Service resource deployment method, device, equipment, medium and product

Abstract

The present application provides a service resource deployment method, apparatus, equipment, medium and product, the method comprising: obtaining a service request sequence corresponding to the service resources of an edge server within a preset time period; the service request sequence is generated based on the first service resource request quantity and service resource request category corresponding to each moment within the preset time period; determining the service heat corresponding to each service resource request category based on the second service resource request quantity corresponding to each service resource request category in the service request sequence; inputting the service request sequence into a preset service deployment model, and using the service heat as a constraint condition to generate a target service resource deployment strategy; deploying service resources to the edge server according to the target service resource deployment strategy. The service resource deployment method of the present application can reduce the feedback time of high-heat services and improve user experience.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a service resource deployment method, apparatus, equipment, medium, and product. Background Art

[0002] Edge computing is suitable for processing latency-sensitive services. Since the request transmission distance is reduced, the request latency is also greatly reduced. At the same time, service requests are diverted, effectively reducing network pressure and avoiding network congestion. However, compared with cloud servers, edge servers have limited capacity and bandwidth, so all types of service resources cannot be deployed on each edge server.

[0003] Therefore, the service resources deployed on each edge server are not exactly the same. Currently, due to the different user demands for different types of service resources, high-demand service resources are more demanding, while low-demand service resources are less demanding. This results in long feedback times for high-demand services and a poor user experience. Summary of the Invention

[0004] The present application provides a service resource deployment method, device, equipment, medium and product to solve the current problem that users have different demands for different types of service resources, with high demand for high-popularity service resources and low demand for low-popularity service resources, resulting in long feedback time for high-popularity services and poor user experience.

[0005] A first aspect of the present application provides a service resource deployment method, comprising:

[0006] Obtaining a service request sequence corresponding to a service resource of the edge server within a preset time period; the service request sequence is generated according to the number of first service resource requests and the service resource request category corresponding to each moment within the preset time period;

[0007] Determining the service heat corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence;

[0008] Inputting the service request sequence into a preset service deployment model, and using the service popularity as a constraint condition to generate a target service resource deployment strategy;

[0009] Deploy the service resources to the edge server according to the target service resource deployment strategy.

[0010] Furthermore, in the above method, obtaining the service request sequence corresponding to the service resources of the edge server within a preset time period includes:

[0011] Obtain the number of service resource requests and service resource request time corresponding to each service resource request category of the edge server within a preset time period;

[0012] The first service resource request quantities corresponding to each moment are arranged in chronological order of the service resource request times to generate the service request sequence; the first service resource request quantities include the service resource request quantities corresponding to each service resource request category at the same moment.

[0013] Furthermore, in the above method, determining the service popularity corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence includes:

[0014] Accumulating the number of each second service resource request according to the service resource request category to generate a corresponding accumulated request number series;

[0015] Using grey prediction and fitting functions to process the accumulated request series to generate a service heat weight factor corresponding to each service resource category;

[0016] The Hacker News algorithm is used to calculate the service popularity weight factor of each service to determine the service popularity corresponding to each service resource category.

[0017] Furthermore, in the above method, the grey prediction and fitting function are used to process the service request series to generate the service heat weight factor corresponding to each service resource category, including:

[0018] Using grey prediction to perform prediction processing on the accumulated request sequence to generate a predicted request sequence; the predicted request sequence includes the predicted number of service resource requests and the accumulated request sequence;

[0019] Using the fitting function to perform curve fitting on the prediction request series to generate the corresponding prediction request curve;

[0020] The service heat weight factor corresponding to each service resource category is determined according to the predicted request curve.

[0021] Furthermore, in the above method, the step of inputting the service request sequence into a preset service deployment model and using the service popularity as a constraint condition to generate a target service resource deployment strategy includes:

[0022] Inputting the service request sequence into a preset service deployment model;

[0023] Using a harmony search algorithm to determine the optimal solution of the service resource deployment strategy corresponding to the preset service deployment model according to the constraints;

[0024] The optimal solution of the service resource deployment strategy is determined as the target service resource deployment strategy.

[0025] Furthermore, in the above method, there are multiple edge servers; before inputting the service request sequence into the preset service deployment model, the method further includes:

[0026] Obtain the location information of each edge server;

[0027] Determine the distance between each edge server based on the location information using a haversine calculation algorithm;

[0028] The preset service deployment model is constructed according to each of the interval distances and the service request sequence.

[0029] A second aspect of the present application provides a service resource deployment device, including:

[0030] An acquisition module, configured to acquire a service request sequence corresponding to a service resource of an edge server within a preset time period; the service request sequence is generated based on the number of first service resource requests and the type of service resource requests corresponding to each moment within the preset time period;

[0031] a determining module, configured to determine the service heat corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence;

[0032] A generation module, configured to input the service request sequence into a preset service deployment model and generate a target service resource deployment strategy using the service popularity as a constraint condition;

[0033] A deployment module is used to deploy service resources to the edge server according to the target service resource deployment strategy.

[0034] Furthermore, in the above device, the acquisition module is specifically configured to:

[0035] Obtain the number of service resource requests and the service resource request time corresponding to each service resource request category of the edge server within a preset time period; arrange the first number of service resource requests corresponding to each moment in the order of the service resource request time to generate the service request sequence; the first number of service resource requests includes the number of service resource requests corresponding to each service resource request category at the same moment.

[0036] Furthermore, in the above-mentioned device, the determining module includes:

[0037] The number of each second service resource request is accumulated in the same category according to the service resource request category to generate a corresponding accumulated request series; the accumulated request series is processed using gray prediction and fitting functions to generate a service heat weight factor corresponding to each service resource category; and the Hacker News algorithm is used to calculate each of the service heat weight factors to determine the service heat corresponding to each service resource category.

[0038] Furthermore, in the above-mentioned device, when the determination module processes the service request series using grey prediction and fitting function to generate the service heat weight factor corresponding to each service resource category, it is specifically used to:

[0039] Gray prediction is used to predict the accumulated request series to generate a predicted request series; the predicted request series includes the predicted number of service resource requests and the accumulated request series; a fitting function is used to perform curve fitting on the predicted request series to generate a corresponding predicted request curve; and a service heat weight factor corresponding to each service resource category is determined based on the predicted request curve.

[0040] Furthermore, in the above device, the generating module is specifically configured to:

[0041] The service request sequence is input into a preset service deployment model; a harmony search algorithm is used to determine an optimal solution of a service resource deployment strategy corresponding to the preset service deployment model according to the constraint conditions; and the optimal solution of the service resource deployment strategy is determined as a target service resource deployment strategy.

[0042] Furthermore, in the above-mentioned device, there are multiple edge servers; the device further includes:

[0043] A construction module is used to obtain the location information of each edge server; use a half-vector calculation algorithm to determine the interval distance between each edge server based on each of the location information; and construct the preset service deployment model based on each of the interval distances and the service request sequence.

[0044] A third aspect of the present application provides an electronic device, comprising: a memory and a processor;

[0045] The memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory to implement the service resource deployment method as described in any one of the first aspects.

[0047] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the service resource deployment method described in any one of the first aspects.

[0048] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the service resource deployment method described in any one of the first aspects.

[0049] The present application provides a service resource deployment method, apparatus, device, medium, and product, the method comprising: obtaining a service request sequence corresponding to a service resource of an edge server within a preset time period; the service request sequence is generated based on a first service resource request quantity and a service resource request category corresponding to each moment within the preset time period; determining the service heat corresponding to each service resource request category based on a second service resource request quantity corresponding to each service resource request category in the service request sequence; inputting the service request sequence into a preset service deployment model, and using the service heat as a constraint condition to generate a target service resource deployment strategy; and deploying service resources to the edge server according to the target service resource deployment strategy. The service resource deployment method of the present application obtains a service request sequence corresponding to a service resource of an edge server within a preset time period, and determines the service heat corresponding to each service resource request category based on a second service resource request quantity corresponding to each service resource request category in the service request sequence. By using a preset service deployment model and using service heat as a constraint condition, the target service resource deployment strategy corresponding to the edge server is determined. Compared with existing service resource deployment methods, the feedback time of high-heat services can be reduced, thereby improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A scenario diagram for implementing the service resource deployment method of an embodiment of the present application;

[0052] Figure 2 Schematic diagram of the service resource deployment method provided for this application Figure 1 ;

[0053] Figure 3 Schematic diagram of the service resource deployment method provided for this application Figure 2 ;

[0054] Figure 4 Schematic diagram of the structure of the service resource deployment device provided by this application Figure 1 ;

[0055] Figure 5 Schematic diagram of the structure of the service resource deployment device provided by this application Figure 2 ;

[0056] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application.

[0057] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0058] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0059] In the technical solutions of the embodiments of this application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0060] It should be noted that the service resource deployment method, apparatus, device, medium, and product disclosed herein can be used in the field of data processing technology. They can also be used in any field other than data processing, such as finance. The application fields of the service resource deployment method, apparatus, device, medium, and product disclosed herein are not limited.

[0061] The technical solution of the present application is described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0062] In order to clearly understand the technical solution of the present application, the solution of the prior art is first introduced in detail. Edge computing is a distributed open platform that integrates network, computing, storage, and application core capabilities on the network side close to the source of objects or data. The edge is considered to be the area between the cloud and the terminal. In an edge computing environment, the edge server is close to the end user, making it suitable for executing delay-sensitive services. However, the capacity, bandwidth, and coverage of edge servers are much lower than those of cloud servers, so all types of services cannot be allocated to each edge server. Therefore, it is crucial to reasonably allocate service resources to meet the needs of users and service providers.

[0063] When allocating service resources in an edge computing environment, service request latency reflects the quality of the service provided by edge servers to end users. The shorter the request latency, the better the user experience. Latency is even more important for latency-sensitive services, making end-user request latency a key consideration. Furthermore, service popularity is a crucial factor that cannot be ignored. Due to the limited capacity and bandwidth of edge servers, ignoring popularity can result in a higher number of low-interest services deployed than high-interest services within a limited space. This forces high-interest service requests to queue for processing, while low-interest service resources remain unused, resulting in low service resource utilization. This in turn leads to prolonged response times for high-interest services and a poor user experience.

[0064] To address the existing issues of long feedback times and poor user experience for high-demand services, the inventors discovered that a comprehensive consideration of the service popularity of service resources can be used to address this issue. This can be achieved by combining actual service resource request data within a preset time period with a preset service deployment model to determine a more optimal service resource deployment strategy. Using this service resource deployment strategy for resource deployment can reduce feedback times for high-demand services and improve user experience.

[0065] Specifically, obtain the service request sequence corresponding to the service resources of the edge server within a preset time period. The service request sequence is generated based on the number of first service resource requests and the service resource request category corresponding to each moment within the preset time period. Determine the service heat corresponding to each service resource request category based on the number of second service resource requests corresponding to each service resource request category in the service request sequence. Input the service request sequence into the preset service deployment model, and use the service heat as a constraint condition to generate a target service resource deployment strategy. Deploy service resources to the edge server according to the target service resource deployment strategy. Compared with the existing service resource deployment method, it can reduce the feedback time of high-heat services and improve the user experience.

[0066] Based on the above creative findings, the inventor proposed the technical solution of this application.

[0067] The following describes the application scenarios of the service resource deployment method provided in the embodiment of the present application. Figure 1 As shown, 1 is an electronic device, 2 is an edge server, and 3 is a user terminal. The network architecture for the application scenario corresponding to the service resource deployment method provided in the embodiment of the present application includes: electronic device 1, edge server 2, and user terminal 3. Edge server 2 can be one or more. Edge server 2 is deployed close to the end user side, storing service resources and processing end user service requests.

[0068] For example, in an edge computing scenario, when user terminal 3 requests a service resource, it sends a service resource request to edge server 2. If edge server 2 is a multi-edge server, the service resource request is sent to one of the edge servers, which then forwards the service resource request to the edge server that stores the corresponding service resource. After receiving the service resource request, edge server 2 responds with the service resource requested by user terminal 3 based on the service resource request.

[0069] When service resource deployment is required, electronic device 1 obtains the number of service resource requests and service resource request categories within a preset time period from edge server 2, and generates a service request sequence corresponding to the service resources of the edge server based on the number of service resource requests and the service resource request categories. At the same time, the service popularity corresponding to each service resource request category is determined based on the number of service resource requests corresponding to each service resource request category in the service request sequence. Electronic device 1 inputs the service request sequence into a preset service deployment model and, using service popularity as a constraint, generates a target service resource deployment strategy. Finally, electronic device 1 deploys the service resources to edge server 2 according to the target service resource deployment strategy.

[0070] In other application scenarios, other electronic devices may store service resource request-related data between the edge server 2 and the user terminal 3, and the electronic device 1 may directly obtain the data from the other electronic devices and perform the subsequent target service resource deployment strategy generation process, thereby further improving the efficiency of target service resource deployment strategy generation. The service resource deployment method of this embodiment may also be applicable to other application scenarios, which are not limited by this embodiment.

[0071] The embodiments of the present application are introduced below with reference to the accompanying drawings.

[0072] Figure 2 Schematic diagram of the service resource deployment method provided for this application Figure 1 ,like Figure 2 As shown, in this embodiment, the execution subject of the embodiment of the present application is a service resource deployment device, which can be integrated into an electronic device. The service resource deployment method provided in this embodiment includes the following steps:

[0073] Step S101: Obtain a service request sequence corresponding to the service resources of the edge server within a preset time period. The service request sequence is generated according to the number of first service resource requests and the service resource request category corresponding to each moment within the preset time period.

[0074] In this embodiment, the preset time period can be within one hour, ten minutes, or one minute before the current moment, etc., and can be set according to actual needs. The first number of service resource requests refers to the total number of service resource requests received at each moment.

[0075] At each moment within a preset time period, the number of second service resource requests corresponding to each service resource request category may be greater than 1. Exemplarily, assuming that the service resource request categories include category A, category B, and category C, and at the same moment, terminal A sends a service resource request corresponding to a service resource of category A, terminal B sends a service resource request corresponding to a service resource of category A, and terminal C sends a service resource request corresponding to a service resource of category B. Then, at that moment, the number of service resource requests corresponding to category A is 2, the number of service resource requests corresponding to category B is 1, and the number of service resource requests corresponding to category C is 0.

[0076] In this embodiment, the service request sequence may be a data queue of service resource request categories and corresponding second service resource request quantities at each moment, arranged in chronological order.

[0077] Step S102 : determining the service heat corresponding to each service resource request category according to the second service resource request quantity corresponding to each service resource request category in the service request sequence.

[0078] In this embodiment, the service heat is related to the number of service resource requests of the service resource request category. For example, if the number of second service resource requests corresponding to category A of the service resource request category is the largest within a preset time period, then category A has the highest service heat.

[0079] Step S103: input the service request sequence into a preset service deployment model, and use the service popularity as a constraint condition to generate a target service resource deployment strategy.

[0080] In this embodiment, the preset service deployment model simulates actual edge server feedback of service resources, and can simulate the service resource feedback duration and service resource feedback process. This model can be constructed based on actual edge server information, the number of first service resource requests corresponding to each moment in a preset time period, and the service resource request category. The target service resource deployment strategy refers to the strategy for deploying service resources to edge servers.

[0081] Step S104: deploy service resources to edge servers according to the target service resource deployment strategy.

[0082] Due to the limited capacity and bandwidth of edge servers, only certain categories of service resources can be deployed. Deploying service resources to edge servers through targeted service resource deployment strategies can reduce the feedback time of high-frequency services and the overall feedback time of service resources on edge servers, thereby improving user experience.

[0083] An embodiment of the present application provides a service resource deployment method, which includes: obtaining a service request sequence corresponding to the service resources of an edge server within a preset time period. The service request sequence is generated based on the first service resource request quantity and service resource request category corresponding to each moment within the preset time period. The service heat corresponding to each service resource request category is determined based on the second service resource request quantity corresponding to each service resource request category in the service request sequence. The service request sequence is input into a preset service deployment model, and the target service resource deployment strategy is generated using the service heat as a constraint condition. The service resources are deployed to the edge server according to the target service resource deployment strategy.

[0084] The service resource deployment method of this application obtains a service request sequence corresponding to the service resources of an edge server within a preset time period and determines the service popularity corresponding to each service resource request category based on the number of second service resource requests corresponding to each service resource request category in the service request sequence. By using a preset service deployment model and using service popularity as a constraint, a target service resource deployment strategy corresponding to the edge server is determined. Compared to existing service resource deployment methods, this can reduce the feedback time for high-popularity services and improve the user experience.

[0085] Figure 3 Schematic diagram of the service resource deployment method provided for this application Figure 2 ,like Figure 3 As shown, the service resource deployment method provided in this embodiment is a further refinement of the service resource deployment method provided in the previous embodiment of this application. The service resource deployment method provided in this embodiment includes the following steps.

[0086] Step S201: Obtain the number of service resource requests and the service resource request time corresponding to each service resource request category of the edge server within a preset time period.

[0087] In this embodiment, when the user terminal sends a service resource request to the edge server, the edge server records the requested service resource request category and service resource request time. When actually acquiring, data preprocessing can be performed in advance to remove unnecessary data.

[0088] Step S202: Arrange the first service resource request quantities corresponding to each moment in chronological order of the service resource request times to generate a service request sequence. The first service resource request quantities include the service resource request quantities corresponding to each service resource request category at the same moment.

[0089] In this embodiment, the first number of service resource requests is the number of service resource requests corresponding to each moment. For example, at a certain moment, there are two requests for service resource request category A and two requests for service resource request category B. The corresponding number of service resource requests at that moment is 4. That is, the first number of service resource requests includes the number of service resource requests corresponding to each service resource request category at the same moment. Using the service request sequence, the second number of service resource requests corresponding to each service resource request category can be more conveniently determined, improving the efficiency of subsequent processes.

[0090] Step S203 : Accumulate the number of second service resource requests according to the service resource request category to generate a corresponding accumulated request number sequence.

[0091] In this embodiment, the second number of service resource requests refers to the number of service resource requests corresponding to each service resource request category in the service request sequence. This number can generally be the number of service resource requests corresponding to a certain category at a certain moment, or the number of service resource requests corresponding to a certain category at multiple moments. For example, if service resource request category A has three service resource requests at a certain moment, the second number of service resource requests can be three.

[0092] The accumulated request number sequence is to accumulate the second service resource request numbers corresponding to each service resource request category to obtain the total number of service resource requests of each service resource request category within a preset time period.

[0093] Step S204 : Processing the accumulated request series using grey prediction and fitting functions to generate a service heat weight factor corresponding to each service resource category.

[0094] Gray prediction refers to a method for forecasting systems with uncertainties, predicting future trends. Fitting functions can be used to perform curve fitting on discrete values. The service popularity weighting factor is the weighting factor for each category in the service popularity area.

[0095] Optionally, in this embodiment, step S204 may specifically be:

[0096] Grey prediction is used to predict the cumulative request series to generate a predicted request series. The predicted request series includes the predicted number of service resource requests and the cumulative request series.

[0097] The fitting function is used to perform curve fitting on the prediction request series to generate the corresponding prediction request curve.

[0098] Determine the service heat weight factor corresponding to each service resource category based on the predicted request curve.

[0099] The gray prediction method can be used to predict the development of the accumulated request series. The fitting function can further determine the development law of the predicted request series, so that the service popularity weight factor of each service resource category can be determined more accurately through partial data, thereby improving the accuracy of the subsequent determination of the service popularity of each service resource category.

[0100] In step S205 , the Hacker News algorithm is used to calculate the service popularity weight factor of each service to determine the service popularity corresponding to each service resource category.

[0101] In this embodiment, the Hacker News algorithm is a heat calculation algorithm for ranking heat. By calculating the heat weight factor of each service through the Hacker News algorithm, the heat of the service can be determined with higher accuracy.

[0102] Step S206: input the service request sequence into the preset service deployment model.

[0103] Optionally, in this embodiment, there are multiple edge servers. Before step S206, the process of building a preset service deployment model is also included, which is as follows:

[0104] Get the location information of each edge server.

[0105] The half-segment calculation algorithm is used to determine the interval distance between each edge server according to the location information.

[0106] Build a preset service deployment model based on each interval distance and service request series.

[0107] In this embodiment, the common edge server sample dataset includes edge server location information, such as the edge server's geographic location, longitude, and latitude. However, it lacks the distance between edge servers, which is closely related to request latency. A haversine calculation algorithm can be used to calculate more accurate distances, thereby improving the accuracy of the preset service deployment model. The preset service deployment model can simulate the transmission of service requests between edge servers in actual edge computing, simulating the process by which service requests arrive at and are processed by edge servers.

[0108] Step S207: using a harmony search algorithm to determine the optimal solution of the service resource deployment strategy corresponding to the preset service deployment model according to the constraint conditions.

[0109] The harmony search algorithm is an algorithm that finds the most suitable strategy among many strategies by imitating musicians to continuously improve the pitch of each instrument. It can quickly iterate and search across all service deployment strategies. The correspondence between the harmony search terminology and service deployment is shown below:

[0110] Instruments: Represent the deployment of various services on each edge server.

[0111] Tone: whether the service is deployed or not (tone 0: service not deployed, tone 1: service deployed).

[0112] Harmony: A strategy for service resource deployment.

[0113] Harmony Memory: Represents a collection of partial service deployment strategies.

[0114] The process of finding the optimal solution for service resource deployment strategy using the harmony search algorithm is as follows:

[0115] After the service request sequence is input into the preset service deployment model, a random number is first generated and compared with the probability of the value in the harmony memory. If the probability is less than the memory value probability, a new harmony is generated in the harmony memory and fine-tuned according to the pitch fine-tuning probability and pitch fine-tuning bandwidth. Otherwise, a new harmony is randomly generated in the solution space. At the same time, the new harmony is adjusted by adjusting the bandwidth and pitch of the new harmony. The newly generated harmony is compared with the worst-performing harmony in the harmony memory. If the newly generated harmony is better than the worst harmony, the harmony memory is updated and the new harmony replaces the worst harmony in the harmony memory. Otherwise, the harmony memory is not updated. A maximum number of iterations is set. If the number of iterations is reached, the iteration stops and the optimal solution for the service resource deployment strategy is output. Otherwise, the iteration continues.

[0116] The speed at which the harmonious search iterates the optimal solution to the service resource deployment strategy is much faster than the optimal solution to the service resource deployment strategy enumerated by the enumeration device in the existing technology, which greatly saves the time of service deployment. Compared with the heuristic service deployment device based on clustering enhancement, the optimal solution to the service resource deployment strategy iterated by the harmonious search avoids falling into local optimality, the strategy has higher accuracy, and the waiting delay of end users is less.

[0117] Step S208: determining the optimal solution of the service resource deployment strategy as the target service resource deployment strategy.

[0118] Step S209: deploy service resources to the edge server according to the target service resource deployment strategy.

[0119] In this embodiment, the implementation of step S209 is similar to that of step S104 in the previous embodiment, and will not be described in detail here.

[0120] In order to further illustrate the service resource deployment method of this embodiment in detail, this implementation method is further illustrated below in the form of a device.

[0121] Based on the above method, the embodiment of the present application forms a service resource deployment device in an edge computing environment, which supports the deployment of latency-sensitive services in the edge computing environment and reduces the service feedback delay by selecting an appropriate deployment strategy. The entire device is divided into a sample preprocessing unit, a service heat calculation unit, and a service deployment unit. The device mainly includes several main components: a service data processor, a server data processor, a service heat factor calculator, a service heat calculator, a service deployment simulator, and a harmony searcher.

[0122] The service data processor extracts and processes data from edge servers to obtain the number and duration of service resource requests for each category within a preset time period. The server data processor calculates the distance between edge servers. The service popularity factor calculator calculates the service popularity factor. The service popularity calculator calculates the service popularity. The service deployment simulator models the service deployment problem in an edge computing environment. The harmony searcher iterates across all deployment strategies to approximate the optimal service deployment strategy.

[0123] The sample preprocessing unit includes: a service data processor and a server data processor.

[0124] Since actual data sets may have problems such as missing and omitted data, it is necessary to pre-process the sample data first and filter out the data information required by this embodiment.

[0125] Service data preprocessing: Process the data to obtain the service resource request time and service resource request quantity corresponding to each service resource request category within a preset time period. Server data preprocessing: Calculate the distance between edge servers based on the half-vector.

[0126] The service heat calculation unit includes: service heat factor calculator, service heat calculator

[0127] The processing flow of the service heat calculation unit is as follows:

[0128] First, input a service request sequence generated by the number of service resource requests corresponding to each service resource request category and the time of the service resource requests. The service request accumulator in the service heat factor calculator adds the number of service resource requests at each moment to the cumulative number of requests from the previous moment to obtain the accumulated request sequence. The gray predictor in the service heat factor calculator then uses gray prediction and a fitting function to determine the service heat weight factor. Finally, the service heat weight factor is input as a parameter into the service heat calculator, and the service heat is finally calculated using the improved Hacker News algorithm.

[0129] The service deployment unit includes a service deployment simulator and a harmony searcher. The service deployment simulator simulates the transmission of service resource requests between edge servers in actual edge computing and builds a service deployment model. The harmony searcher determines the optimal service deployment strategy using a harmony search algorithm. The details are similar to those in the previous embodiment and will not be repeated here.

[0130] Figure 4 Schematic diagram of the structure of the service resource deployment device provided by this application Figure 1 ,like Figure 4 As shown, in this embodiment, the service resource deployment device 300 can be set in an electronic device, and the service resource deployment device 300 includes:

[0131] The acquisition module 301 is configured to acquire a service request sequence corresponding to the service resources of the edge server within a preset time period. The service request sequence is generated based on the number of first service resource requests and the service resource request category corresponding to each moment within the preset time period.

[0132] The determination module 302 is configured to determine the service heat corresponding to each service resource request category according to the second service resource request quantity corresponding to each service resource request category in the service request sequence.

[0133] The generation module 303 is used to input the service request sequence into the preset service deployment model and generate a target service resource deployment strategy using the service popularity as a constraint condition.

[0134] The deployment module 304 is configured to deploy service resources to edge servers according to a target service resource deployment strategy.

[0135] The service resource deployment device provided in this embodiment can execute Figure 2 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The method embodiments shown are similar and will not be described in detail here.

[0136] Figure 5 Schematic diagram of the structure of the service resource deployment device provided by this application Figure 2 ,like Figure 5 As shown, the service resource deployment device provided in this application is further refined based on the service resource deployment device provided in the previous embodiment. For the convenience of description, the service resource deployment device of this embodiment is described in the form of a service resource deployment device 400, and the service resource deployment device 400 includes:

[0137] Optionally, in this embodiment, the acquisition module 301 is specifically configured to:

[0138] Obtain the number of service resource requests and the service resource request time corresponding to each service resource request category on the edge server within a preset time period. Arrange the first number of service resource requests corresponding to each moment in chronological order of the service resource request time to generate a service request sequence. The first number of service resource requests includes the number of service resource requests corresponding to each service resource request category at the same moment.

[0139] Optionally, in this embodiment, the determining module 302 includes:

[0140] The number of second service resource requests is accumulated by category to generate a corresponding cumulative request series. This cumulative request series is processed using gray prediction and fitting functions to generate a service popularity weight factor for each service resource category. The Hacker News algorithm is used to calculate each service popularity weight factor to determine the service popularity of each service resource category.

[0141] Optionally, in this embodiment, when the determination module 302 processes the service request series using the grey prediction and fitting function to generate the service heat weight factor corresponding to each service resource category, it is specifically configured to:

[0142] Gray prediction is used to predict the cumulative request series to generate a predicted request series. The predicted request series includes the predicted number of service resource requests and the cumulative request series. A fitting function is used to perform curve fitting on the predicted request series to generate the corresponding predicted request curve. Based on the predicted request curve, the service popularity weight factor corresponding to each service resource category is determined.

[0143] Optionally, in this embodiment, the generating module 303 is specifically configured to:

[0144] Input the service request sequence into the preset service deployment model. Use the harmony search algorithm to determine the optimal solution for the service resource deployment strategy corresponding to the preset service deployment model based on the constraints. This optimal solution is then determined as the target service resource deployment strategy.

[0145] Optionally, in this embodiment, there are multiple edge servers. The service resource deployment device 400 further includes:

[0146] The construction module 401 is configured to obtain location information of each edge server, determine the distance between each edge server based on the location information using a haversine calculation algorithm, and construct a preset service deployment model based on the distances and the service request sequence.

[0147] The service resource deployment device provided in this embodiment can execute Figure 2-Figure 3 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2-Figure 3The method embodiments shown are similar and will not be described in detail here.

[0148] According to an embodiment of the present application, the present application also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0149] like Figure 6 As shown, Figure 6 is a schematic diagram of the structure of the electronic device provided by the present application. The electronic device is intended to be various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application as described and / or claimed herein.

[0150] like Figure 6 As shown, the electronic device includes: a processor 501 and a memory 502. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the electronic device.

[0151] Memory 502 is the non-transitory computer-readable storage medium provided in this application. The memory stores instructions executable by at least one processor to cause the at least one processor to perform the service resource deployment method provided in this application. The non-transitory computer-readable storage medium of this application stores computer instructions for causing a computer to perform the service resource deployment method provided in this application.

[0152] The memory 502 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as the program instructions / modules corresponding to the service resource deployment method in the embodiment of the present application (for example, the attached Figure 4 The processor 501 executes the non-transient software programs, instructions, and modules stored in the memory 502 to execute various functional applications and data processing of the electronic device, thereby implementing the service resource deployment method in the above method embodiment.

[0153] At the same time, this embodiment also provides a computer product. When instructions in the computer product are executed by a processor of an electronic device, the electronic device can execute the service resource deployment method of the above embodiment.

[0154] Those skilled in the art will readily conceive of other implementations of the embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the embodiments of the present application, which follow the general principles of the embodiments of the present application and include common knowledge or customary technical means in the art not disclosed in the embodiments of the present application.

[0155] It should be understood that the embodiments of the present application are not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the embodiments of the present application is limited only by the appended claims.

Claims

1. A service resource deployment method, characterized in that: include: Obtain the service request sequence corresponding to the service resources of the edge server within a preset time period; The service request sequence is generated according to the number of first service resource requests and the category of service resource requests corresponding to each moment in a preset time period; Determining the service heat corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence; Inputting the service request sequence into a preset service deployment model, and using the service popularity as a constraint condition to generate a target service resource deployment strategy; Deploy service resources to edge servers according to the target service resource deployment strategy; The determining the service heat corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence includes: Accumulating the number of each second service resource request according to the service resource request category to generate a corresponding accumulated request number series; Using grey prediction and fitting functions to process the accumulated request series to generate a service heat weight factor corresponding to each service resource category; The Hacker News algorithm is used to calculate the service popularity weight factor of each service to determine the service popularity corresponding to each service resource category.

2. The method according to claim 1, characterized in that The obtaining of a service request sequence corresponding to the service resources of the edge server within a preset time period includes: Obtain the number of service resource requests and service resource request time corresponding to each service resource request category of the edge server within a preset time period; The first service resource request quantities corresponding to each moment are arranged in chronological order of the service resource request times to generate the service request sequence; the first service resource request quantities include the service resource request quantities corresponding to each service resource request category at the same moment.

3. The method according to claim 1, characterized in that The grey prediction and fitting function are used to process the service request sequence to generate the service heat weight factor corresponding to each service resource category, including: Using grey prediction to perform prediction processing on the accumulated request sequence to generate a predicted request sequence; the predicted request sequence includes the predicted number of service resource requests and the accumulated request sequence; Using the fitting function to perform curve fitting on the prediction request series to generate the corresponding prediction request curve; The service heat weight factor corresponding to each service resource category is determined according to the predicted request curve.

4. The method according to claim 3, characterized in that The step of inputting the service request sequence into a preset service deployment model and using the service popularity as a constraint condition to generate a target service resource deployment strategy includes: Inputting the service request sequence into a preset service deployment model; Using a harmony search algorithm to determine the optimal solution of the service resource deployment strategy corresponding to the preset service deployment model according to the constraints; The optimal solution of the service resource deployment strategy is determined as the target service resource deployment strategy.

5. The method according to any one of claims 1 to 4, characterized in that There are multiple edge servers; and before inputting the service request sequence into a preset service deployment model, the method further includes: Obtain the location information of each edge server; Determine the distance between each edge server based on the location information using a haversine calculation algorithm; The preset service deployment model is constructed according to each of the interval distances and the service request sequence.

6. A service resource deployment device, characterized in that: include: An acquisition module is used to obtain a service request sequence corresponding to the service resources of the edge server within a preset time period; The service request sequence is generated according to the number of first service resource requests and the category of service resource requests corresponding to each moment in a preset time period; a determining module, configured to determine the service heat corresponding to each service resource request category according to the number of second service resource requests corresponding to each service resource request category in the service request sequence; A generation module, configured to input the service request sequence into a preset service deployment model and generate a target service resource deployment strategy using the service popularity as a constraint condition; A deployment module, configured to deploy service resources to edge servers according to the target service resource deployment strategy; The determination module is specifically used to accumulate the number of each second service resource request in the same category according to the service resource request category to generate a corresponding accumulated request series; use gray prediction and fitting functions to process the accumulated request series to generate a service heat weight factor corresponding to each service resource category; use the Hacker News algorithm to calculate each of the service heat weight factors to determine the service heat corresponding to each service resource category.

7. An electronic device, characterized in that: include: memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the service resource deployment method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the service resource deployment method according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the service resource deployment method according to any one of claims 1 to 5 is implemented.

Images