Resource elastic expansion and contraction method, device, equipment, medium and product

Abstract

The invention provides a resource elastic scaling method and device, equipment, a medium and a product, and relates to the technical field of resource management. The method comprises the following steps: obtaining a first adjustment strategy according to one adjustment strategy in a plurality of preset adjustment strategies and a preset first user demand; collecting first data in a past first preset time period; according to the first data, predicting to obtain second data in a second preset time period in the future; generating a resource adjustment scheme according to the second data and a plurality of preset resource adjustment indexes; and according to the first adjustment strategy and the resource adjustment scheme, adjusting computing resources, storage resources and network resources of the virtual machine of the cloud resource pool to obtain resource configuration required by the first user. Through a plurality of adjustment strategies, the system can select different adjustment strategies according to different emphasis of the user, so that user requirements are met during resource adjustment. The balance between the resource utilization efficiency and the user experience is realized, and the user satisfaction is improved.

Description

Technical Field

[0001] The present application relates to the field of resource management technology, and in particular to a resource elastic scaling method, device, equipment, medium and product. Background Art

[0002] Resource elasticity refers to dynamically adjusting computing, storage or network resources according to actual demand to optimize resource utilization and cost. In cloud computing environments, this is usually achieved through automation tools that can automatically increase resources when the load increases, or reduce resources when the load decreases, thereby ensuring application performance while avoiding resource waste.

[0003] In existing technologies, most resource elastic scaling is based on resource indicators for prediction and adjustment. The system monitors the resource usage of applications and regularly collects and analyzes application performance data, such as memory consumption and network traffic. The system then uses the performance data to predict future resources required and adds or removes resources based on the prediction results.

[0004] However, the resource elasticity of the existing technology has the problem of poor user experience. Resource elasticity based on resource indicators mainly focuses on the use of system resources rather than user experience. Although this method can optimize resource utilization efficiency to a certain extent, it fails to fully consider factors related to user experience such as the response time of user requests, service stability and availability. The system may still be unable to provide high-quality services even when resources are sufficient, resulting in poor user experience. Summary of the invention

[0005] The present application provides a resource elastic scaling method, device, equipment, medium and product to solve the problem of poor user experience in the prior art.

[0006] In a first aspect, the present application provides a resource elastic scaling method, the method comprising:

[0007] According to one of the preset multiple adjustment strategies and the preset first user demand, a first adjustment strategy is obtained; wherein the first user demand refers to a set of computing resource demand, storage resource demand and network resource demand, and the adjustment strategy is used to indicate the adjustment order of computing resources, storage resources and network resources;

[0008] Collecting first data within a first preset time period in the past; wherein the first data refers to a resource usage set of the computing resources, the storage resources, and the network resources;

[0009] Predicting and obtaining second data within a second preset time period in the future based on the first data; wherein the second data refers to a set of predicted resource usage of the computing resources, the storage resources, and the network resources;

[0010] Generate a resource adjustment plan according to the second data and a plurality of preset resource adjustment indicators; wherein the plurality of resource adjustment indicators include computing resource indicators, storage resource indicators and network resource indicators, and the resource adjustment plan is used to indicate the increase or decrease of computing resources, storage resources and network resources of virtual machines in the cloud resource pool;

[0011] According to the first adjustment strategy and the resource adjustment scheme, the computing resources, the storage resources and the network resources of the virtual machines in the cloud resource pool are adjusted to obtain the resource configuration required by the first user.

[0012] In a possible design, obtaining the first adjustment strategy according to one of the preset multiple adjustment strategies and a preset first user requirement includes:

[0013] Acquire a strategy selection model; wherein the strategy selection model is trained based on multiple second user needs in a third preset time period in the past, and a second adjustment strategy corresponding to each second user need, the end time of the third preset time period is earlier than the start time of the first preset time period, and the strategy selection model is preset with the multiple adjustment strategies;

[0014] Inputting the first user requirement into the strategy selection model to obtain a third adjustment strategy output by the strategy selection model; wherein the third adjustment strategy refers to an adjustment strategy among the multiple adjustment strategies that has the highest degree of matching with the first user requirement;

[0015] The third adjustment strategy is optimized to obtain the first adjustment strategy.

[0016] In a possible design, optimizing the third adjustment strategy to obtain the first adjustment strategy includes:

[0017] According to the third adjustment strategy, at least one target strategy factor is determined from a plurality of preset strategy factors; wherein the strategy factor is used to optimize the third adjustment strategy;

[0018] Setting a weight for each of the target strategy factors;

[0019] The first adjustment strategy is generated according to the third adjustment strategy, the at least one target strategy factor, and the weight of each target strategy factor.

[0020] In a possible design, after adjusting the computing resources, the storage resources, and the network resources of each virtual machine in the cloud resource pool according to the first adjustment strategy and the resource adjustment scheme to obtain the resource configuration required by the first user, the method further includes:

[0021] Collect feedback data, and calculate the deviation between the feedback data and the second data; wherein the feedback data refers to the actual usage of the computing resources, the storage resources, and the network resources of the virtual machine, and the deviation refers to the difference between the actual usage of the resources of the virtual machine and the predicted usage of the resources;

[0022] The first adjustment strategy is iteratively optimized according to the deviation.

[0023] In a possible design, predicting, based on the first data, second data within a second preset time period in the future includes:

[0024] Decomposing the first data to obtain trend data, periodic data and residual data; wherein the trend data is used to describe the resource change trend of the virtual machine, the periodic data is used to describe the periodic change of the resources of the virtual machine, and the residual data refers to the data in the first data other than the trend data and the periodic data;

[0025] According to the trend data, the period data and the residual data, respectively predict the trend prediction data, the period prediction data and the residual prediction data within a second preset time period in the future;

[0026] The trend prediction data, the period prediction data and the residual prediction data are aggregated to obtain the second data.

[0027] In a possible design, predicting trend prediction data, period prediction data and residual prediction data within a second preset time period in the future respectively according to the trend data, the period data and the residual data includes:

[0028] Substituting the trend data into a preset linear regression model to obtain the trend prediction data output by the linear regression model;

[0029] Substituting the periodic data into a preset Fourier transform model to obtain the periodic prediction data output by the Fourier transform model;

[0030] The residual data is substituted into a preset support vector machine model to obtain the residual prediction data output by the support vector machine model.

[0031] In a second aspect, the present application provides a resource elastic scaling device, the device comprising:

[0032] A determination module, configured to obtain a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user requirement; wherein the first user requirement refers to a set of computing resource requirements, storage resource requirements, and network resource requirements, and the adjustment strategy is used to indicate an adjustment order of computing resources, storage resources, and network resources;

[0033] A collection module, used to collect first data within a first preset time period in the past; wherein the first data refers to a resource usage set of the computing resources, the storage resources, and the network resources;

[0034] A prediction module, configured to predict, based on the first data, second data within a second preset time period in the future; wherein the second data refers to a set of predicted resource usage of the computing resources, the storage resources, and the network resources;

[0035] a scheme generating module, configured to generate a resource adjustment scheme according to the second data and a plurality of preset resource adjustment indicators; wherein the plurality of resource adjustment indicators include a computing resource indicator, a storage resource indicator, and a network resource indicator, and the resource adjustment scheme is used to indicate to increase or decrease the computing resources, storage resources, and network resources of the virtual machines in the cloud resource pool;

[0036] An adjustment module is used to adjust the computing resources, storage resources and network resources of the virtual machines in the cloud resource pool according to the first adjustment strategy and the resource adjustment plan to obtain the resource configuration required by the first user.

[0037] In one possible design, the determining module includes:

[0038] an acquisition unit, configured to acquire a strategy selection model; wherein the strategy selection model is obtained by training based on a plurality of second user requirements in a third preset time period in the past and a second adjustment strategy corresponding to each second user requirement, the end time of the third preset time period is earlier than the start time of the first preset time period, and the strategy selection model is pre-set with the plurality of adjustment strategies;

[0039] A selection unit, configured to input the first user requirement into the policy selection model to obtain a third adjustment policy output by the policy selection model; wherein the third adjustment policy refers to an adjustment policy among the multiple adjustment policies that has the highest degree of matching with the first user requirement;

[0040] An optimization unit is used to optimize the third adjustment strategy to obtain the first adjustment strategy.

[0041] In a possible design, the optimization unit includes:

[0042] A determination component, configured to determine at least one target strategy factor from a plurality of preset strategy factors according to the third adjustment strategy; wherein the strategy factor is used to optimize the third adjustment strategy;

[0043] A weight setting component, used to set a weight for each of the target strategy factors;

[0044] A strategy generation component is used to generate the first adjustment strategy according to the third adjustment strategy, the at least one target strategy factor, and the weight of each target strategy factor.

[0045] In a possible design, the resource elastic scaling device further includes:

[0046] a collection module, configured to collect feedback data and calculate a deviation between the feedback data and the second data; wherein the feedback data refers to actual resource usage of the computing resources, the storage resources, and the network resources of the virtual machine, and the deviation refers to a difference between the actual resource usage of the virtual machine and the predicted resource usage;

[0047] An iteration module is used to iteratively optimize the first adjustment strategy according to the deviation.

[0048] In a possible design, the prediction module includes:

[0049] a decomposition unit, configured to decompose the first data to obtain trend data, periodic data, and residual data; wherein the trend data is used to describe the resource change trend of the virtual machine, the periodic data is used to describe the periodic change of the resources of the virtual machine, and the residual data refers to the data in the first data other than the trend data and the periodic data;

[0050] A prediction unit, configured to respectively predict trend prediction data, cycle prediction data and residual prediction data within a second preset time period in the future according to the trend data, the cycle data and the residual data;

[0051] An aggregation unit is used to aggregate the trend prediction data, the period prediction data and the residual prediction data to obtain the second data.

[0052] In a possible design, the prediction unit includes:

[0053] A first prediction component, used for substituting the trend data into a preset linear regression model to obtain the trend prediction data output by the linear regression model;

[0054] A second prediction component, used for substituting the periodic data into a preset Fourier transform model to obtain the periodic prediction data output by the Fourier transform model;

[0055] The third prediction component is used to substitute the residual data into a preset support vector machine model to obtain the residual prediction data output by the support vector machine model.

[0056] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0057] The memory stores computer-executable instructions;

[0058] When the processor executes the computer-executable instructions stored in the memory, it is used to implement the resource elastic scaling method as described in any one of the first aspects.

[0059] In a fourth aspect, 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 resource elastic scaling method as described in any one of the first aspects.

[0060] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the resource elastic scaling method as described in any one of the first aspects.

[0061] The present application provides a resource elastic scaling method, device, equipment, medium and product, the method comprising: obtaining a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user demand; collecting first data in a first preset time period in the past; predicting second data in a second preset time period in the future according to the first data; generating a resource adjustment plan according to the second data and a plurality of preset resource adjustment indicators; adjusting the computing resources, storage resources and network resources of the virtual machine in the cloud resource pool according to the first adjustment strategy and the resource adjustment plan to obtain the resource configuration required by the first user. The resource elastic scaling method of the present application introduces user demand and multiple adjustment strategies. The method not only predicts and adjusts based on resource usage, but also takes user demand into consideration to formulate a resource adjustment strategy that meets the actual needs of the user. This method collects and analyzes past resource usage data, predicts future resource demand, and generates a resource adjustment plan based on these predictions. The adjustment strategy ensures that user demand is met first during the resource adjustment process, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 A schematic diagram of the system architecture of the resource elastic scaling method provided in an embodiment of the present application;

[0064] Figure 2 A schematic diagram of an application scenario of the resource elastic scaling method provided in an embodiment of the present application;

[0065] Figure 3 A schematic diagram of a process for elastically scaling resources provided in an embodiment of the present application;

[0066] Figure 4 A schematic diagram of the structure of a resource elastic scaling device provided in an embodiment of the present application;

[0067] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0068] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0069] In the embodiments of the present application, words such as "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. Those skilled in the art will understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit the difference. It should be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in this application should not be interpreted as being more preferred or more advantageous than other embodiments or design. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way. In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more.

[0070] It should be noted that the "at..." in the embodiments of the present application can be the instant when a certain situation occurs, or can be a period of time after a certain situation occurs, and the embodiments of the present application do not specifically limit this. In addition, the data backup method provided in the embodiments of the present application is only used as an example, and the data backup method can also include more or less content.

[0071] In order to clearly describe the technical solutions of the embodiments of the present application, some terms and technologies involved in the embodiments of the present application are briefly introduced below:

[0072] Resource elastic scaling: is a technology that dynamically adjusts the configuration of computing resources. Its purpose is to optimize resource utilization efficiency and reduce operating costs. By monitoring system load and performance indicators, resource elastic scaling can provide additional resource support during peak demand periods and release excess resources during low demand periods, thereby achieving efficient resource management.

[0073] Adjustment strategy: A set of plans or methods developed to achieve specific goals and guide the modification and optimization of systems, processes or resources. In the field of technology, adjustment strategies may include algorithms, rules or procedures to ensure that the system can flexibly respond to changing needs and conditions in a dynamic environment.

[0074] Adjustment indicators: are key parameters or standards used to guide system adjustments. These indicators provide a quantitative benchmark to help determine when and how to make adjustments to optimize performance or meet specific needs. By monitoring these indicators, you can effectively guide the implementation of adjustment strategies to ensure that the system or process remains efficient and stable in a dynamic environment.

[0075] Trend data: refers to continuous data points collected over a period of time, reflecting the changing pattern and direction of a variable or phenomenon. By analyzing trend data, long-term growth, decline or cyclical fluctuation patterns can be identified, thus providing a basis for predicting future developments, formulating strategies and making decisions.

[0076] Cyclic data: refers to a data set with patterns or regularities that repeat at fixed time intervals, reflecting the cyclical change characteristics of a phenomenon or variable. By identifying and understanding the repetitive patterns in cyclical data, decision makers can make more accurate predictions to optimize resource allocation and strategy formulation.

[0077] Residual data: refers to the difference or deviation between the actual observed value and the model predicted value in data analysis. It reflects the part that the model cannot explain and is usually used to evaluate the accuracy and fitting effect of the model. By analyzing the residual data, it is possible to identify the systematic errors or outliers that may exist in the model, thus providing a basis for the improvement and optimization of the model.

[0078] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0079] The technical solution of the present invention is described in detail with specific embodiments below. 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 invention will be described below in conjunction with the accompanying drawings.

[0080] In order to clearly understand the technical solution of the present application, the solution of the prior art is first introduced in detail. In the prior art, most resource elastic scaling is achieved by implementing resource monitoring and data analysis. The system continuously monitors the resource usage of the application and regularly collects performance indicator data, such as processor usage, memory consumption, and network traffic. Subsequently, this data is input into the prediction algorithm to predict future resource requirements. Based on the prediction results, the system automatically adjusts the resource configuration to optimize performance by increasing or decreasing resources, thereby achieving dynamic response to changes in application load. This approach ensures efficient use of resources and improves the availability and responsiveness of the application.

[0081] However, the resource elastic scaling methods in the prior art have the problem of poor user experience. Resource elastic scaling based on resource indicators mainly focuses on the use of system resources rather than user experience. Although this method can improve resource utilization efficiency to a certain extent, due to the failure to fully consider the actual experience of user requests, the system may still be unable to provide high-quality services even when resources are sufficient. This resource management method that lacks user experience orientation may cause users to experience delays, service interruptions or instability during use, thereby reducing user satisfaction and overall experience.

[0082] Therefore, in order to solve the problem of poor user experience in resource elastic scaling methods in existing technologies, the study found that in order to solve this problem, user experience indicators can be incorporated into resource elastic scaling strategies to dynamically adjust resource allocation: ① Machine learning and data analysis technologies can be used to predict changes in user demand and adjust resource allocation in advance to ensure that the system can still provide high-quality services under high load. ② A real-time monitoring system can be established to collect user experience data, provide timely feedback and adjust resource allocation strategies to quickly respond to user experience issues. ③ A hybrid cloud architecture can be adopted to transfer part of the load to the public cloud so that resources can be quickly expanded when resource demand surges.

[0083] Specific:

[0084] An intelligent resource management platform can be built that integrates user experience indicators and machine learning algorithms to predict changes in user demand and dynamically adjust resource configuration by real-time monitoring of user behavior and system performance. The system can not only automatically expand or reduce resources according to the current load, but also prioritize resources for critical tasks or high-priority users to ensure service quality. At the same time, by implementing a real-time feedback mechanism and continuously collecting user experience data, the platform can continuously optimize resource allocation strategies, quickly respond to potential problems, and ultimately achieve a dual improvement in resource utilization efficiency and user experience.

[0085] The resource elastic scaling method of the embodiment of the present application determines the adjustment strategy according to the preset adjustment strategy and user needs, collects and analyzes the past resource usage data, predicts the future resource needs, and generates a resource adjustment plan to dynamically adjust the computing, storage and network resources of the virtual machine in the cloud resource pool, and finally meets the resource configuration required by the user. An improved resource elastic scaling method is proposed, which solves the problem of poor user experience in the prior art by introducing the consideration of adjustment strategy and user needs. The method not only pays attention to the usage of system resources, but also takes into account the specific needs and priorities of users. Through the preset multiple adjustment strategies, the system can select appropriate strategies according to the different focuses of users, so as to meet the actual needs of users when adjusting resources. In this way, a balance between resource utilization efficiency and user experience is achieved, and the responsiveness of the system and user satisfaction are enhanced.

[0086] Based on the above creative findings, the technical solution of the present application is proposed.

[0087] Figure 1 A schematic diagram of the system architecture of the resource elastic scaling method provided in the embodiment of the present application. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present application can be applied, in order to help those skilled in the art understand the technical content of the present application, but it does not mean that the embodiments of the present application cannot be used in other devices, systems, environments or scenarios.

[0088] like Figure 1As shown, the system architecture in which the method is located includes a front-end application module 100, a business gateway module 101, a scaling instance management module 102, a scaling configuration management module 103, a scaling policy management module 104, a resource monitoring module 105, a demand prediction module 106, a resource optimization module 107, an elastic scaling execution module 108 and a cloud unified service module 109. The front-end application module 100 can receive user needs and present the processing results to the user; the business gateway module 101 can perform interface authority control for identity authentication and authorization management; the scaling instance management module 102 can be responsible for the creation, deletion and query of scaling group instances, and maintain the move-in and move-out operations of cloud server instances; the scaling configuration management module 103 can be responsible for the definition and management of scaling group scaling configurations, corresponding to the management of basic configurations of cloud servers, including images, cloud hard disks and security groups; the scaling policy management module 104 can be responsible for the basic operations of creating, deleting and querying scaling policies and rules; the resource monitoring module 105 can collect relevant data in real time by monitoring changes in business volume, and provide a basis for demand prediction and resource optimization. The module also has the function of real-time monitoring of cloud resources, providing monitoring data for scaling execution and ensuring the smooth progress of the resource adjustment process. The demand prediction module 106 can predict business demand by collecting historical business data and establishing a prediction model based on model decomposition. The prediction results can reflect the trend of business demand changes in the future, and provide the predicted business demand to the resource optimization module 107 to provide a basis for resource adjustment. The resource optimization module 107 can establish a queuing network model for each application, analyze the current operating status and performance through the queuing network system, and finally find the most suitable number of service copies in the future to adapt to the predicted request sequence, ensuring the maximum utilization of resources and the stable operation of the business. The elastic scaling execution module 108 can be responsible for automatically executing the increase or decrease of resources according to the adjustment plan output by the resource optimization module. The cloud unified service module 109 can be responsible for encapsulating the cloud merchant cloud resource operation interface, forming a unified cloud resource operation interface, and providing it to the elastic scaling execution module 108 for calling.

[0089] In an embodiment of the present application, the front-end application module 100 can be a user interface application that can receive user input requirements and communicate with the back-end through a port, and present the processing results to the user in a graphical manner. The business gateway module 101 can be a gateway that uses identity authentication and authorization technology to manage interface permissions to ensure that only authorized users and applications can access specific services. The scaling instance management module 102 can be an automated management tool that uses the cloud service provider's port to create, delete, and query cloud server instances, and manages the addition and removal of instances through programming logic. The scaling configuration management module 103 can be a configuration management system that uses templated configuration files to define and manage the configuration of scaling groups, including basic settings such as images, cloud hard disks, and security groups. The scaling policy management module 104 can be a policy management service that provides a user interface to create, delete, and query scaling policies and rules, and stores policy information through a database to support automated decision-making. The resource monitoring module 105 can be a real-time monitoring system that uses a data collection agent and a time series database to monitor business volume changes, collect and store relevant data, and provide real-time monitoring information to support resource adjustment. The demand prediction module 106 can be a prediction service based on machine learning, which uses historical business data to train models, predicts future business needs through model decomposition technology, and provides the results to the resource optimization module. The resource optimization module 107 can be an optimization analysis tool that uses a queuing theory model to analyze the current system state and calculate the optimal number of service copies to maximize resource utilization and stable business operation. The elastic scaling execution module 108 can be an automated execution engine that automatically executes resource increase or decrease operations based on the output of the resource optimization module by calling the port of the cloud unified service module to achieve dynamic adjustment of cloud resources. The cloud unified service module 109 can be an encapsulation layer that provides a unified cloud resource operation interface for the elastic scaling execution module to call to achieve resource management across cloud platforms.

[0090] The following introduces the application scenarios of the resource elastic scaling method provided by the embodiment of the present invention. Figure 2 Schematic diagram of application scenarios of the resource elastic scaling method provided in the embodiment of the present application. Figure 2 As shown, the application scenario includes a mobile terminal 201 and a server 202. The mobile terminal 201 obtains user requirements and sends the user requirements to the server 202. The server 202 predicts the resources required after a preset time period in the future based on the user requirements and existing resources, and sends the predicted resource configuration to the mobile terminal 201.

[0091] The embodiments of the present invention are described below in conjunction with the accompanying drawings.

[0092] Figure 3The following is a flow chart of a resource elastic scaling method provided in an embodiment of the present application. Figure 3 As shown, in this embodiment, the execution subject of the embodiment of the present invention is a server. Then the resource elastic scaling method provided by this embodiment includes the following steps:

[0093] S301: Obtain a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user requirement.

[0094] Specifically, a selection model can be trained based on historical data, the first user demand can be input into the selection model, and then the output of the selection model can be optimized to obtain the first adjustment strategy. By dynamically selecting and applying the most appropriate adjustment strategy, the system can respond to user needs more effectively, optimize the resource allocation process, avoid over-allocation or under-allocation of resources, improve the flexibility and efficiency of the overall system, and ensure that user needs are met in a timely manner. Among them, the first user demand refers to the collection of computing resource requirements, storage resource requirements, and network resource requirements, and the adjustment strategy is used to indicate the adjustment order of computing resources, storage resources, and network resources.

[0095] S302: Collect first data within a first preset time period in the past.

[0096] Specifically, the first data within the first preset time period in the past can be collected through monitoring tools. These tools are deployed in the cloud resource pool to continuously collect and record the usage of computing resources, storage resources, and network resources. These data usually include indicators such as processor utilization, memory usage, storage space occupancy, and network bandwidth usage. In this way, the system can obtain detailed historical resource usage information and provide data support for subsequent resource demand prediction and adjustment. By analyzing historical data, the system can predict future resource requirements, optimize resource allocation, and ensure that user needs are met. Among them, the first data refers to the resource usage collection of computing resources, storage resources, and network resources.

[0097] S303: Predict and obtain second data within a second preset time period in the future based on the first data.

[0098] Specifically, the second data within the second preset time period in the future can be predicted by applying machine learning algorithms or time series analysis methods. These algorithms use the first data collected in the past to identify patterns and trends in resource usage, thereby predicting future resource requirements. By accurately predicting future resource requirements, the system can adjust resource configuration in advance to avoid over-allocation or under-allocation of resources, improve the system's responsiveness and efficiency, and optimize cost management to ensure efficient use of resources. Among them, the second data refers to the set of predicted resource usage of computing resources, storage resources, and network resources.

[0099] S304: Generate a resource adjustment plan according to the second data and a plurality of preset resource adjustment indicators.

[0100] Specifically, a resource adjustment plan can be generated through a decision algorithm or an optimization model. These algorithms formulate a resource adjustment strategy based on the predicted second data and the preset resource adjustment indicators. By generating a resource adjustment plan, the system can dynamically adapt to changing user needs, ensure the efficient allocation and use of resources, improve the overall service quality, and control operating costs to avoid resource waste. Among them, the multiple resource adjustment indicators include computing resource indicators, storage resource indicators, and network resource indicators. The resource adjustment plan is used to indicate the increase or decrease of computing resources, storage resources, and network resources of virtual machines in the cloud resource pool.

[0101] S305: According to the first adjustment strategy and resource adjustment plan, the computing resources, storage resources, and network resources of the virtual machines in the cloud resource pool are adjusted to obtain resource configuration required by the first user.

[0102] Specifically, the computing resources, storage resources, and network resources of the virtual machines in the cloud resource pool can be adjusted through automated resource management tools or platforms to obtain the resource configuration required by the first user. These tools automatically perform resource allocation operations on virtual machines based on the generated resource adjustment plan and preset adjustment strategy, such as starting or shutting down virtual machine instances, adjusting memory allocation, expanding or reducing storage volumes, and adjusting network bandwidth. Through automated and strategic resource adjustments, the system can quickly respond to changing user needs, ensure that resource configuration matches actual needs, improve resource utilization, and service performance, while reducing manual intervention and management complexity.

[0103] The present embodiment provides a resource elastic scaling method, which includes: obtaining a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user demand; collecting first data in a first preset time period in the past; predicting second data in a second preset time period in the future according to the first data; generating a resource adjustment plan according to the second data and a plurality of preset resource adjustment indicators; adjusting the computing resources, storage resources and network resources of the virtual machine in the cloud resource pool according to the first adjustment strategy and the resource adjustment plan to obtain the resource configuration required by the first user. A resource elastic scaling method achieves the following technical effects: by introducing the consideration of adjustment strategies and user needs, the problem of poor user experience in the prior art is solved. The method not only focuses on the use of system resources, but also considers the specific needs and priorities of users. Through the preset multiple adjustment strategies, the system can select appropriate strategies according to the different focuses of users, so as to meet the actual needs of users when adjusting resources. In this way, a balance between resource utilization efficiency and user experience is achieved, and user satisfaction is enhanced.

[0104] In a possible design, S301 obtains a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user requirement, including:

[0105] S3011. Obtain a strategy selection model.

[0106] Specifically, a policy selection model can be obtained through machine learning technology. Collect user needs and corresponding adjustment strategies in the past period of time, and use these data to train a policy selection model so that it can predict the most matching adjustment strategy based on the input user needs. The model can automatically select the adjustment strategy that best suits the current user needs, improve the efficiency and accuracy of resource allocation, thereby optimizing the use of the cloud resource pool and reducing resource waste and response time. Among them, the policy selection model is trained based on multiple second user needs in the past third preset time period, and the second adjustment strategy corresponding to each second user need. The end time of the third preset time period is earlier than the start time of the first preset time period, and multiple adjustment strategies are preset in the policy selection model.

[0107] S3012: Input the first user demand into the strategy selection model to obtain a third adjustment strategy output by the strategy selection model.

[0108] Specifically, the third adjustment strategy output by the strategy selection model can be obtained by passing the user demand data as input features to a pre-trained machine learning model. The model will calculate based on the input features and output a best-matching adjustment strategy as the third adjustment strategy. Through automated model prediction, the adjustment strategy that best suits the current user needs can be quickly identified and selected, thereby improving the efficiency and flexibility of resource management and reducing human intervention and decision-making time. Among them, the third adjustment strategy refers to the adjustment strategy that best matches the first user's needs among multiple adjustment strategies.

[0109] S3013. Optimize the third adjustment strategy to obtain the first adjustment strategy.

[0110] Specifically, the first adjustment strategy can be obtained by adjusting the parameters of the third adjustment strategy by applying a machine learning algorithm or an optimization algorithm. This may involve analyzing historical data, current resource usage, and predicted resource demand to dynamically adjust the priority and order of resource allocation. This processing improves the efficiency and accuracy of resource allocation, ensuring that the overall performance of the cloud resource pool is optimized while meeting user needs.

[0111] In a possible design, S3013 optimizes the third adjustment strategy to obtain a first adjustment strategy, including:

[0112] S30131. According to a third adjustment strategy, determine at least one target strategy factor from a plurality of preset strategy factors.

[0113] Specifically, by analyzing the characteristics and requirements of the third adjustment strategy and matching it with the preset policy factors, the key factors that can optimize the strategy can be identified. Machine learning algorithms or heuristic rules can be used to evaluate the degree of influence of each policy factor on the third adjustment strategy, so as to select the most influential factor as the target policy factor. By accurately identifying and applying these target policy factors, the adaptability and efficiency of the adjustment strategy can be improved, the dynamic changes of user needs and resource usage can be met, and the resource allocation process can be optimized. Among them, the policy factor is used to optimize the third adjustment strategy.

[0114] S30132. Set a weight for each target strategy factor.

[0115] Specifically, historical data analysis and machine learning techniques can be used to evaluate the relative importance of each target strategy factor to the resource adjustment effect, and then weights can be set for each target strategy factor based on the importance. Through regression analysis or expert experience, the contribution of each factor to the optimization of the adjustment strategy can be quantified, and the corresponding weights can be assigned accordingly. By reasonably allocating weights, the adjustment strategy can be optimized so that resource allocation is more in line with actual needs, and resource utilization efficiency and system responsiveness can be improved.

[0116] S30133. Generate a first adjustment strategy according to the third adjustment strategy, at least one target strategy factor, and the weight of each target strategy factor.

[0117] Specifically, an optimization model can be constructed, which takes the third adjustment strategy as the basis and adjusts it in combination with the target strategy factors and their weights. The target strategy factors and their weights are input into the optimization algorithm to adjust and optimize the parameters and priorities of the third adjustment strategy, thereby generating a more accurate and efficient first adjustment strategy. By comprehensively considering the influence and weight of each factor, the generated first adjustment strategy can adapt to dynamic resource requirements and improve the resource management efficiency and response speed of the system.

[0118] The technical effect of the scheme in this embodiment is: by analyzing past user needs and corresponding adjustment strategies, a model that can identify demand characteristics is trained, so that the system can quickly select the optimal adjustment strategy when facing new user needs, and further optimize to adapt to current specific needs. This method not only improves the accuracy and efficiency of resource adjustment, but also reduces the necessity of human intervention, and enhances the system's adaptive ability and response speed when dealing with complex and changing resource needs; by selecting the factors that have the greatest impact on strategy optimization and reasonably allocating their weights, it ensures that when generating the first adjustment strategy, the contribution and priority of each factor are fully considered. This optimization process not only improves the efficiency and effect of resource allocation, but also enables the system to cope with changes in different user needs and resource usage scenarios, ultimately achieving higher resource utilization and faster response time, thereby enhancing the overall performance of the system and user satisfaction.

[0119] In a possible design, after adjusting the computing resources, storage resources, and network resources of the virtual machines in the cloud resource pool according to the first adjustment strategy and the resource adjustment scheme to obtain the resource configuration required by the first user, S305 further includes:

[0120] S306: Collect feedback data, and calculate the deviation between the feedback data and the second data.

[0121] Specifically, feedback data can be collected by monitoring the actual resource usage of the virtual machine in the cloud resource pool, and the deviation between the feedback data and the second data can be calculated. The actual usage data of the computing resources, storage resources and network resources of the virtual machine can be collected in real time using a monitoring tool or an agent, and these data can be compared with the previously predicted resource usage to calculate the difference or deviation between the actual usage and the predicted usage. By identifying and quantifying the deviation between the predicted and actual usage, data support can be provided for subsequent adjustment strategies, thereby achieving iterative optimization of the adjustment strategy, improving the accuracy and efficiency of resource configuration, and ultimately meeting the dynamic needs of users. Among them, feedback data refers to the actual resource usage of the computing resources, storage resources and network resources of the virtual machine, and deviation refers to the difference between the actual resource usage of the virtual machine and the predicted resource usage.

[0122] S307: Iteratively optimize the first adjustment strategy according to the deviation.

[0123] Specifically, the first adjustment strategy can be iteratively optimized by using a feedback control mechanism. Based on the calculated deviation, the system can adjust the parameters of the policy selection model or optimization algorithm to gradually reduce the difference between the prediction and actual use. For example, a machine learning algorithm can be used to retrain the policy selection model, or adjust the weights of the target policy factors to better adapt to changes in actual resource requirements. By continuously optimizing the adjustment strategy, the system can dynamically adapt to changes in the environment and demand, improve the accuracy and response speed of resource configuration, and effectively utilize computing, storage, and network resources to meet user needs.

[0124] The technical effect of the solution in this embodiment is that by collecting the actual resource usage of the virtual machine and calculating the deviation from the predicted data, the system can obtain feedback information on the accuracy of resource configuration. This feedback mechanism allows the system to identify deficiencies in the prediction model or adjustment strategy and perform iterative optimization, thereby gradually improving the accuracy and efficiency of resource configuration. Ultimately, this adaptive optimization process can ensure that the dynamic adjustment of resources is more in line with actual user needs, improving the overall performance of the system and user satisfaction.

[0125] In a possible design, S303 predicts, based on the first data, second data within a second preset time period in the future, including:

[0126] S3031. Decompose the first data to obtain trend data, period data and residual data.

[0127] Specifically, the first data can be decomposed by time series decomposition technology to obtain trend data, periodic data and residual data, for example, using an addition or multiplication model to decompose the time series data into long-term trends, periodic fluctuations and random residual parts. This processing can accurately capture different change patterns of resource usage, thereby improving the accuracy of the prediction model. By analyzing and predicting trend, periodic and residual data separately, it is possible to adapt to dynamic changes in resource demand and improve the efficiency and response speed of resource allocation. Among them, trend data is used to describe the resource change trend of the virtual machine, periodic data is used to describe the periodic changes of the resources of the virtual machine, and residual data refers to the data in the first data other than trend data and periodic data.

[0128] S3032. According to the trend data, the period data and the residual data, respectively predict the trend prediction data, the period prediction data and the residual prediction data within a second preset time period in the future.

[0129] Specifically, different forecasting models can be applied to predict the trend forecast data, cycle forecast data, and residual forecast data for the second preset time period in the future. Trend data can usually be predicted using models such as linear regression or polynomial regression; cycle data can be predicted using methods such as Fourier transform or seasonal decomposition; residual data can be predicted using an autoregressive integrated moving average model. This separate forecasting method can apply the most suitable forecasting model to data with different characteristics, thereby improving the accuracy and reliability of the overall forecast and ultimately optimizing the decision-making process of resource allocation.

[0130] S3033. Aggregate the trend prediction data, the period prediction data, and the residual prediction data to obtain second data.

[0131] Specifically, the second data can be obtained by superimposing the prediction results of each part. In the additive model, the aggregation process is to simply add the trend prediction data, period prediction data and residual prediction data; in the multiplicative model, these data are multiplied or added after logarithmic transformation. This aggregation method can comprehensively consider the long-term trend, periodic fluctuations and random changes in resource use, generate comprehensive prediction data, thereby providing a more accurate basis for resource adjustment plans and improving the efficiency and accuracy of resource allocation.

[0132] The technical effect of the scheme in this embodiment is: by decomposing the historical resource usage data into three parts: trend, cycle and residual, and predicting them separately, different change patterns of resource demand can be captured and analyzed more accurately. This method allows the most suitable prediction models to be applied to long-term trends, cyclical fluctuations and random changes, respectively, thereby improving the accuracy of the overall prediction. Ultimately, this refined prediction capability can support effective resource adjustment plans, optimize resource allocation, improve the system's response speed and resource utilization efficiency, and meet the dynamically changing needs of users.

[0133] In a possible design, S3032 predicts trend prediction data, cycle prediction data, and residual prediction data in a second preset time period in the future according to the trend data, the cycle data, and the residual data, respectively, including:

[0134] S30321. Substitute the trend data into a preset linear regression model to obtain trend prediction data output by the linear regression model.

[0135] Specifically, historical trend data can be preprocessed first to ensure data quality and consistency. Then, the preprocessed trend data is input into a pre-trained linear regression model, which predicts future trend changes by learning the linear relationship in the data. The linear regression model can effectively capture the long-term linear trend in the data, thereby providing accurate predictions of future resource demand, helping to optimize resource allocation and improve resource utilization efficiency.

[0136] S30322. Substitute the periodic data into a preset Fourier transform model to obtain periodic prediction data output by the Fourier transform model.

[0137] Specifically, the periodic data can be sampled and preprocessed first to ensure the integrity and accuracy of the data. Then, the processed periodic data is input into a preset Fourier transform model, which identifies and extracts the periodic components in the data by converting the time domain data into the frequency domain. Fourier transform can effectively decompose complex periodic signals and identify the main periodic patterns, thereby providing accurate predictions of periodic resource requirements and helping to achieve reasonable allocation of resources.

[0138] S30323. Substitute the residual data into a preset support vector machine model to obtain residual prediction data output by the support vector machine model.

[0139] Specifically, the residual data can be standardized or normalized first to improve the model training effect and prediction accuracy. Then, the processed residual data is input into the pre-trained support vector machine model, which predicts future residual changes by learning the nonlinear relationship in the data. The support vector machine model can capture the complex nonlinear patterns in the data, thereby improving the prediction accuracy of resource demand fluctuations, supplementing the deficiencies of trend and cycle predictions, and ultimately optimizing resource allocation strategies.

[0140] The technical effect of the scheme in this embodiment is: by substituting trend data, periodic data and residual data into the linear regression model, Fourier transform model and support vector machine model respectively, different components in resource usage can be predicted. The linear regression model can effectively capture long-term trend changes, the Fourier transform model can identify periodic fluctuations, and the support vector machine model can handle complex nonlinear residuals. This multi-model combination method improves the accuracy and reliability of prediction.

[0141] Figure 4 This is a schematic diagram of the structure of the resource elastic scaling device provided in the embodiment of the present application. Figure 4 As shown, the resource elastic expansion device includes:

[0142] Determination module 401 is used to obtain a first adjustment strategy based on one of multiple preset adjustment strategies and a preset first user demand; wherein the first user demand refers to a set of computing resource requirements, storage resource requirements and network resource requirements, and the adjustment strategy is used to indicate the adjustment order of computing resources, storage resources and network resources.

[0143] The collection module 402 is used to collect first data within a first preset time period in the past; wherein the first data refers to a collection of resource usage conditions of computing resources, storage resources, and network resources.

[0144] The prediction module 403 is used to predict the second data within a second preset time period in the future based on the first data; wherein the second data refers to a set of predicted resource usage of computing resources, storage resources and network resources.

[0145] The plan generation module 404 is used to generate a resource adjustment plan based on the second data and multiple preset resource adjustment indicators; wherein the multiple resource adjustment indicators include computing resource indicators, storage resource indicators and network resource indicators, and the resource adjustment plan is used to indicate the increase or decrease of computing resources, storage resources and network resources of the virtual machines in the cloud resource pool.

[0146] The adjustment module 405 is used to adjust the computing resources, storage resources and network resources of the virtual machines in the cloud resource pool according to the first adjustment strategy and the resource adjustment plan to obtain the resource configuration required by the first user.

[0147] In one possible design, the determination module 401 includes:

[0148] An acquisition unit is used to acquire a strategy selection model; wherein the strategy selection model is trained based on multiple second user needs within a third preset time period in the past, and a second adjustment strategy corresponding to each second user need, the end time of the third preset time period is earlier than the start time of the first preset time period, and multiple adjustment strategies are preset in the strategy selection model.

[0149] The selection unit is used to input the first user demand into the strategy selection model to obtain a third adjustment strategy output by the strategy selection model; wherein the third adjustment strategy refers to the adjustment strategy with the highest matching degree with the first user demand among multiple adjustment strategies.

[0150] The optimization unit is used to optimize the third adjustment strategy to obtain the first adjustment strategy.

[0151] In one possible design, the optimization unit includes:

[0152] The determination component is used to determine at least one target strategy factor from a plurality of preset strategy factors according to the third adjustment strategy; wherein the strategy factor is used to optimize the third adjustment strategy.

[0153] The weight setting component is used to set the weight for each target strategy factor.

[0154] The strategy generation component is used to generate a first adjustment strategy according to the third adjustment strategy, at least one target strategy factor, and a weight of each target strategy factor.

[0155] In a possible design, the resource elastic scaling device further includes:

[0156] A collection module is used to collect feedback data and calculate the deviation between the feedback data and the second data; wherein the feedback data refers to the actual resource usage of the virtual machine's computing resources, storage resources and network resources, and the deviation refers to the difference between the actual resource usage of the virtual machine and the predicted resource usage.

[0157] The iteration module is used to iteratively optimize the first adjustment strategy according to the deviation.

[0158] In one possible design, the prediction module 403 includes:

[0159] A decomposition unit is used to decompose the first data to obtain trend data, periodic data and residual data; wherein the trend data is used to describe the resource change trend of the virtual machine, the periodic data is used to describe the periodic change of the resources of the virtual machine, and the residual data refers to the data in the first data other than the trend data and the periodic data.

[0160] The prediction unit is used to predict the trend prediction data, the cycle prediction data and the residual prediction data within a second preset time period in the future according to the trend data, the cycle data and the residual data.

[0161] The aggregation unit is used to aggregate the trend prediction data, the period prediction data and the residual prediction data to obtain the second data.

[0162] In a possible design, the prediction unit includes:

[0163] The first prediction component is used to substitute the trend data into a preset linear regression model to obtain the trend prediction data output by the linear regression model.

[0164] The second prediction component is used to substitute the periodic data into a preset Fourier transform model to obtain the periodic prediction data output by the Fourier transform model.

[0165] The third prediction component is used to substitute the residual data into a preset support vector machine model to obtain the residual prediction data output by the support vector machine model.

[0166] The resource elastic scaling device provided in this embodiment can be executed Figure 3 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 3 The method embodiments shown are similar and will not be described in detail here.

[0167] Figure 5 The hardware structure diagram of the electronic device provided in the embodiment of the present application is shown in FIG. Figure 5 As shown, the electronic device includes: at least one processor 510 and a memory 520. The electronic device also includes a communication component 530. The processor 510, the memory 520 and the communication component 530 are connected via a bus 540.

[0168] In the specific implementation process, at least one processor 510 executes the computer execution instructions stored in the memory 520, so that the at least one processor 510 is used to implement the resource elastic scaling method of the above embodiment.

[0169] The specific implementation process of the processor 510 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0170] In the above embodiments, it should be understood that the processor 510 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The steps of the method disclosed in the invention may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0171] The memory 520 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0172] The bus 540 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus 540 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus 540 in the drawings of the present application is not limited to only one bus or one type of bus.

[0173] The above-mentioned functions implemented by the electronic device and the main control device introduce the scheme provided by the embodiment of the present invention. It can be understood that in order to implement the above-mentioned functions, the electronic device or the main control device includes a hardware structure and / or software module corresponding to the execution of each function. In combination with the units and algorithm steps of each example described in the embodiment disclosed in the embodiment of the present invention, the embodiment of the present invention can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present invention.

[0174] The embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the resource elastic scaling method of the above embodiment. In the specific implementation of the above resource elastic scaling method, each module can be implemented as a processor.

[0175] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0176] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in an electronic device or a main control device.

[0177] The embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the resource elastic scaling method of the above embodiment.

[0178] The computer program is stored in a readable storage medium. At least one processor can read the computer program from the readable storage medium. At least one processor executes the computer program to execute the solution provided in any of the above embodiments.

[0179] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by hardware related to program instructions. The above program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above method embodiments are executed; and the above storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0180] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments, and the above embodiments are only used to illustrate the technical solution of the present application rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A resource elastic scaling method, characterized in that: The method comprises: According to one of the preset multiple adjustment strategies and the preset first user demand, a first adjustment strategy is obtained; wherein the first user demand refers to a set of computing resource demand, storage resource demand and network resource demand, and the adjustment strategy is used to indicate the adjustment order of computing resources, storage resources and network resources; Collecting first data within a first preset time period in the past; wherein the first data refers to a resource usage set of the computing resources, the storage resources, and the network resources; Predicting and obtaining second data within a second preset time period in the future based on the first data; wherein the second data refers to a set of predicted resource usage of the computing resources, the storage resources, and the network resources; Generate a resource adjustment plan according to the second data and a plurality of preset resource adjustment indicators; wherein the plurality of resource adjustment indicators include computing resource indicators, storage resource indicators and network resource indicators, and the resource adjustment plan is used to indicate the increase or decrease of computing resources, storage resources and network resources of virtual machines in the cloud resource pool; According to the first adjustment strategy and the resource adjustment scheme, the computing resources, the storage resources and the network resources of the virtual machines in the cloud resource pool are adjusted to obtain the resource configuration required by the first user.

2. The resource elastic scaling method according to claim 1, characterized in that: The obtaining of a first adjustment strategy according to one of the preset multiple adjustment strategies and a preset first user requirement includes: Acquire a strategy selection model; wherein the strategy selection model is trained based on multiple second user needs in a third preset time period in the past, and a second adjustment strategy corresponding to each second user need, the end time of the third preset time period is earlier than the start time of the first preset time period, and the strategy selection model is preset with the multiple adjustment strategies; Inputting the first user requirement into the strategy selection model to obtain a third adjustment strategy output by the strategy selection model; wherein the third adjustment strategy refers to an adjustment strategy among the multiple adjustment strategies that has the highest degree of matching with the first user requirement; The third adjustment strategy is optimized to obtain the first adjustment strategy.

3. The resource elastic scaling method according to claim 2, characterized in that: The step of optimizing the third adjustment strategy to obtain the first adjustment strategy includes: According to the third adjustment strategy, at least one target strategy factor is determined from a plurality of preset strategy factors; wherein the strategy factor is used to optimize the third adjustment strategy; Setting a weight for each of the target strategy factors; The first adjustment strategy is generated according to the third adjustment strategy, the at least one target strategy factor, and the weight of each target strategy factor.

4. The resource elastic scaling method according to claim 2, characterized in that: After adjusting the computing resources, the storage resources, and the network resources of each virtual machine in the cloud resource pool according to the first adjustment strategy and the resource adjustment scheme to obtain the resource configuration required by the first user, the method further includes: Collect feedback data, and calculate the deviation between the feedback data and the second data; wherein the feedback data refers to the actual usage of the computing resources, the storage resources, and the network resources of the virtual machine, and the deviation refers to the difference between the actual usage of the resources of the virtual machine and the predicted usage of the resources; The first adjustment strategy is iteratively optimized according to the deviation.

5. The resource elastic scaling method according to claim 1, characterized in that: The step of predicting, based on the first data, second data within a second preset time period in the future includes: Decomposing the first data to obtain trend data, periodic data and residual data; wherein the trend data is used to describe the resource change trend of the virtual machine, the periodic data is used to describe the periodic change of the resources of the virtual machine, and the residual data refers to the data in the first data other than the trend data and the periodic data; According to the trend data, the period data and the residual data, respectively predict the trend prediction data, the period prediction data and the residual prediction data within a second preset time period in the future; The trend prediction data, the period prediction data and the residual prediction data are aggregated to obtain the second data.

6. The resource elastic scaling method according to claim 5, characterized in that: The step of respectively predicting trend prediction data, period prediction data and residual prediction data within a second preset time period in the future according to the trend data, the period data and the residual data comprises: Substituting the trend data into a preset linear regression model to obtain the trend prediction data output by the linear regression model; Substituting the periodic data into a preset Fourier transform model to obtain the periodic prediction data output by the Fourier transform model; The residual data is substituted into a preset support vector machine model to obtain the residual prediction data output by the support vector machine model.

7. A resource elastic expansion device, characterized in that: include: A determination module, configured to obtain a first adjustment strategy according to one of a plurality of preset adjustment strategies and a preset first user requirement; wherein the first user requirement refers to a set of computing resource requirements, storage resource requirements, and network resource requirements, and the adjustment strategy is used to indicate an adjustment order of computing resources, storage resources, and network resources; A collection module, used to collect first data within a first preset time period in the past; wherein the first data refers to a resource usage set of the computing resources, the storage resources, and the network resources; A prediction module, configured to predict, based on the first data, second data within a second preset time period in the future; wherein the second data refers to a set of predicted resource usage of the computing resources, the storage resources, and the network resources; a scheme generating module, configured to generate a resource adjustment scheme according to the second data and a plurality of preset resource adjustment indicators; wherein the plurality of resource adjustment indicators include a computing resource indicator, a storage resource indicator, and a network resource indicator, and the resource adjustment scheme is used to indicate to increase or decrease the computing resources, storage resources, and network resources of the virtual machines in the cloud resource pool; An adjustment module is used to adjust the computing resources, storage resources and network resources of the virtual machines in the cloud resource pool according to the first adjustment strategy and the resource adjustment plan to obtain the resource configuration required by the first user.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; When the processor executes the computer-executable instructions stored in the memory, it is used to implement the resource elastic scaling method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: 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 resource elastic scaling method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program is used to implement the resource elastic scaling method according to any one of claims 1 to 6.

Images