs, and using the real-time collected user satisfaction information as outputs, training the updated private user satisfaction detection model;

[0111] When the trained private user satisfaction detection model does not converge, obtaining a gradient update vector according to the trained private user satisfaction detection model, and sending the gradient update vector. The edge server is used to aggregate the gradient update vector, and update the weight parameters of the common user satisfaction detection model of the edge server according to the aggregated gradient update vector to obtain the updated weight parameters;

[0112] When the trained private user satisfaction detection model converges, obtaining the target user satisfaction detection model according to the trained private user satisfaction detection model.

[0113] In an embodiment of the present disclosure, both the private user satisfaction detection model and the common user satisfaction detection model can be a neural network model, a convolutional neural network model, or a long short-term memory network model, etc.

[0114] In an embodiment of the present disclosure, obtaining a target user satisfaction detection model according to the trained private user satisfaction detection model can be understood as storing the trained private user satisfaction detection model as the target user satisfaction detection model, or can be understood as directly identifying the trained private user satisfaction detection model as the target user satisfaction detection model.

[0115] In this embodiment, the terminal device receives the updated weight parameters sent by the edge server, which are obtained by the edge server aggregating the gradient update vectors sent by multiple terminal devices and updating the weight parameters of the common user satisfaction detection model of the edge server according to the aggregated gradient update vectors. Therefore, the updated common user satisfaction detection model can reflect the common user satisfaction law between the human-computer interaction delay information, error message information, repeated operation information, customer evaluation information, and user satisfaction information of multiple terminal devices collected in real time and learned by the common user satisfaction detection model of the edge server in the previous round of training. Then, by using the human-computer interaction delay information, error message information, repeated operation information, and customer evaluation information collected in real time as inputs and the user satisfaction information collected in real time as outputs to train the updated private user satisfaction detection model, the updated private user satisfaction detection model can, on the basis of learning the common user satisfaction law, also learn personalized information about the human-computer interaction delay information, error message information, repeated operation information, customer evaluation information, and user satisfaction information collected in real time by the terminal device itself, so that the trained private user satisfaction detection model can learn the private user satisfaction law between the human-computer interaction delay information, error message information, repeated operation information, customer evaluation information, and user satisfaction information collected by the terminal device itself. When the trained private user satisfaction detection model does not converge, it means that the trained private user satisfaction detection model still needs to be further trained. By obtaining the gradient update vector according to the trained private user satisfaction detection model and sending the gradient update vector, the edge server can continue to obtain the corresponding updated weight parameters based on the gradient update vectors uploaded by multiple terminal devices without disclosing the user personal data of the terminal device, so as to continue training the private user satisfaction detection models of each terminal device. When the trained private user satisfaction detection model converges, the converged private user satisfaction detection model can obtain relatively accurate user satisfaction detection information based on the input human-computer interaction delay information, error message information, repeated operation information, and customer evaluation information. Therefore, the converged private user satisfaction detection model is used as the target user satisfaction detection model.

[0116] In an implementation of the present disclosure, before receiving the updated weight parameters sent by the edge server and updating the private user satisfaction detection model according to the updated weight parameters, the method further includes:

[0117] Receiving a private data upload instruction;

[0118] In response to the private data upload instruction, sending the pre-collected human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service;

[0119] Receiving the initial weight parameters sent by the edge server;

[0120] Updating the initial user satisfaction detection model according to the initial weight parameters to obtain a private user satisfaction detection model.

[0121] In this implementation, by receiving the private data upload instruction and, in response to the private data upload instruction, sending the pre-collected human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service, the edge server can perform preliminary training on its own initial user satisfaction detection model based on the pre-collected human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service to obtain a common user satisfaction detection model. At this time, the common user satisfaction detection model can be understood as a model that has preliminarily learned the common user satisfaction law between the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service pre-collected by multiple terminal devices. Then, the terminal device updates the initial user satisfaction detection model according to the initial weight parameters obtained by the common user satisfaction detection model to obtain a private user satisfaction detection model. The private user satisfaction detection model can be understood as a model that has learned the law learned by the common user satisfaction detection model, that is, at this time, the private user satisfaction detection model can also be understood as a model that has preliminarily learned the common user satisfaction law of multiple terminal devices, thus facilitating subsequent multi-round training of the private user satisfaction detection model without training based on the initial user satisfaction detection model, reducing the training difficulty.

[0122] In an implementation of the present disclosure, when sending the pre-collected human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service, the method further includes:

[0123] Add noise to the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information to obtain the human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information after adding noise;

[0124] Send the pre - collected human - machine interaction delay information of the target service, error message of the target service, repeated operation information of the target service, customer evaluation information of the target service, and user satisfaction information of the target service, including:

[0125] Send the human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information after adding noise.

[0126] Among them, adding noise to the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information can be adding random noise to the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information, or adding Laplace noise to the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information.

[0127] In this embodiment, by adding noise to the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information to obtain the human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information after adding noise, and sending the human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information after adding noise, it is possible to avoid the leakage of relatively sensitive user personal information in the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information on the premise of not reducing the user satisfaction law of the terminal device reflected by the pre - collected human - machine interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information as much as possible, and improve the security of user personal information.

[0128] The present disclosure also discloses an electronic device, Figure 5 showing a schematic structural block diagram of an electronic device according to an embodiment of the present disclosure, as Figure 5 shown, the electronic device 400 includes a memory 401 and a processor 402; among them,

[0129] The memory 401 is used to store one or more computer instructions, and among them, the one or more computer instructions are executed by the processor 402 to implement any method in the embodiments of the present disclosure.

[0130] Figure 6It is a schematic structural diagram of an electronic device suitable for implementing the data caching control method according to an embodiment of the present disclosure.

[0131] As Figure 6 shown, the electronic device 500 includes a processing unit 501, which can be implemented as a processing unit such as a CPU, GPU, FPGA, NPU, etc. The processing unit 501 can execute various processes in the embodiments of any of the above methods of the present disclosure according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage section 508 into the random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0132] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read from it can be installed into the storage section 508 as needed.

[0133] Specifically, according to an embodiment of the present disclosure, any of the above methods with reference to the embodiments of the present disclosure can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program tangibly embodied on a machine-readable medium, and the computer program includes program code for executing any of the methods in the embodiments of the present disclosure. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 509, and / or installed from the removable medium 511.

[0134] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0135] The units or modules described in the embodiments of the present disclosure can be implemented in software or in hardware. The described units or modules can also be provided in a processor, and the names of these units or modules do not, in some cases, constitute a limitation to the units or modules themselves.

[0136] On the other hand, the present disclosure also provides a computer-readable storage medium, which can be the computer-readable storage medium included in the device described in the above embodiments; or it can be a computer-readable storage medium that exists separately and is not assembled into the device. The computer-readable storage medium stores one or more programs, and the programs are used by one or more processors to execute the methods described in the present disclosure.

[0137] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, technical solutions formed by mutually replacing the above features with (but not limited to) technical features having similar functions disclosed in the present disclosure.

Claims

1. A data caching control method, which is applied to a data caching control device, including: Obtaining cache system information, and determining an application server, a plurality of first cache server clusters corresponding to the application server, and a plurality of second cache server clusters according to the cache system information. The first cache server clusters include a plurality of first cache servers located in the same computer room as the application server, and the second cache server clusters include a plurality of second cache servers located in different computer rooms from the application server. The first cache servers and the second cache servers are used to receive data sent by the application server and cache it; Obtaining the service type information of the target service corresponding to the data caching of the application server and the service characteristic information of the target service. The service type information is used to indicate the service type of the service, and the service types include live broadcast service, streaming media service, online retail service, game service, and chat service. The service characteristic data is used to indicate the write data rate, read data rate, number of concurrent threads, number of error reports per unit time, and number of restarts per unit time of the service; Obtaining a pre-trained cache configuration model, and using the service type information and the service characteristic information as inputs, and inputting them into the cache configuration model to obtain cache configuration information. The cache configuration information is used to indicate the order of the cache server clusters to which the application server turns to read data when the cache server cluster for the application server to read data fails; Sending the cache configuration information to the application server.

2. The data caching control method according to claim 1, before obtaining the cache system information, the method further includes: Obtaining the historical cache log of the application server, where the historical cache log includes the service type information of the target service corresponding to the data caching of the application server, the service characteristic information of the target service, the user satisfaction information of the target service, and the cache server cluster identifier for the application server to perform data caching, where the user satisfaction information is information uploaded by the terminal device executing the target service; Obtaining a target cache order according to the cache server cluster identifier and the user satisfaction information, and generating target cache configuration information according to the target cache order; Obtaining an initial cache configuration model, using the service type information of the target service and the service characteristic information of the target service as inputs, and using the target cache configuration information as outputs, and training the initial cache configuration model to obtain the cache configuration model.

3. The data caching control method according to claim 1, obtaining the historical cache log of the application server, including: Sending a historical cache log upload instruction; Receiving the historical cache log sent by the application server in response to the historical cache log upload instruction.

4. A data caching control method, which is applied to the application server described in claim 2, the method including: Collect the service type information of the target service corresponding to data caching by the application server, the service feature information of the target service, and the cache server cluster identifier for data caching by the application server. The service type information is used to indicate the service type of the service, and the service types include live broadcast service, streaming media service, online retail service, game service, and chat service. The service feature data is used to indicate the write data rate, read data rate, concurrent thread count, number of error reports per unit time, and number of restarts per unit time of the service; Determine the target terminal device that matches the target service, and send a user satisfaction upload instruction; Receive the user satisfaction information of the target service sent by the target terminal device in response to the user satisfaction upload instruction; Send the service type information of the target service, the service feature information of the target service, the user satisfaction information of the target service, and the cache server cluster identifier for data caching by the application server.

5. A data caching control method, which is applied to the terminal device described in claim 2, and the method includes: Receive a user satisfaction upload instruction; In response to the user satisfaction upload instruction, obtain the human-computer interaction delay information of the target service indicated by the user satisfaction upload instruction, the service error information of the target service, the service repeated operation information of the target service, and the service customer evaluation information of the target service; Use the human-computer interaction delay information, error information, repeated operation information, and customer evaluation information of the target service as inputs, and input them into a pre-trained user satisfaction detection model to obtain the user satisfaction information of the target service; Send the user satisfaction information of the target service.

6. According to the data caching control method described in claim 5, before receiving the user satisfaction upload instruction, the method further includes: Receive the updated weight parameter sent by the edge server, and update the private user satisfaction detection model according to the updated weight parameter; Collect in real time the human-computer interaction delay information, error information, repeated operation information, customer evaluation information, and user satisfaction information of the target service; Use the human-computer interaction delay information, error information, repeated operation information, and customer evaluation information collected in real time as inputs, and use the user satisfaction information collected in real time as outputs to train the updated private user satisfaction detection model; When the trained private user satisfaction detection model does not converge, obtain a gradient update vector according to the trained private user satisfaction detection model, and send the gradient update vector. The edge server is used to aggregate the gradient update vector and update the weight parameter of the common user satisfaction detection model of the edge server according to the aggregated gradient update vector to obtain the updated weight parameter; When the trained private user satisfaction detection model converges, obtain the target user satisfaction detection model according to the trained private user satisfaction detection model.

7. The data caching control method according to claim 6, before receiving the updated weight parameter sent by the edge server and updating the private user satisfaction detection model according to the updated weight parameter, the method further includes: Receiving a private data upload instruction; In response to the private data upload instruction, sending the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance; Receiving the initial weight parameter sent by the edge server; Updating the initial user satisfaction detection model according to the initial weight parameter to obtain the private user satisfaction detection model.

8. The data caching control method according to claim 7, when sending the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance, the method further includes: Adding noise to the pre-collected human-computer interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information to obtain the human-computer interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information with noise added; The step of sending the human-computer interaction delay information of the target service, the error information of the target service, the repeated operation information of the target service, the customer evaluation information of the target service, and the user satisfaction information of the target service collected in advance includes: Sending the human-computer interaction delay information, repeated operation information, customer evaluation information, and user satisfaction information with noise added.

9. An electronic device, the electronic device includes a memory, a processor, and a computer program stored on the memory, wherein, The processor executes the computer program to implement the method according to any one of claims 1-8.

10. A computer-readable storage medium, on which computer instructions are stored, wherein, When the computer instructions are executed by the processor, the method according to any one of claims 1-8 is implemented.

Images