Calculation network host application capacity expansion and contraction scheduling system and method based on service occupation

Abstract

The invention discloses a business occupation-based computing network host application capacity expansion and contraction scheduling system and method, and relates to the technical field of resource scheduling, and the system comprises a mirror image management module which is responsible for receiving a mirror image submitted by a developer through a business platform, and carrying out the management and storage; the application management module is responsible for processing application resource requirements and capacity expansion and contraction strategies submitted by the service platform, and deploying and recycling service application examples on the computing network host; the service perception module is responsible for registering and recording instance information and perceiving an instance occupancy state; the service scheduling module is responsible for distributing nearby instances according to the IP of the client, processing occupation and release requests of the instances, and informing the service awareness module of state changes of the instances; and the capacity expansion and contraction scheduling module is responsible for periodically querying the state of the instance, triggering a capacity expansion and contraction strategy according to the occupation proportion, and adjusting the number of instances on the computing network host. The resource can be dynamically adjusted according to the actual service occupation condition, and the resource utilization rate and the service stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of resource scheduling, and in particular to a system and method for scheduling the expansion and contraction of computing network host applications based on business occupancy. Background Art

[0002] In today's era of surging digitalization, computing networks and edge computing are booming at an unprecedented speed, profoundly changing the landscape of information technology. With the continuous evolution of these two cutting-edge technologies, the services carried by computing network hosts are becoming increasingly complex and diverse.

[0003] The computing network breaks the traditional model of separation between computing resources and network resources, and flexibly allocates computing power like network bandwidth, realizing the deep integration of computing and network. Edge computing sinks computing tasks to the edge of the network, close to the data source, effectively reducing the delay of data transmission and improving real-time response capabilities. In this context, the computing network host, as the core carrier, must not only cope with the processing needs of massive data, but also meet the strict requirements of different business scenarios in terms of real-time and reliability.

[0004] However, the traditional management and operation support system is unable to cope with these complex and ever-changing businesses. Traditionally, the system mainly relies on resource indicators, such as CPU usage and memory usage, to schedule stateless applications. This scheduling method may have played a certain role in relatively simple business environments in the past. However, at present, business scenarios have become extremely complex, and the disadvantages of this incomplete scheduling solution have gradually become apparent.

[0005] Taking business occupancy as an example, many modern businesses use long connection technology. Long connection means that the client and the server will maintain a continuous connection state so that data can be exchanged at any time. In this case, the business instance will be occupied by the client for a long time. However, traditional scheduling schemes fail to fully consider this key factor. When the system is scaling down, it is very likely to mistakenly delete business instances that are busy, maintain a long connection with the client, and process data because it is judged only based on resource indicators. This will undoubtedly lead to the interruption of ongoing business, seriously affect the user experience, and may even cause economic losses and reputation damage to the enterprise.

[0006] In summary, with the continuous development of computing networks and edge computing, the scheduling scheme of traditional management and operation support systems based on resource indicators without considering business occupancy has become difficult to adapt to the increasingly complex business carrying needs of computing network hosts, and urgently needs to be improved and optimized to ensure the stable operation of the business and a good user experience. Summary of the invention

[0007] Traditional management and operation support systems usually use resource indicators (such as CPU and memory) to schedule stateless applications. This traditional scheduling solution does not take into account business occupancy (such as when a business uses long connection technology, it will be occupied by the client). As a result, there is a problem that busy business instances may be deleted when capacity is reduced. Based on this, the present invention provides a computing network host application expansion and contraction scheduling system and method based on business occupancy.

[0008] In the first aspect, the present invention provides a computing network host application expansion and contraction scheduling system based on business occupancy, and the technical solution adopted to solve the above technical problems is as follows:

[0009] A computing network host application expansion and contraction scheduling system based on business occupancy, comprising:

[0010] The image management module is responsible for receiving, managing and storing images submitted by developers through the business platform;

[0011] The application management module is responsible for processing the application resource requirements and expansion and contraction strategies submitted by the business platform, and deploying and retrieving business application instances on the computing network host;

[0012] The business perception module is responsible for registering and recording business application instance information, sensing the occupancy status of business application instances, and providing data basis for capacity expansion and contraction scheduling;

[0013] The service scheduling module is responsible for allocating the nearest service application instance based on the client IP address, processing the occupation and release requests of service application instances, and notifying the service perception module of the status changes of service application instances;

[0014] The capacity expansion and contraction scheduling module is responsible for periodically querying the status of business application instances, triggering the capacity expansion and contraction strategy based on the occupancy ratio, and adjusting the number of business application instances on the computing network host.

[0015] Optionally, the image management module involved specifically includes:

[0016] Image interface unit, used for uploading images to the business platform;

[0017] The image management unit is used to perform integrity verification and version management on the images uploaded by the business platform to ensure the availability and accuracy of the images and provide basic image resources for subsequent business application deployment;

[0018] The image storage unit is used to store the image after integrity verification and version management in a designated storage area, and to record and manage the storage path or storage location of the image.

[0019] Optionally, the application management modules involved specifically include:

[0020] The receiving and selecting unit is used to receive the application resource requirements and expansion and contraction strategies of the business platform, select the optimal nodes and resources on the computing network host according to the resource requirements, start the business application instance, and also include the retry strategy in case of deployment failure;

[0021] The receiving and sending unit is used to receive the resource recovery application from the business platform after the business life cycle ends, send instructions to the computing network host to delete the business application instance, release related resources, and handle cleanup operations when resources are not completely released.

[0022] Further optionally, the business perception module involved specifically includes:

[0023] The database establishment unit is used to receive the business application instance information registered by the business platform, establish the instance information database, including the instance number, the computing network host node, resource usage, running time and historical usage data, and use transaction processing to ensure data consistency;

[0024] The update statistics unit is used to update the instance occupancy status in real time when the business scheduling module allocates or releases business application instances, count the total number of instances, the number of occupied instances, the instance running time, historical usage data, and calculate the occupancy ratio for query by the scaling scheduling module.

[0025] Further optionally, the business scheduling module involved specifically includes:

[0026] The selection allocation unit is used to select the available business application instance closest to the client and with the best performance for allocation when the business client applies for an idle business application instance, based on the client IP address, combined with the network topology, the location information of the computing network host node and the performance index of the computing network host, and notify the update statistics unit of the allocation result. The update statistics unit updates the occupation status of the business application instance to occupied, and handles the abnormal situation when the allocation fails;

[0027] The notification receiving unit is used to receive a release request when the business client is closed and applies for releasing the instance, and to notify the statistics updating unit to update the occupation state of the corresponding business application instance to release.

[0028] Further optionally, the expansion and contraction scheduling module involved specifically includes:

[0029] A timing setting unit, used to set a periodic query time interval;

[0030] A timing execution unit, used to query the business perception module for the occupancy of application instances at regular intervals according to a set periodic query time interval, including the total number of instances, the number of occupied instances and the occupancy ratio;

[0031] A threshold setting unit, used to set a capacity expansion threshold and a capacity reduction threshold;

[0032] The expansion judgment unit is used to calculate the number of business application instances to be added according to the target replica number formula when the occupancy ratio exceeds the set expansion threshold, and then notify the application management module to deploy the corresponding number of new business application instances on the computing and networking host. After the new business application instances are started, it ensures that the information of the new business application instances is registered and perceived by the service awareness module;

[0033] The contraction judgment unit is used to trigger the contraction strategy when the occupancy ratio is less than the set contraction threshold and continuously meets the conditions within the time window. During contraction, it first filters out the unoccupied business application instances, and then notifies the application management module to delete the unoccupied business application instances, and at the same time updates the instance information and occupancy status in the service awareness module.

[0034] Preferably, the target replica number formula is as follows:

[0035] Target replica number = ceiling(current replica number * (current occupancy / target occupancy)).

[0036] In a second aspect, the present invention provides a computing and networking host application scaling scheduling method based on service occupancy. The technical solution adopted to solve the above technical problems is as follows:

[0037] A computing and networking host application scaling scheduling method based on service occupancy, which is based on the system described in the first aspect, includes the following implementation steps:

[0038] S1. The developer submits an image to the image management module through the service platform. Subsequently, the service platform submits application resource requirements and scaling policies, and the application management module deploys business application instances on the computing and networking host accordingly;

[0039] After the business application instances are started, they report information to the service platform, and the service platform registers the business application instance information with the service awareness module;

[0040] When the business client applies for an idle instance, the service scheduling module allocates a business application instance nearby according to the client IP and notifies the service awareness module that the business application instance is occupied;

[0041] When the business client closes and applies to release the instance, the service scheduling module notifies the service awareness module that the business application instance is released, so that the service awareness module can master the total number of instances and the occupancy situation, providing data support for scheduling;

[0042] S3. The scaling scheduling module periodically queries the status of business application instances in the service awareness module. When the occupancy ratio exceeds the set expansion threshold, it triggers the expansion strategy, calculates the target replica number according to the formula and increases the number of business application instances. After the new business application instances are started, it ensures that the information of the new business application instances is registered and perceived by the service awareness module;

[0043] S4. The capacity expansion and contraction scheduling module periodically queries the business application instance status of the business perception module. When the occupancy ratio is less than the set capacity reduction threshold and continues to meet the conditions within the time window, the capacity reduction strategy is triggered to reduce the number of business application instances. When reducing capacity, the occupied instances are screened out to ensure that the business is not interrupted.

[0044] S5. After the service life cycle ends, the service platform submits a resource recovery application to the application management module, and the application management module sends a deletion command to the computing network host to complete the closed loop of the entire process.

[0045] The system and method for scheduling the expansion and contraction of computing network host applications based on business occupancy of the present invention have the following beneficial effects compared with the prior art:

[0046] 1. The present invention records the business instance status data by sensing the client's occupancy of the business instance. When the business instance occupancy is greater than the threshold, expansion scheduling is performed, and when the business instance occupancy is less than the threshold, reduction scheduling is performed. At the same time, the reduction scheduling will exclude the occupied instance to prevent business interruption, thus realizing the expansion and reduction scheduling of the computing network host application based on business occupancy, and being able to dynamically adjust resources according to the actual business occupancy, thereby improving resource utilization and business stability;

[0047] 2. The present invention standardizes and normalizes the process of scaling up and down the computing network host application based on business occupancy, thereby solving the problem that the traditional management and operation support system only supports the use of resource indicators to scale up and down stateless applications, and cannot fully consider the business occupancy of the application, thus causing business interruptions. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Attached Figure 1 is a system architecture flow chart of Embodiment 1 of the present invention;

[0049] Attached Figure 2 It is a system module connection block diagram of the first embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the technical solution, the technical problem solved and the technical effect of the present invention more clearly understood, the technical solution of the present invention is clearly and completely described below in conjunction with specific embodiments.

[0051] Embodiment 1:

[0052] Combined with Figure 1 , 2 This embodiment proposes a computing network host application expansion and contraction scheduling system based on business occupancy, which includes:

[0053] The image management module is responsible for receiving, managing and storing images submitted by developers through the business platform;

[0054] The application management module is responsible for processing the application resource requirements and expansion and contraction strategies submitted by the business platform, and deploying and retrieving business application instances on the computing network host;

[0055] The business perception module is responsible for registering and recording business application instance information, sensing the occupancy status of business application instances, and providing data basis for capacity expansion and contraction scheduling;

[0056] The service scheduling module is responsible for allocating the nearest service application instance based on the client IP address, processing the occupation and release requests of service application instances, and notifying the service perception module of the status changes of service application instances;

[0057] The capacity expansion and contraction scheduling module is responsible for periodically querying the status of business application instances, triggering the capacity expansion and contraction strategy based on the occupancy ratio, and adjusting the number of business application instances on the computing network host.

[0058] In this embodiment, the image management module involved specifically includes:

[0059] Image interface unit, used for uploading images to the business platform;

[0060] The image management unit is used to perform integrity verification and version management on the images uploaded by the business platform to ensure the availability and accuracy of the images and provide basic image resources for subsequent business application deployment;

[0061] The image storage unit is used to store the image after integrity verification and version management in a designated storage area, and to record and manage the storage path or storage location of the image.

[0062] In this embodiment, the application management module involved specifically includes:

[0063] The receiving and selecting unit is used to receive the application resource requirements and expansion and contraction strategies of the business platform, select the optimal nodes and resources on the computing network host according to the resource requirements, start the business application instance, and also include the retry strategy in case of deployment failure;

[0064] The receiving and sending unit is used to receive the resource recovery application from the business platform after the business life cycle ends, send instructions to the computing network host to delete the business application instance, release related resources, and handle cleanup operations when resources are not completely released.

[0065] In this embodiment, the service perception module involved specifically includes:

[0066] The database establishment unit is used to receive the business application instance information registered by the business platform, establish the instance information database, including the instance number, the computing network host node, resource usage, running time and historical usage data, and use transaction processing to ensure data consistency;

[0067] The update statistics unit is used to update the instance occupancy status in real time when the business scheduling module allocates or releases business application instances, count the total number of instances, the number of occupied instances, the instance running time, historical usage data, and calculate the occupancy ratio for query by the scaling scheduling module.

[0068] In this embodiment, the service scheduling module involved specifically includes:

[0069] The selection allocation unit is used to select the available business application instance closest to the client and with the best performance for allocation when the business client applies for an idle business application instance, based on the client IP address, combined with the network topology, the location information of the computing network host node and the performance index of the computing network host, and notify the update statistics unit of the allocation result. The update statistics unit updates the occupation status of the business application instance to occupied, and handles the abnormal situation when the allocation fails;

[0070] The notification receiving unit is used to receive a release request when the business client is closed and applies for releasing the instance, and to notify the statistics updating unit to update the occupation state of the corresponding business application instance to release.

[0071] In this embodiment, the expansion and contraction scheduling module involved specifically includes:

[0072] A timing setting unit, used to set a periodic query time interval;

[0073] A timing execution unit, used to query the business perception module regularly for the occupancy status of application instances according to a set periodic query time interval, including the total number of instances, the number of occupied instances and the occupancy ratio;

[0074] A threshold setting unit, used to set a capacity expansion threshold and a capacity reduction threshold;

[0075] The capacity expansion judgment unit is used to calculate the number of business application instances that need to be added according to the target copy number formula when the occupancy ratio exceeds the set capacity expansion threshold, and then notify the application management module to deploy a corresponding number of new business application instances on the computing network host, and after the new business application instance is started, ensure that the information of the new business application instance is registered and perceived by the business perception module; wherein the formula for the target copy number involved is as follows: target copy number = round up (current copy number * (current occupied number / target occupied number));

[0076] The capacity reduction judgment unit is used to trigger the capacity reduction strategy when the occupancy ratio is less than the set capacity reduction threshold and continues to meet the conditions within the time window; when reducing capacity, first filter out the unoccupied business application instances, then notify the application management module to delete the unoccupied business application instances, and update the instance information and occupancy status in the business perception module.

[0077] Embodiment 2:

[0078] Reference Figure 1 , 2 Based on the method of the first embodiment, this embodiment proposes a computing network host application expansion and contraction scheduling system based on business occupancy, including the following implementation steps:

[0079] S1. The developer submits the image to the image management module through the business platform. The business platform then submits the application resource requirements and scaling strategies. The application management module deploys the business application instance on the computing network host accordingly.

[0080] Specifically, the developer submits the image to the image management module through the business platform. The image interface unit in the image management module receives the image. The image management unit performs integrity verification, version management and other operations on the image and then stores it in the designated storage area, and records and manages the image storage location. Subsequently, the business platform submits application resource requirements and scaling strategies. The receiving and selecting unit in the application management module selects the optimal node and resources on the computing network host based on the resource requirements and the resource load of the computing network host (including CPU, memory, storage usage, network bandwidth and latency, etc.), and starts the business application instance.

[0081] S2. After the business application instance is started, it reports information to the business platform, and the business platform registers the business application instance information with the business perception module;

[0082] When a business client applies for an idle instance, the business scheduling module allocates the nearest business application instance based on the client IP address and notifies the business perception module that the business application instance is occupied.

[0083] When the business client is closed and applies to release the instance, the business scheduling module notifies the business perception module that the business application instance is released, so that the business perception module can grasp the total number of instances and occupancy status, providing data support for scheduling.

[0084] Specifically, after the business application instance is started, it reports information to the business platform. The business platform registers the business application instance information with the database establishment unit in the business perception module, including the instance number, the computing network host node, resource usage, running time and historical usage data, etc., and establishes an instance information database based on MySQL or Redis, using transaction processing to ensure data consistency.

[0085] When a service client applies for an idle instance, the selection and allocation unit in the service scheduling module selects the available service application instance closest to the client and with the best performance for allocation based on the client IP address, combined with the network topology, the location information of the computing network host node and the performance indicators of the computing network host (such as the current load and historical performance data), and notifies the update statistics unit in the service perception module of the allocation result, which updates the occupancy status of the service application instance to occupied;

[0086] When the business client closes and applies to release the instance, the receiving notification unit in the business scheduling module receives the release request and notifies the updating statistics unit to update the occupancy status of the corresponding business application instance to released, so that the business perception module can grasp the total number of instances, the number of occupied instances, the instance running time, historical usage data and occupancy ratio, etc., to provide comprehensive data support for scheduling.

[0087] S3. The capacity expansion and contraction scheduling module periodically queries the business application instance status of the business perception module. When the occupancy ratio exceeds the set capacity expansion threshold, the capacity expansion strategy is triggered, the target number of replicas is calculated according to the formula, and the number of business application instances is increased. After the new business application instance is started, ensure that the information of the new business application instance is registered and perceived by the business perception module.

[0088] Specifically, the timing setting unit in the expansion and contraction scheduling module sets a periodic query time interval, and the timing execution unit periodically queries the business perception module for the occupancy of the business application instances according to the set periodic query time interval, including the total number of instances, the number of occupied instances, and the occupancy ratio;

[0089] When the occupancy ratio exceeds the set expansion threshold, the expansion strategy is triggered, and the target number of copies is calculated according to the formula (target number of copies = round up (current number of copies * (current occupancy number / target occupancy number))). For example, the current number of copies is 10 and the target occupancy is 80%. If 9 instances are detected to be occupied, that is, the occupancy is 90%, which exceeds the threshold of 80%, the expansion strategy is triggered, and the target number of copies is 12, that is, ceil (10 * (90% / 80%)). Then the judgment and expansion unit in the expansion and contraction scheduling module notifies the application management module, and the receiving and selection unit of the application management module deploys the corresponding number of new business application instances on the computing network host. After the new business application instance is started, ensure that the information of the new business application instance is registered by the database establishment unit in the business perception module and perceived by the update statistics unit.

[0090] S4. The scaling scheduling module periodically queries the business application instance status of the business perception module. When the occupancy ratio is less than the set scaling threshold and continues to meet the conditions within the time window, the scaling strategy is triggered to reduce the number of business application instances. When scaling down, the occupied instances are screened out to ensure that the business is not interrupted.

[0091] Specifically, the scaling scheduling module continues to periodically query the business application instance status of the business perception module. When the occupancy ratio is less than the set scaling threshold and continues to meet the conditions within a time window (such as 5 minutes), the scaling strategy is triggered to reduce the number of business application instances. When scaling down, the occupied instances are screened out by filtering out the occupied instances according to the screening method marked by the occupancy status to ensure that the business is not interrupted. Then the scaling judgment unit in the scaling scheduling module notifies the receiving and sending unit in the application management module to delete the unoccupied business application instances, and at the same time updates the instance information and occupancy status in the business perception module.

[0092] S5. After the service life cycle ends, the service platform submits a resource recovery application to the application management module, and the application management module sends a deletion command to the computing network host to complete the closed loop of the entire process.

[0093] The above scheme explains in more detail the specific operations and interaction processes of each module and its sub-units in each step, making the entire computing network host application expansion and contraction scheduling process based on business occupancy clearer, complete and easy to understand. It is also closely integrated with the previous detailed description of the system modules to form a coherent whole.

[0094] In summary, the system and method for scaling up and down of computing network host applications based on business occupancy of the present invention can realize scaling up and down of computing network host applications based on business occupancy, dynamically adjust resources according to the actual occupancy of the business, improve resource utilization and business stability, and solve the problem that traditional management and operation support systems only support the use of resource indicators to scale up and down stateless applications, and cannot fully consider the occupancy of the business on the application, thereby causing business interruptions.

[0095] The above specific examples are used to explain the principles and implementation methods of the present invention in detail. These examples are only used to help understand the core technical content of the present invention. Based on the above specific embodiments of the present invention, any improvements and modifications made by technicians in this technical field without departing from the principles of the present invention should fall within the scope of patent protection of the present invention.

Claims

1. A computing network host application expansion and contraction scheduling system based on business occupancy, characterized in that: It includes: The image management module is responsible for receiving, managing and storing images submitted by developers through the business platform; The application management module is responsible for processing the application resource requirements and expansion and contraction strategies submitted by the business platform, and deploying and retrieving business application instances on the computing network host; The business perception module is responsible for registering and recording business application instance information, sensing the occupancy status of business application instances, and providing data basis for capacity expansion and contraction scheduling; The service scheduling module is responsible for allocating the nearest service application instance based on the client IP address, processing the occupation and release requests of service application instances, and notifying the service perception module of the status changes of service application instances; The capacity expansion and contraction scheduling module is responsible for periodically querying the status of business application instances, triggering the capacity expansion and contraction strategy based on the occupancy ratio, and adjusting the number of business application instances on the computing network host.

2. According to claim 1, a method for scheduling the expansion and contraction of computing network host applications based on business occupancy, characterized in that: The image management module specifically includes: Image interface unit, used for uploading images to the business platform; The image management unit is used to perform integrity verification and version management on the images uploaded by the business platform to ensure the availability and accuracy of the images and provide basic image resources for subsequent business application deployment; The image storage unit is used to store the image after integrity verification and version management in a designated storage area, and to record and manage the storage path or storage location of the image.

3. A method for scheduling expansion and contraction of computing network host applications based on business occupancy according to claim 2, characterized in that: The application management module specifically includes: The receiving and selecting unit is used to receive the application resource requirements and expansion and contraction strategies of the business platform, select the optimal nodes and resources on the computing network host according to the resource requirements, start the business application instance, and also include the retry strategy in case of deployment failure; The receiving and sending unit is used to receive the resource recovery application from the business platform after the business life cycle ends, send instructions to the computing network host to delete the business application instance, release related resources, and handle cleanup operations when resources are not completely released.

4. A method for scheduling expansion and contraction of computing network host applications based on business occupancy according to claim 3, characterized in that: The business perception module specifically includes: The database establishment unit is used to receive the business application instance information registered by the business platform, establish the instance information database, including the instance number, the computing network host node, resource usage, running time and historical usage data, and use transaction processing to ensure data consistency; The update statistics unit is used to update the instance occupancy status in real time when the business scheduling module allocates or releases business application instances, count the total number of instances, the number of occupied instances, the instance running time, historical usage data, and calculate the occupancy ratio for query by the scaling scheduling module.

5. A method for scheduling expansion and contraction of computing network host applications based on business occupancy according to claim 4, characterized in that: The service scheduling module specifically includes: The selection allocation unit is used to select the available business application instance closest to the client and with the best performance for allocation when the business client applies for an idle business application instance, based on the client IP address, combined with the network topology, the location information of the computing network host node and the performance index of the computing network host, and notify the update statistics unit of the allocation result. The update statistics unit updates the occupation status of the business application instance to occupied, and handles the abnormal situation when the allocation fails; The notification receiving unit is used to receive a release request when the business client is closed and applies for releasing the instance, and to notify the statistics updating unit to update the occupation state of the corresponding business application instance to release.

6. A method for scheduling expansion and contraction of computing network host applications based on business occupancy according to claim 5, characterized in that: The expansion and contraction scheduling module specifically includes: A timing setting unit, used to set a periodic query time interval; A timing execution unit, used to query the business perception module regularly for the occupancy status of application instances according to a set periodic query time interval, including the total number of instances, the number of occupied instances and the occupancy ratio; A threshold setting unit, used to set a capacity expansion threshold and a capacity reduction threshold; The capacity expansion judgment unit is used to calculate the number of business application instances that need to be added according to the target copy number formula when the occupancy ratio exceeds the set capacity expansion threshold, and then notify the application management module to deploy the corresponding number of new business application instances on the computing network host, and after the new business application instance is started, ensure that the information of the new business application instance is registered and perceived by the business perception module; The capacity reduction judgment unit is used to trigger the capacity reduction strategy when the occupancy ratio is less than the set capacity reduction threshold and continues to meet the conditions within the time window; when reducing capacity, first filter out the unoccupied business application instances, then notify the application management module to delete the unoccupied business application instances, and update the instance information and occupancy status in the business perception module.

7. A method for scheduling expansion and contraction of computing network host applications based on business occupancy according to claim 6, characterized in that: The formula for the target number of copies is as follows: Target number of replicas = round up (current number of replicas * (current number of occupancy / target number of occupancy)).

8. A method for scheduling the expansion and contraction of computing network host applications based on business occupancy, characterized in that: It is based on the system as claimed in claim 1, and includes the following implementation steps: S1. The developer submits the image to the image management module through the business platform. The business platform then submits the application resource requirements and scaling strategies. The application management module deploys the business application instance on the computing network host accordingly. S2. After the business application instance is started, it reports information to the business platform, and the business platform registers the business application instance information with the business perception module; When a business client applies for an idle instance, the business scheduling module allocates the nearest business application instance based on the client IP address and notifies the business perception module that the business application instance is occupied. When the business client is closed and applies to release the instance, the business scheduling module notifies the business perception module that the business application instance is released, so that the business perception module can grasp the total number of instances and occupancy status, providing data support for scheduling; S3. The capacity expansion and contraction scheduling module periodically queries the business application instance status of the business perception module. When the occupancy ratio exceeds the set capacity expansion threshold, the capacity expansion strategy is triggered, the target number of replicas is calculated according to the formula, and the number of business application instances is increased. After the new business application instance is started, ensure that the information of the new business application instance is registered and perceived by the business perception module; S4. The capacity expansion and contraction scheduling module periodically queries the business application instance status of the business perception module. When the occupancy ratio is less than the set capacity reduction threshold and continues to meet the conditions within the time window, the capacity reduction strategy is triggered to reduce the number of business application instances. When reducing capacity, the occupied instances are screened out to ensure that the business is not interrupted. S5. After the service life cycle ends, the service platform submits a resource recovery application to the application management module, and the application management module sends a deletion command to the computing network host to complete the closed loop of the entire process.

Images