A multi-scenario service elastic scaling method based on a policy model and related components

Abstract

The present invention discloses a multi-scenario service elastic scaling method and related components based on a policy model, which relates to the field of application service instance allocation. A server instance allocation strategy for each preset application service is provided, and the corresponding strategy for these application services is determined according to the received user instructions. If it is a direct allocation strategy, each application service is allocated a corresponding number of server instances. If it is a time allocation strategy, each application service is allocated a corresponding number of server instances at the current time according to the correspondence between the current time and the server instance. If it is a load allocation strategy, each application service is allocated a corresponding number of server instances under the current load according to the load of each application service. There is no need for manual real-time detection of the load of these application services, and no need for manual adjustment of the number of server instances allocated to each application service, thereby reducing the cost and workload of application service operation and maintenance.

Description

Technical Field

[0001] The present invention relates to the field of application service operation and maintenance, and in particular to a multi-scenario service elastic scaling method based on a policy model and related components. Background Art

[0002] When a client accesses a specific application service, the application service processes the access through its server instances. When the number of accesses is too high, the load on these server instances will also increase. If the total load of the server instances exceeds the load quota allocated to the application service, the application service will crash. If the number of server instances allocated to the application service is simply increased, when the total load of the application service is low, it will waste server instance resources. When allocating server instances to a specific application service, the existing technology usually manually detects the real-time total load of the server instances of the application service and manually adjusts the number of server instances allocated to the application service based on the total load. However, this manual detection method increases the cost of service operation and maintenance, and the manual adjustment is labor-intensive. Summary of the Invention

[0003] The purpose of the present invention is to provide a multi-scenario service elastic scaling method and related components based on a policy model, which does not require manual real-time detection of the load of these application services, nor does it require manual adjustment of the number of server instances allocated to each application service, thereby reducing the cost and workload of application service operation and maintenance.

[0004] To solve the above technical problems, the present invention provides a multi-scenario service elastic scaling method based on a policy model, comprising:

[0005] Determine the server instance allocation strategy corresponding to each pre-specified application service according to the received user instruction, wherein the server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load distribution strategy;

[0006] When the server instance allocation strategy is the direct allocation strategy, a preset number of server instances corresponding to each application service is allocated to each application service;

[0007] When the server instance allocation strategy is the time allocation strategy, based on the correspondence between time and the number of server instances of each application service, a preset number of server instances corresponding to the application service at the current time is allocated to each application service;

[0008] When the server instance allocation strategy is the load distribution strategy, based on the correspondence between the load of each application service and the number of server instances, a preset number of server instances corresponding to the current load of each application service is allocated to each application service.

[0009] Preferably, allocating a preset number of server instances corresponding to each application service at the current time to each application service based on the corresponding relationship between time and the number of server instances of each application service includes:

[0010] Determine whether the current time exists in a preset allocation time library;

[0011] If so, a preset number of server instances corresponding to each application service at the current time are allocated to each application service.

[0012] Preferably, it also includes:

[0013] When an update instruction is received, the preset quantity corresponding to each of the application services corresponding to the update instruction is changed to the preset update quantity corresponding to the application service, the update instruction including N application services and the preset quantity corresponding to N application services, where N is a positive integer.

[0014] Preferably, before determining the corresponding server instance allocation strategy according to the received user instruction, the method further includes:

[0015] Utilize Vue and iView to build a policy model including the server instance allocation policy;

[0016] Determine the corresponding server instance allocation strategy based on the received user instructions, including:

[0017] According to the received user instruction, a server instance allocation policy corresponding to the user instruction is determined in the policy model.

[0018] Preferably, based on the correspondence between the load of each application service and the number of server instances, allocating a preset number of server instances corresponding to the current load of each application service to each application service includes:

[0019] Determining the current load of each of the server instances in each of the application services;

[0020] For any of the application services, determining an average value of current loads of all the server instances in the application service;

[0021] Based on the correspondence between the average value of the current loads of all server instances in each of the application services and the number of server instances, a preset number of server instances corresponding to the average value of each application service is allocated to each of the application services.

[0022] Preferably, determining the current load of each of the server instances in each of the application services includes:

[0023] The current load of each of the server instances in each of the application services is determined using a metrics-server.

[0024] Preferably, based on the correspondence between the load of each application service and the number of server instances, allocating a preset number of server instances corresponding to the current load of each application service to each application service includes:

[0025] HPA is utilized to allocate a preset number of server instances corresponding to the current load of each application service to each application service based on the corresponding relationship between the load of each application service and the number of server instances.

[0026] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-scenario service elastic scaling method based on the policy model as described above are implemented.

[0027] This application also provides a multi-scenario service elastic scaling system based on a policy model, including:

[0028] a strategy determination unit, configured to determine a corresponding server instance allocation strategy according to a received user instruction, wherein the server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load distribution strategy;

[0029] a first allocation unit, configured to allocate a preset number of server instances corresponding to each pre-specified application service to each pre-specified application service when the server instance allocation strategy is the direct allocation strategy;

[0030] a second allocation unit, configured to allocate, to each application service, a preset number of server instances corresponding to each application service at a current time based on a correspondence between time and the number of server instances of each application service when the server instance allocation strategy is the time allocation strategy;

[0031] The third allocation unit is used to allocate a preset number of server instances corresponding to the current load of each application service based on the correspondence between the load of each application service and the number of server instances when the server instance allocation strategy is the load distribution strategy.

[0032] This application also provides a multi-scenario service elastic scaling device based on a policy model, including:

[0033] memory for storing computer programs;

[0034] A processor is used to implement the steps of the multi-scenario service elastic scaling method based on the policy model as described above when executing the computer program.

[0035] The present invention provides a multi-scenario service elastic scaling method and related components based on a policy model, which relates to the field of instance allocation for application services. A server instance allocation strategy is set for each preset application service, and the strategy includes a direct allocation strategy, a time allocation strategy and a load allocation strategy. The corresponding strategy for these application services is determined according to the received user instructions. If it is a direct allocation strategy, each application service is allocated a corresponding number of server instances. If it is a time allocation strategy, each application service is allocated a corresponding number of server instances at the current time according to the correspondence between the current time and the number of server instances. If it is a load allocation strategy, each application service is allocated a corresponding number of server instances under the current load according to the correspondence between the load of each application service and the number of server instances. There is no need for manual real-time detection of the load of these application services, and no need for manual adjustment and allocation of the number of server instances allocated to each application service, thereby reducing the cost and workload of application service operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the prior art and the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 A flowchart of a multi-scenario service elastic scaling method based on a policy model provided in this application;

[0038] Figure 2 A flow chart of a direct allocation strategy provided for this application;

[0039] Figure 3 A flowchart of a time allocation strategy provided for this application;

[0040] Figure 4 A flow chart of a load distribution strategy provided for this application;

[0041] Figure 5 A schematic diagram of the structure of a multi-scenario service elastic scaling system based on a policy model provided in this application;

[0042] Figure 6 This is a schematic diagram of the structure of a multi-scenario service elastic scaling device based on a policy model provided in this application. DETAILED DESCRIPTION

[0043] The core of the present invention is to provide a multi-scenario service elastic scaling method and related components based on a policy model. It does not require manual real-time detection of the load of these application services, nor does it require manual adjustment of the number of server instances allocated to each application service, thereby reducing the cost and workload of application service operation and maintenance.

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0045] Please refer to Figure 1 , Figure 1 The flowchart of a multi-scenario service elastic scaling method based on a policy model provided in this application includes:

[0046] S11: Determine the server instance allocation strategy corresponding to each pre-specified application service according to the received user instruction, where the server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load allocation strategy;

[0047] S12: When the server instance allocation strategy is a direct allocation strategy, a preset number of server instances corresponding to each application service are allocated to each application service;

[0048] S13: When the server instance allocation strategy is a time allocation strategy, based on the correspondence between time and the number of server instances of each application service, a preset number of server instances corresponding to each application service at the current time is allocated to each application service;

[0049] S14: When the server instance allocation strategy is a load distribution strategy, based on the correspondence between the load of each application service and the number of server instances, a preset number of server instances corresponding to the current load of each application service is allocated to each application service.

[0050] Application services refer to applications installed on a server that can provide certain functions or services to clients, such as social applications and online shopping applications. Different application services have different load conditions. For example, near holidays, the load of online shopping applications and ticket purchasing applications will be higher than usual; the load of social applications will be higher at night than during the day. A server instance is a virtual environment (including but not limited to containers) that has computing resources such as CPU (central processing unit), memory, and network, which can be easily and dynamically created and destroyed for the normal operation of the service.

[0051] To flexibly allocate server instances for pre-specified application services in different ways, this application pre-sets server instance allocation policies. These policies include three types: a direct allocation policy that directly allocates a preset number of server instances to an application service; a time allocation policy that allocates a preset number of server instances to each application service based on time; and a load allocation policy that allocates a preset number of server instances to each application service based on its load. Users can determine the server allocation policy for these application services based on user instructions. Specifically, they can apply the same policy to all application services or define a separate policy for each application service.

[0052] When the user selects a direct allocation strategy for a pre-specified application service, please refer to Figure 2 , Figure 2 A flowchart of a direct allocation strategy provided for this application will determine the current number of server instances allocated to the application service. If the current number of server instances allocated to the application service is equal to the number corresponding to the application service under the direct allocation strategy, since the numbers are equal, no further allocation is required. If they are not equal, the application service will be allocated. When a user selects a time allocation strategy for a pre-specified application service, the strategy sets multiple time points, and the number of server instances corresponding to the application service is set at each of these time points. When it is detected that the current time point is the time point set in the strategy, the application service will be allocated the corresponding number of server instances at the current time point. When a user selects a load distribution strategy for a pre-specified application service, the strategy sets the number of server instances corresponding to the application service under different loads, and the corresponding number of server instances will be allocated according to the current load of the application service.

[0053] To summarize, by setting up a server instance allocation strategy for each preset application service, the strategy includes a direct allocation strategy, a time allocation strategy and a load allocation strategy, and determining the corresponding strategy for these application services based on the received user instructions. If it is a direct allocation strategy, each application service is allocated a corresponding number of server instances. If it is a time allocation strategy, each application service is allocated a corresponding number of server instances at the current time based on the correspondence between the current time and the number of server instances. If it is a load allocation strategy, each application service is allocated a corresponding number of server instances under the current load based on the correspondence between the load of each application service and the number of server instances. There is no need for manual real-time detection of the load of these application services, and there is no need for manual adjustment of the number of server instances allocated to each application service, thereby reducing the cost and workload of application service operation and maintenance.

[0054] Based on the above embodiment:

[0055] As a preferred embodiment, based on the correspondence between time and the number of server instances of each application service, allocating a preset number of server instances corresponding to each application service at the current time to each application service includes:

[0056] S21: Determine whether the current time exists in the preset allocation time library;

[0057] S22: If yes, allocate a preset number of server instances corresponding to each application service at the current time to each application service.

[0058] Taking into account that the load of the server instance of a specified application service may change over time, for example, the load of the application service may be higher during certain holidays or evenings, and lower during working days or daytimes, in order to allocate server instances to the application service according to the time pattern of the client accessing the application service, in this application, a preset instance allocation time library and a correspondence between time points and the allocated number of server instances are set in advance according to the time pattern of the client accessing the application service, wherein multiple time points are set as the time points for executing the allocated tasks, and each time point for executing the allocated task has its corresponding number of server instances, please refer to Figure 3 , Figure 3A flowchart of a time allocation strategy provided for this application, when the current time point is any time point for executing an assigned task, the corresponding number of server instances at the current time is allocated to the application service. For example, 9:00 is pre-set as the time point for executing the assigned task, and the number of server instances corresponding to 9:00 is 10. Therefore, when the current time point is 9:00, 10 server instances are allocated to the application service. Based on this, according to the time pattern of the client accessing the specified application service, a larger number of server instances can be allocated to the application service when the application service is at a peak access period, and a smaller number of server instances can be allocated to the application service when the application service is at a low access period.

[0059] As a preferred embodiment, the present invention further comprises:

[0060] When an update instruction is received, the preset quantity corresponding to each application service corresponding to the update instruction is changed to the preset update quantity corresponding to the application service. The update instruction includes N application services and the preset quantity corresponding to the N application services, where N is a positive integer.

[0061] In order to flexibly change the number of server instances assigned to a pre-specified application service, in this application, for some situations that can be intuitively predicted, such as predictable short-term emergencies, the number of server instances to be assigned to the application service can be updated before these situations occur. Upon receiving a user instruction indicating a direct allocation strategy, a preset number of server instances is allocated to the application service to flexibly respond to such predictable short-term emergencies. For example, if the preset update number is 10, when it is predicted that a large number of clients may access the specified application service at the same time at a certain point in time, the direct allocation strategy can be selected to allocate 10 server instances to the application service.

[0062] As a preferred embodiment, before determining the corresponding server instance allocation strategy according to the received user instruction, the method further includes:

[0063] Use Vue and iView to build a policy model that includes server instance allocation strategy;

[0064] Determine the corresponding server instance allocation strategy based on the received user instructions, including:

[0065] According to the received user instruction, a server instance allocation policy corresponding to the user instruction is determined in the policy model.

[0066] Vue is a progressive framework for building user interfaces. Its core library focuses solely on the view layer, making it both easy to get started and conveniently integrated with third-party libraries and existing projects. Furthermore, when combined with other applications and supporting libraries, Vue is fully capable of powering complex single-page applications. iView provides a rich set of input and selection components that cater to the aesthetics and usage habits of most users, making it easy to build strategic models. This makes it easy to configure server instance allocation policies.

[0067] As a preferred embodiment, based on the correspondence between the load of each application service and the number of server instances, allocating a third preset number of server instances corresponding to the current load of each application service to each application service includes:

[0068] S31: Determine the current load of each server instance in each application service;

[0069] S32: For any application service, determine the average value of the current load of all server instances in the application service;

[0070] S33: Based on the correspondence between the average value of the current loads of all server instances in each application service and the number of server instances, a third preset number of server instances corresponding to the average value of each application service is allocated to each application service.

[0071] Please refer to Figure 4 , Figure 4A flow chart of a load distribution strategy provided for this application. Considering that the load of a specified application service may fluctuate irregularly and continuously, in order to allocate server instances to the application service based on the real-time load of the application service, in this application, the average value of the computing load of each server instance can more accurately describe the load pressure of the server instance than the sum of the computing loads of all server instances, as the actual computing loads of each server instance vary. Therefore, the average value is used for judgment. Specifically, an average value interval and the number of server instances corresponding to each interval are pre-set. When the interval in which the average value lies is determined, the server instances corresponding to the interval are allocated to the application service. For example, 0% to 30% load is pre-set as the first interval, which corresponds to 3 server instances; 31% to 70% load is pre-set as the second interval, which corresponds to 5 server instances; 71% to 100% load is pre-set as the third interval, which corresponds to 10 server instances. If it is determined that the average value of the computing load of each server instance in the application service is 50%, it can be determined that the average value of the application service at this time belongs to the second interval, and 5 server instances are allocated to the application service, so as to realize the allocation of different numbers of server instances to the application service according to the real-time load of the application service.

[0072] As a preferred embodiment, determining the current load of each server instance in each application service includes:

[0073] Use metrics-server to determine the current load of each server instance in each application service.

[0074] The metrics-server can periodically obtain the current load of each server instance in each pre-specified application service. Moreover, the metrics-server can determine the load of these application services even when the servers where these application services are located are offline. In addition, the metrics-server has a low resource usage and a simple and stable implementation method.

[0075] As a preferred embodiment, based on the correspondence between the load of each application service and the number of server instances, each application service is allocated a preset number of server instances corresponding to the current load of each application service, including:

[0076] HPA is used to allocate a preset number of server instances corresponding to the current load of each application service to each application service based on the corresponding relationship between the load of each application service and the number of server instances.

[0077] HPA (Horizontal Pod Autoscaler) automatically allocates server instances to pre-specified application services based on their load. It can also automatically allocate server instances based on the CPU (central processing unit) utilization of the servers hosting these application services or other custom metrics. Furthermore, it can automatically and periodically perform server instance allocation tasks. This allows for easy allocation of server instances to each pre-specified application service under varying loads by simply setting a preset number of server instances to be allocated under varying loads.

[0078] The present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-scenario service elastic scaling method based on the policy model as described above are implemented.

[0079] For a detailed introduction to a computer-readable storage medium provided by this application, please refer to the above embodiments, and this application will not go into details here.

[0080] Please refer to Figure 5 , Figure 5 The schematic diagram of the structure of a multi-scenario service elastic scaling system based on a policy model provided in this application includes:

[0081] A strategy determination unit 11 is configured to determine a corresponding server instance allocation strategy according to a received user instruction. The server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load allocation strategy.

[0082] The first allocation unit 12 is configured to allocate a preset number of server instances corresponding to each pre-specified application service to each pre-specified application service when the server instance allocation strategy is a direct allocation strategy;

[0083] The second allocation unit 13 is configured to allocate a preset number of server instances corresponding to each application service at the current time based on the correspondence between time and the number of server instances of each application service when the server instance allocation strategy is a time allocation strategy;

[0084] The third allocation unit 14 is configured to allocate a preset number of server instances corresponding to the current load of each application service based on the correspondence between the load of each application service and the number of server instances when the server instance allocation strategy is a load distribution strategy.

[0085] For a detailed introduction to the multi-scenario service elastic scaling system based on a policy model provided by this application, please refer to the above embodiments, and this application will not go into details here.

[0086] Please refer to Figure 6 , Figure 6 A schematic diagram of the structure of a multi-scenario service elastic scaling device based on a policy model provided in this application includes:

[0087] Memory 21, for storing computer programs;

[0088] The processor 22 is configured to implement the steps of the above-mentioned multi-scenario service elastic scaling method based on a policy model when executing a computer program.

[0089] For a detailed introduction to the multi-scenario service elastic scaling device based on a policy model provided by this application, please refer to the above embodiments, and this application will not go into details here.

[0090] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0091] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0092] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-scenario service elastic scaling method based on a policy model, characterized in that: include: Determine the server instance allocation strategy corresponding to each pre-specified application service according to the received user instruction, wherein the server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load distribution strategy; When the server instance allocation strategy is the direct allocation strategy, a preset number of server instances corresponding to each application service is allocated to each application service; When the server instance allocation strategy is the time allocation strategy, based on the correspondence between time and the number of server instances of each application service, a preset number of server instances corresponding to each application service at the current time is allocated to each application service; When the server instance allocation strategy is the load distribution strategy, based on the correspondence between the load of each application service and the number of server instances, a preset number of server instances corresponding to the current load of each application service is allocated to each application service; The user instruction is used to determine the server instance allocation strategy corresponding to the application service, and different application services correspond to the same or different server instance allocation strategies; The multi-scenario service elastic scaling method based on the policy model further includes: When an update instruction is received, the preset quantity corresponding to each of the application services corresponding to the update instruction is changed to the preset update quantity corresponding to the application service, the update instruction including N application services and the preset quantity corresponding to N application services, where N is a positive integer.

2. The multi-scenario service elastic scaling method based on the policy model according to claim 1 is characterized in that: Allocating a preset number of server instances corresponding to each application service at the current time to each application service based on a correspondence between time and the number of server instances of each application service includes: Determine whether the current time exists in a preset allocation time library; If so, a preset number of server instances corresponding to each application service at the current time are allocated to each application service.

3. The multi-scenario service elastic scaling method based on a policy model according to claim 1, characterized in that: Before determining the corresponding server instance allocation strategy according to the received user instruction, the method further includes: Utilize Vue and iView to build a policy model including the server instance allocation policy; Determine the corresponding server instance allocation strategy based on the received user instructions, including: According to the received user instruction, a server instance allocation policy corresponding to the user instruction is determined in the policy model.

4. The multi-scenario service elastic scaling method based on a policy model according to any one of claims 1 to 3, characterized in that: Allocating a preset number of server instances corresponding to the current load of each application service to each application service based on the corresponding relationship between the load of each application service and the number of server instances includes: Determining the current load of each of the server instances in each of the application services; For any of the application services, determining an average value of current loads of all the server instances in the application service; Based on the correspondence between the average value of the current loads of all server instances in each of the application services and the number of server instances, a preset number of server instances corresponding to the average value of each application service is allocated to each of the application services.

5. The multi-scenario service elastic scaling method based on the policy model according to claim 4 is characterized in that: Determining the current load of each of the server instances in each of the application services includes: The current load of each of the server instances in each of the application services is determined using a metrics-server.

6. The multi-scenario service elastic scaling method based on a policy model according to claim 4, characterized in that: Allocating a preset number of server instances corresponding to the current load of each application service to each application service based on the corresponding relationship between the load of each application service and the number of server instances includes: HPA is utilized to allocate a preset number of server instances corresponding to the current load of each application service to each application service based on the corresponding relationship between the load of each application service and the number of server instances.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-scenario service elastic scaling method based on a policy model as described in any one of claims 1 to 6.

8. A multi-scenario service elastic scaling system based on a policy model, characterized by: include: a strategy determination unit, configured to determine a corresponding server instance allocation strategy according to a received user instruction, wherein the server instance allocation strategy includes a direct allocation strategy, a time allocation strategy, and a load distribution strategy; a first allocation unit, configured to allocate a preset number of server instances corresponding to each pre-specified application service to each pre-specified application service when the server instance allocation strategy is the direct allocation strategy; a second allocation unit, configured to allocate, to each application service, a preset number of server instances corresponding to each application service at a current time based on a correspondence between time and the number of server instances of each application service when the server instance allocation strategy is the time allocation strategy; a third allocation unit, configured to allocate, to each application service, a preset number of server instances corresponding to the current load of each application service based on a correspondence between the load of each application service and the number of server instances when the server instance allocation strategy is the load distribution strategy; The user instruction is used to determine the server instance allocation strategy corresponding to the application service, and different application services correspond to the same or different server instance allocation strategies; The multi-scenario service elastic scaling method based on the policy model further includes: When an update instruction is received, the preset quantity corresponding to each of the application services corresponding to the update instruction is changed to the preset update quantity corresponding to the application service, the update instruction including N application services and the preset quantity corresponding to N application services, where N is a positive integer.

9. A multi-scenario service elastic scaling device based on a policy model, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the policy model-based multi-scenario service elastic scaling method according to any one of claims 1 to 6 when executing the computer program.

Images