Method and system for calling service capability in application software

Abstract

The invention provides a service capability calling method and system in application software. The method comprises the following steps of: monitoring real-time temperature data and load state data of each available service node in a server cluster in real time, and combining a preset load balancing strategy and a virtual machine dynamic migration technology to ensure the stable operation of the server cluster in a high-concurrency scene. And when it is detected that the real-time temperature of the available server node exceeds the preset temperature threshold value or the capacity of the server node reaches the capacity upper limit, the service capability of the available server node is migrated to the standby node, and service interruption is avoided. Meanwhile, the real-time temperature data, the load state data and the execution frequency of the migration operation are subjected to correlation analysis, and a load balancing strategy and a migration trigger threshold value are dynamically optimized, so that service continuity is maintained, and the temperature fluctuation range of hardware resources is controlled. According to the technical scheme provided by the invention, continuous response of service requests and temperature stability of hardware resources in a high-concurrency scene can be realized.

Description

Technical Field

[0001] This application relates to the technical field of application software development, and in particular, to a method and system for invoking service capabilities in application software. Background Art

[0002] In a high-concurrency cloud computing environment, a server cluster needs to process a large number of service requests simultaneously, resulting in a sharp increase in the load of hardware resources and a rapid rise in temperature. To ensure system stability and service continuity, a technical solution that can monitor the hardware status in real time and dynamically adjust resource allocation is required to cope with instantaneous request peaks and hardware overheating problems.

[0003] The load balancing strategies provided by existing solutions mainly rely on load status data to make resource allocation and migration decisions. The load balancing strategy can ensure the rationality of resource allocation to a certain extent. However, in the case of hardware overheating, high-load server nodes still provide service capabilities in application software, which may cause certain problems such as security and service efficiency. Moreover, the existing solutions adopt a passive heat dissipation method for the hardware overheating problem, and the combination of this method and the load balancing strategy is difficult to effectively constrain the temperature fluctuation range of hardware resources while ensuring service continuity. Summary of the Invention

[0004] Embodiments of this application provide a method and system for invoking service capabilities in application software to solve the problems in the prior art that when there is hardware overheating, relying on passive heat dissipation and load balancing makes it difficult to ensure service continuity and effectively control the hardware temperature fluctuation at the same time, resulting in security and service efficiency problems.

[0005] In a first aspect, embodiments of this application provide a method for invoking service capabilities in application software, including: Obtain the real-time temperature data and load status data of each available server node in a server cluster deploying application software, and a preset load balancing strategy, where the load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate; When it is detected that the real-time temperature of any one of the available server nodes exceeds a preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches a preset capacity upper limit, combine the virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, where the standby server node is another available server node in the server cluster; Perform correlation analysis on the real-time temperature data, the load status data, and the execution frequency of the migration operation of all available server nodes to obtain a correlation analysis result; Based on the correlation analysis results, dynamically optimize the node screening rules in the preset load balancing policy and the triggering threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing policy can maintain the continuous response of service requests in high-concurrency scenarios and restrict the temperature fluctuation range of the hardware resources of the server cluster.

[0006] Optionally, in combination with the virtual machine dynamic migration technology, migrate the service capabilities running on the available server nodes to the standby server nodes, where the standby server nodes are another available server nodes in the server cluster, including: Determine the priority of the service capabilities running on the available server nodes, where the priority is dynamically allocated based on the resource requirement characteristics and business types of the service capabilities, and the resource requirement characteristics include the number of CPU cores required, the memory capacity required, and the network bandwidth occupancy rate required; Based on the real-time temperature data and the load status data of all available server nodes in the server cluster, screen the candidate server nodes with a temperature lower than the preset temperature threshold and a load not reaching the capacity limit; When the number of candidate server nodes is multiple, sort the candidate server nodes according to the remaining resource capacity to generate a candidate server node list; According to the priority of the service capabilities and the resource requirement characteristics, match the candidate server node with the most suitable remaining resource capacity and the smallest difference from the candidate server node list as the standby server node; Before the migration execution, perform a secondary verification on the real-time temperature data and the load status data of the standby server node to ensure that the temperature fluctuation of the standby server node during the migration process is within the temperature fluctuation range, and the load status data is within the preset load status safety range; Migrate the service capabilities to the standby server node through the virtual machine dynamic migration technology. The migration process of the service capabilities includes migrating each service capability in order of priority, and updating the load status data and temperature data of the standby server node in real time after each migration is completed.

[0007] Optionally, when migrating the service capabilities to the standby server node through the virtual machine dynamic migration technology, the migration process of the service capabilities includes migrating each service capability in order of priority, including: According to the priority of each service capability, arrange all the service capabilities running on the available server nodes in descending order to generate a migration task queue; Select the service capabilities to be migrated from the migration task queue in sequence, and calculate whether the remaining resource capacity of the standby server node meets the resource requirements of the service capabilities to be migrated based on the resource requirement characteristics of the service capabilities to be migrated; If the resource requirements are met, start the virtual machine dynamic migration technology to migrate the service capabilities to be migrated from the available server node to the standby server node; Or, if the resource requirements are not met, suspend the current migration task, and re-screen the standby server nodes from the candidate server node list to update the migration targets of the unmigrated tasks in the migration task queue.

[0008] Optionally, if the resource requirements are met, starting the virtual machine dynamic migration technology to migrate the service capabilities to be migrated from the available server node to the standby server node includes: Generate migration configuration parameters according to the resource requirement characteristics of the service capabilities to be migrated, and pre-allocate the resources corresponding to the migration configuration parameters on the standby server node; Start the virtual machine dynamic migration technology to synchronize the memory state, central processing unit state, and network connection information of the service capabilities to be migrated from the available server node to the standby server node; During the synchronization process, monitor the standby server node in real time. If it is detected that the resource occupancy rate of the standby server node exceeds the preset occupancy rate threshold or the temperature exceeds the preset temperature threshold, suspend the synchronization process and trigger the resource reallocation mechanism to re-evaluate and adjust the pre-allocated resources; After the synchronization is completed, switch the running state of the service capabilities to be migrated from the available server node to the standby server node, and release the corresponding resources on the available server node; After the switching is completed, update the load status data and temperature data of the standby server node in real time. The temperature data is periodically collected at a preset sampling frequency by the temperature sensor deployed on the standby server node.

[0009] Optionally, based on the correlation analysis result, dynamically optimize the node selection rules in the preset load balancing strategy and the trigger threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing strategy can maintain the continuous response of service requests in a high-concurrency scenario and constrain the temperature fluctuation range of the server cluster hardware resources, including: Extract the coupling characteristics of temperature fluctuation and load surge from the correlation analysis result, and optimize the node selection rules in the preset load balancing strategy based on the coupling characteristics; According to the execution frequency of the migration operation in the association analysis result, in combination with the temperature change curve and the load change curve of the available server node and the standby server node respectively during the migration process, calculate the attenuation function parameter, and dynamically adjust the sensitivity of the migration trigger threshold through the attenuation function parameter to dynamically optimize the migration trigger threshold; Based on the optimized node screening rule and migration trigger threshold, obtain the optimized preset load balancing policy, and apply the optimized preset load balancing policy to the server cluster to form a closed-loop feedback mechanism.

[0010] Optionally, generating migration configuration parameters according to the resource demand characteristics of the service capabilities to be migrated includes: Generate initial pre-allocation parameters according to the required number of Central Processing Unit (CPU) cores, required memory capacity, and required network bandwidth occupancy rate; Dynamically adjust the initial pre-allocation parameters in combination with the real-time temperature data and load status data of the standby server node to obtain the adjusted pre-allocation parameters; Based on the migration policy constraints, set the corresponding fault tolerance and efficiency parameters, and combine the fault tolerance, the efficiency parameters with the adjusted pre-allocation parameters to obtain the migration configuration parameters.

[0011] Optionally, obtaining the optimized preset load balancing policy based on the optimized node screening rule and migration trigger threshold includes: Add an edge distance factor corresponding to a specified scenario to the optimized node screening rule, where the specified scenarios include high-concurrency scenarios and preset high-temperature sensitive scenarios; Based on the node screening rule, migration trigger threshold, and exception fusing mechanism with the added edge distance factor, use the federated learning model to generate the optimized preset load balancing policy.

[0012] In a second aspect, an embodiment of the present application provides a service capability invocation system in an application software, including: An acquisition module, configured to acquire real-time temperature data and load status data of each available server node in a server cluster where the application software is deployed, and a preset load balancing policy, where the load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate; A detection and migration module, configured to, when detecting that the real-time temperature of any one of the available server nodes exceeds a preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches a preset capacity upper limit, combine with the virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, where the standby server node is another available server node in the server cluster; An association analysis module, configured to perform an association analysis on the real-time temperature data, the load status data, and the execution frequency of the migration operation of all available server nodes to obtain an association analysis result; An optimization module, configured to dynamically optimize the node screening rule in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology based on the association analysis result, so that the optimized preset load balancing policy can maintain continuous response to service requests in a high-concurrency scenario and constrain the temperature fluctuation range of the hardware resources of the server cluster.

[0013] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a service capability calling method in an application software as described in any one of the first aspects.

[0014] In a fourth aspect, an embodiment of the present application provides a computer storage medium, storing a computer program, where when the computer program is executed by a computer, it implements a service capability calling method in an application software as described in any one of the first aspects.

[0015] In an embodiment of the present application, a service capability calling method in an application software is provided. The method includes: obtaining real-time temperature data and load status data of each available server node in a server cluster where the application software is deployed, and a preset load balancing policy, where the load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate; when detecting that the real-time temperature of any one of the available server nodes exceeds a preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches a preset capacity upper limit, combine with the virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, where the standby server node is another available server node in the server cluster; performing an association analysis on the real-time temperature data, the load status data, and the execution frequency of the migration operation of all available server nodes to obtain an association analysis result; dynamically optimizing the node screening rule in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology based on the association analysis result, so that the optimized preset load balancing policy can maintain continuous response to service requests in a high-concurrency scenario and constrain the temperature fluctuation range of the hardware resources of the server cluster.

[0016] In the embodiments of the present application, by obtaining the temperature data and load status data of each node in the server cluster in real time, before detecting that an available server node is about to reach the performance bottleneck (temperature exceeding the limit or resources being fully loaded), the virtual machine dynamic migration mechanism is automatically triggered to seamlessly transfer service requests to standby nodes. This mechanism can ensure that under the impact of high-concurrency traffic, effects such as seamless service switching, improved failure avoidance rate, and improved service level agreement achievement rate are achieved, thereby ensuring service continuity in high-concurrency scenarios. By dynamically adjusting the migration strategy through the correlation analysis of temperature and load, effects such as suppressing temperature fluctuations, optimizing heat dissipation costs, and extending the hardware life are achieved, thereby realizing intelligent temperature control of hardware resources. Further, the service capacity is migrated to standby server nodes in combination with the virtual machine dynamic migration technology. During the migration process, standby nodes are dynamically selected based on the priority of service capacity and resource demand characteristics, and the temperature and load status of standby nodes are verified twice before migration. After migration, the load status data and temperature data of standby nodes are updated in real time, and the load balancing strategy and migration trigger threshold are optimized through correlation analysis, so as to maintain the continuous response of service requests in high-concurrency scenarios and restrict the temperature fluctuation range of hardware resources. Further, through dynamic migration and load balancing optimization, continuous response of service requests and temperature stability of hardware resources in high-concurrency scenarios are achieved.

[0017] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a flowchart of a method for invoking service capabilities in an application software provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a system for invoking service capabilities in an application software provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. Detailed Embodiments

[0020] In order to enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application.

[0021] In some processes described in the specification, claims, and the above-mentioned drawings of this application, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The operation numbers such as 11, 12, etc. are only used to distinguish different operations, and the numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0022] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application.

[0023] The prior art relies on passive cooling and load balancing when the hardware overheats, and it is difficult to ensure service continuity and effectively control the temperature fluctuation of the hardware at the same time, resulting in safety and service efficiency problems. To solve the above problems, the embodiments of the present application provide a method for invoking service capabilities in an application software. The specific concept includes: real-time collecting the real-time temperature data and load status data (including the number of service requests and resource occupancy rate) of each available server node in the server cluster through sensors or monitoring software, and performing a preliminary analysis in combination with a preset load balancing strategy. When it is detected that the temperature of a certain node exceeds the preset temperature threshold or the resource occupancy rate reaches the capacity limit, the virtual machine dynamic migration technology is triggered to migrate the service capabilities of the available server node to the standby node in the server cluster to ensure service continuity. Correlation analysis is performed on the temperature, load data, and migration frequency of all available server nodes to form an analysis result. Based on the analysis result, the node screening rules and migration trigger thresholds in the load balancing strategy are dynamically optimized to maintain continuous response of the service in a high-concurrency scenario, while controlling the temperature fluctuation range of the hardware resources and improving system stability and resource utilization.

[0024] Figure 1 The flowchart of a method for invoking service capabilities in an application software provided by the embodiments of the present application is as Figure 1 shown, and the method includes: S11. Obtain the real-time temperature data and load status data of each available server node in the server cluster deployed with the application software, and the preset load balancing strategy. The load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate.

[0025] Among them, the server nodes in the server cluster can be simply referred to as nodes. The real-time temperature data is the hardware temperature information collected by temperature sensors deployed on the server nodes, which is used to reflect the heat dissipation status of the corresponding server nodes. The load status data includes the number of service requests currently processed by the server nodes and the resource occupancy rate, such as CPU occupancy rate, memory usage rate, etc., which is used to reflect the load situation of the nodes. The preset load balancing policy is a pre-defined rule for allocating service requests to appropriate nodes according to the load status data. The number of service requests refers to the total number of service requests currently being processed by the server nodes, which is used to reflect the load situation of the nodes. The resource occupancy rate refers to the proportion of the hardware resources (such as CPU, memory, network bandwidth) of the server nodes currently being used, which is used to quantify the load status of the nodes.

[0026] S12. When it is detected that the real-time temperature of any available server node exceeds the preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches the upper limit of the capacity, in combination with the virtual machine dynamic migration technology, the service capabilities running on the available server node are migrated to the standby server node, and the standby server node is another available server node in the server cluster.

[0027] Among them, the preset temperature threshold can refer to the safety upper limit of the hardware temperature of the server node. Exceeding this temperature threshold may cause a decrease in hardware performance or downtime. The upper limit of the capacity refers to the maximum value of the resource occupancy rate of the server node. Exceeding this maximum value may cause delays or failures in service request processing. The standby server node is another available server node in the server cluster that meets the conditions of having a temperature lower than the preset temperature threshold and a load not reaching the upper limit of the capacity. The virtual machine dynamic migration technology is a technology that migrates a running virtual machine (or container instance) from one physical server to another physical server without interrupting the service. In the embodiments of the present application, what is migrated is the service capabilities related to the application software. During the migration process, the memory status, CPU status, and network connection information of the virtual machine are synchronously updated to the target server in real time to ensure service continuity. Through the virtual machine dynamic migration technology in the embodiments of the present application, the physical heat dissipation policy of the server cluster can be synchronously adjusted to reduce the temperature fluctuation of the available server node.

[0028] S13. Correlate and analyze the real-time temperature data, load status data, and execution frequency of the migration operation of all available server nodes to obtain the correlation analysis result.

[0029] Among them, correlation analysis refers to analyzing the correlation between temperature data, load status data, and migration frequency through available server nodes, and is used to identify the coupling relationship between temperature fluctuations and sudden increases in load. The results of correlation analysis include features such as the time difference between the peak value of temperature fluctuations and sudden load increase events, and the correlation between the amplitude of temperature fluctuations and the intensity of sudden load increases. There is a strong correlation among temperature data, load status data, and the execution frequency of migration operations: temperature fluctuations are closely related to sudden increases in load, sudden increases in load may cause the temperature to rise, and exceeding the temperature standard may trigger migration operations. The migration frequency reflects the severity of load fluctuations and temperature fluctuations, and is an important basis for optimizing load balancing strategies and migration trigger thresholds. By constructing a correlation model, the relationship among the three can be quantified, and the system strategy can be dynamically optimized, so as to maintain the continuous response of service requests in high-concurrency scenarios and restrict the temperature fluctuation range of hardware resources.

[0030] S14. Based on the results of correlation analysis, dynamically optimize the node screening rules in the preset load balancing strategy and the trigger threshold of virtual machine dynamic migration technology, so that the optimized preset load balancing strategy can maintain the continuous response of service requests in high-concurrency scenarios and restrict the temperature fluctuation range of the hardware resources of the server cluster.

[0031] Among them, the node screening rule is a rule for selecting appropriate nodes to allocate service requests, including the weighting coefficients of temperature thresholds and load thresholds. The trigger threshold refers to the critical value for determining whether to trigger a certain operation (such as virtual machine dynamic migration). In the embodiments of the present application, the trigger threshold includes a temperature trigger threshold and a load trigger threshold, which are respectively used to determine whether the temperature or load of the server node reaches the condition for migration.

[0032] The following is a specific example: Suppose an e-commerce platform faces high-concurrency access during a certain event, and the CPU temperature of a certain node in the server cluster reaches 90°C, and the resource occupancy rate reaches 95%. At this time, the migration trigger module detects an abnormality and starts virtual machine dynamic migration to migrate the service capacity of this node to a standby node. The correlation analysis module finds that the temperature abnormality of this node is closely related to high load, and the policy optimization module adjusts the CPU temperature threshold to 80°C and preferentially selects low-load nodes to process new requests. Finally, the server cluster maintains the continuous response of service requests in high-concurrency scenarios and controls the temperature fluctuation range of hardware resources within a reasonable range.

[0033] By executing S11~S14, the embodiments of the present application achieve continuous response to service requests and temperature stability of hardware resources in high-concurrency scenarios by monitoring temperature and load data in real time, dynamically migrating service capabilities, associatively analyzing data, and optimizing load balancing strategies. Through priority allocation and resource requirement matching, critical services are ensured to be migrated first, enhancing system stability. Through real-time monitoring and secondary verification, it is avoided that the temperature or load of the standby node exceeds the limit during the migration process, guaranteeing the migration success rate. Through associative analysis and dynamic optimization, the migration frequency and resource waste are reduced, maximizing resource utilization. In the closed-loop feedback mechanism, the load balancing strategy and migration trigger threshold are continuously optimized to adapt to the changing load and hardware status.

[0034] In a possible embodiment, S12, in combination with virtual machine dynamic migration technology, migrates the service capabilities running on the available server nodes to the standby server node, where the standby server node is another available server node in the server cluster, including: Step 121, determine the priority of the service capabilities running on the available server nodes, where the priority is dynamically allocated based on the resource requirement characteristics and business types of the service capabilities. The resource requirement characteristics include the number of central processing unit (CPU) cores required, the memory capacity required, and the network bandwidth occupancy rate required.

[0035] Among them, the resource requirement characteristics include the number of CPU cores required, the memory capacity required, and the network bandwidth occupancy rate required, which are used to quantify the demand of the service capabilities for hardware resources. The business type is the business category to which the service capabilities belong (such as payment service, log service), which is used to dynamically allocate priorities. The number of CPU cores refers to the number of cores in the CPU that independently process tasks. Each core can execute one thread simultaneously, and the more cores there are, the stronger the parallel processing ability of the CPU. The network bandwidth occupancy rate refers to the proportion of the network bandwidth currently used by the server node to its total bandwidth, which is used to reflect the network resource usage of the node.

[0036] Step 122, based on the real-time temperature data and load status data of all available server nodes in the server cluster, filter out candidate server nodes with temperatures lower than the preset temperature threshold and loads not reaching the capacity limit.

[0037] Among them, the candidate server node is a node that meets the conditions of having a temperature lower than the preset threshold and a load not reaching the capacity limit. The number of candidate server nodes can be one or more. When there is one candidate server node, it can be used as the standby server node. Otherwise, execute step 123.

[0038] Step 123, when the number of candidate server nodes is multiple, sort the candidate server nodes according to the remaining resource capacity to generate a list of candidate server nodes.

[0039] Among them, the remaining resource capacity refers to the currently available number of CPU cores, memory capacity, and network bandwidth of the candidate server nodes. The candidate server node list refers to the list of nodes sorted by the remaining resource capacity, which is used to select the most suitable standby server node. In other words, the candidate server node list refers to the set of nodes filtered from the server cluster that meet specific conditions (such as the temperature being lower than the preset threshold and the load not reaching the capacity limit). These nodes serve as potential targets for the dynamic migration of virtual machines and are used to receive the service capabilities migrated from overloaded or overheated nodes.

[0040] Step 124: According to the priority of the service capabilities and the resource requirement characteristics, match the candidate server node with the most suitable remaining resource capacity from the candidate server node list and select the candidate server node with the smallest difference as the standby server node.

[0041] Among them, matching the most suitable remaining resource capacity means selecting a node from the candidate server node list whose remaining resource capacity best matches the resource requirement characteristics of the service capabilities. Specifically, it means selecting a node whose remaining number of CPU cores, memory capacity, and network bandwidth can maximize the satisfaction of the resource requirements of the service capabilities to be migrated while minimizing resource waste. The difference refers to the difference between the resource requirement characteristics of the service capabilities and the remaining resource capacity of the candidate server node. The following is an example: Suppose there is a service capability that requires the following resources: 4 CPU cores, 8 GB of core memory capacity, and a network bandwidth occupancy rate of 20%. There are two candidate server nodes. Among them, Node A: The available number of CPU cores is 5 cores, the available memory capacity is 10 GB, and the available network bandwidth occupancy rate is 25%. Node B: The available number of CPU cores is 6 cores, the available memory capacity is 12 GB, and the available network bandwidth occupancy rate is 30%. Calculate the difference. Among them, the difference between Node A and the resource requirement characteristics of the service capabilities is as follows: The difference in the number of CPU cores is 4 - 5 = -1, and the absolute value is 1. The difference in memory capacity is 8 - 10 = -2, and the absolute value is 2. The difference in network bandwidth occupancy rate is 20 - 25 = -5, and the absolute value is 5. Therefore, the total difference is 1 + 2 + 5 = 8. The difference between Node B and the resource requirement characteristics of the service capabilities is as follows: The difference in the number of CPU cores is 4 - 6 = -2, and the absolute value is 2. The difference in memory capacity is 8 - 12 = -4, and the absolute value is 4. The difference in network bandwidth occupancy rate is 20 - 30 = -10, so the absolute value is 10. The total difference is 2 + 4 + 10 = 16. Compare the total differences: The total difference between Node A and the resource requirement characteristics of the service capabilities is 8, and the total difference between Node B and the resource requirement characteristics of the service capabilities is 16. The difference between Node A and the resource requirement characteristics of the service capabilities is smaller. Therefore, select Node A as the standby server node.

[0042] Step 125: Before the migration execution, perform secondary verification on the real-time temperature data and load status data of the standby server node to ensure that the temperature fluctuation of the standby server node during the migration process is within the temperature fluctuation range and the load status data is within the preset load status safety range.

[0043] Among them, secondary verification refers to re-checking the temperature and load status of the standby server node before migration to ensure migration safety. The temperature fluctuation range includes a temperature lower limit and a temperature upper limit. The temperature lower limit refers to the lowest safe temperature of the standby server node, usually defined by the hardware manufacturer, such as 10°C. Temperatures below this may cause abnormal hardware performance. The temperature upper limit refers to the highest safe temperature of the standby server node, usually defined by the hardware manufacturer, such as 80°C. Temperatures exceeding this may cause hardware damage or performance degradation. The load status safety range includes a load lower limit and a load upper limit. The load lower limit refers to the lowest load status of the standby server node, usually 0%, indicating that the node is in an idle state. The load upper limit refers to the highest load status of the standby server node, usually defined by the system administrator, such as 80%. Exceeding this value may cause node overload and affect service performance.

[0044] Step 126: Migrate the service capabilities to the standby server node through virtual machine live migration technology. The migration process of the service capabilities includes migrating each service capability in order of priority and updating the load status data and temperature data of the standby server node in real time after each migration is completed.

[0045] Among them, migrating each service capability in order of priority means migrating the service capabilities from the source node to the standby server node in order from high to low according to the importance (priority) of the service capabilities. The service capabilities with higher priority are migrated first to ensure that critical services are not affected during the migration process.

[0046] The following is a specific example: An online education platform faced high concurrent access during the exam season. The CPU temperature of a certain node in the server cluster reached 90°C, and the resource occupancy rate reached 95%. The platform dynamically assigns priorities according to the resource demand characteristics of service capabilities (such as the number of CPU cores, memory capacity, network bandwidth occupancy rate) and business types (such as real-time video, online exams). For example, the priority of the real-time video service is higher than that of the online exam service because the video service is more sensitive to latency. Through temperature sensors and load monitoring tools, candidate nodes with a temperature below 85°C and a load not reaching 90% are screened out to generate a list of candidate nodes. The candidate nodes are sorted according to the remaining resource capacity (such as CPU, memory, bandwidth) to generate a list of candidate nodes. According to the service capability priority and resource demand characteristics, the standby node with the most suitable remaining resource capacity is matched from the list of candidate nodes. For example, the node with the remaining CPU cores and memory capacity closest to the service capability requirements is selected. Before migration, the temperature and load data of the standby node are verified again to ensure that the temperature fluctuation is within the preset range (such as ±5°C) and the load status data is within the safe range (such as CPU usage < 80%). Through virtual machine dynamic migration technology, the service capabilities are migrated sequentially in the order of priority, and the load status data and temperature data of the standby node are updated in real time after each migration. The platform successfully migrates the service capabilities to the standby node, ensuring the high concurrent access requirements during the exam season and controlling the temperature fluctuation range of the hardware resources within a reasonable range.

[0047] By executing steps 121 to 126, in the embodiment of the present application, by dynamically determining the service capability priority, screening candidate nodes, matching standby nodes, performing secondary verification, and executing migration, the embodiment of the present application effectively solves the problems of resource allocation and temperature control in the server cluster under high concurrent scenarios. The optimized strategy not only improves the stability and availability of the system, but also reduces the temperature fluctuation of the hardware resources, extends the device life, and improves the user experience at the same time.

[0048] In a possible embodiment, in step 126, the service capabilities are migrated to the standby server node through virtual machine dynamic migration technology. The migration process of the service capabilities includes migrating each service capability sequentially in the order of priority, including: Step a1: According to the priorities of each service capability, all the service capabilities running on the available server nodes are arranged in descending order to generate a migration task queue.

[0049] Among them, the migration task queue is the core data structure in the dynamic optimization process of the server cluster, used to manage and schedule the migration tasks of service capabilities.

[0050] Step a2: Sequentially select the service capabilities to be migrated from the migration task queue, and based on the resource requirement characteristics of the service capabilities to be migrated, calculate whether the remaining resource capacity of the standby server node meets the resource requirements of the service capabilities to be migrated.

[0051] Among them, the service capabilities to be migrated refer to the service instances or applications that need to be migrated from the current node to the standby node in the server cluster. These service capabilities usually include virtual machines, containers, or application programs running on the server nodes, and their migration requirements may be triggered by various factors, such as too high node temperature, too heavy load, or uneven resource allocation.

[0052] Step a3: If the resource requirements are met, start the virtual machine dynamic migration technology to migrate the service capabilities to be migrated from the available server node to the standby server node.

[0053] Step a4: Or, if the resource requirements are not met, suspend the current migration task, and re-screen the standby server nodes from the list of candidate server nodes to update the migration targets of the unmigrated tasks in the migration task queue.

[0054] Among them, the migration target refers to the standby server node as the final destination of the service capabilities to be migrated during the dynamic optimization process of the server cluster. Specifically, the migration target is determined based on the resource requirement characteristics of the service capabilities to be migrated (such as the number of CPU cores, memory capacity, network bandwidth occupancy rate) and the remaining resource capacity of the standby node, after screening and matching. The selection of the migration target needs to ensure that the standby node can meet the resource requirements of the service capabilities to be migrated and maintain the stability and continuity of the system during the migration process.

[0055] By executing steps a1 - a4, the embodiments of the present application achieve continuous response to service requests and temperature stability of hardware resources in a high-concurrency scenario through priority sorting, resource capacity verification, virtual machine dynamic migration, and dynamic screening of standby nodes. Through priority allocation, it ensures that critical services are migrated first, improving system stability; then resource capacity verification to avoid node overload after migration and ensure service performance; based on the virtual machine dynamic migration technology, it ensures that services are not interrupted and improves the user experience; and dynamic screening of standby nodes supports dynamic adjustment of migration tasks to ensure the migration success rate.

[0056] In a possible embodiment, step a3: If the resource requirements are met, start the virtual machine dynamic migration technology to migrate the service capabilities to be migrated from the available server node to the standby server node, including: Step b1: Generate migration configuration parameters according to the resource requirement characteristics of the service capabilities to be migrated, and pre-allocate the resources corresponding to the migration configuration parameters on the standby server node.

[0057] Among them, the migration configuration parameters refer to the configuration information generated according to the resource requirement characteristics of the service capabilities to be migrated (such as the number of CPU cores, memory capacity, network bandwidth occupancy rate, etc.). These parameters are used to guide the migration process to ensure that the standby server node can meet the resource requirements of the service capabilities to be migrated. Pre-allocating the resources corresponding to the migration configuration parameters means pre-allocating resources on the standby server node according to the migration configuration parameters. This step ensures that the standby node has sufficient resources to carry the service capabilities to be migrated before the migration starts.

[0058] Step b2: Start the virtual machine dynamic migration technology to synchronize the memory state, central processing unit state, and network connection information of the service capabilities to be migrated from the available server node to the standby server node.

[0059] Among them, the synchronization process includes the copying of memory pages, the saving of CPU states, and the switching of network connections.

[0060] Step b3: During the synchronization process, monitor the standby server node in real time. If it is detected that the resource occupancy rate of the standby server node exceeds the preset occupancy rate threshold or the temperature exceeds the preset temperature threshold, pause the synchronization process and trigger the resource reallocation mechanism to re-evaluate and adjust the pre-allocated resources.

[0061] Among them, the resource reallocation mechanism refers to the mechanism that, during the synchronization process, if the resource occupancy rate of the standby node exceeds the preset occupancy rate threshold or the temperature exceeds the preset temperature threshold, pauses the synchronization process and re-evaluates and adjusts the pre-allocated resources.

[0062] Step b4: After the synchronization is completed, switch the running state of the service capabilities to be migrated from the available server node to the standby server node and release the corresponding resources on the available server node.

[0063] Among them, the corresponding resources refer to the hardware resources pre-allocated and used on the standby server node to carry the service capabilities to be migrated. These resources include: CPU resources, memory resources, and network bandwidth resources.

[0064] Step b5: After the switching is completed, update the load status data and temperature data of the standby server node in real time. The temperature data is periodically collected at a preset sampling frequency by the temperature sensors deployed on the standby server node.

[0065] Among them, the update of the load status data and temperature data refers to the process of updating the load status data and temperature data of the standby node in real time after the switching is completed.

[0066] The following is a specific example: Suppose an online education platform faces high-concurrency access during the exam season. The CPU temperature of a certain node in the server cluster reaches 90°C, and the resource occupancy rate reaches 95%. At this time, the platform generates migration configuration parameters according to the resource demand characteristics of the service capabilities to be migrated, including a 4-core CPU, 8GB of memory, and a 100Mbps network bandwidth. After pre-allocating these resources on the standby node, the virtual machine dynamic migration technology is started to migrate the service capabilities from the current node to the standby node. During the synchronization process, the resource occupancy rate and temperature data of the standby node are monitored in real time. If the preset threshold is exceeded, the synchronization process is paused and the pre-allocated resources are re-evaluated and adjusted. Finally, the platform successfully migrates the service capabilities to the standby node, ensuring the high-concurrency access requirements during the exam season.

[0067] By performing steps b1~b5, in the embodiment of the present application, by dynamically determining the migration configuration parameters, pre-allocating resources, starting the virtual machine dynamic migration technology, monitoring the standby node in real time, switching the running state, and updating the load status data, this solution effectively solves the problems of resource allocation and temperature control in the server cluster under high-concurrency scenarios, improves the stability and availability of the system, reduces the temperature fluctuation of the hardware resources, extends the device life, and at the same time improves the user experience.

[0068] In a possible embodiment, S14. Based on the correlation analysis result, dynamically optimize the node screening rule in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing policy can maintain the continuous response of service requests in high-concurrency scenarios and constrain the temperature fluctuation range of the hardware resources of the server cluster, including: Step 141. Extract the coupling characteristics of temperature fluctuation and sudden load increase from the correlation analysis result, and optimize the node screening rule in the preset load balancing policy based on the coupling characteristics.

[0069] Among them, the coupling characteristic refers to the correlation between the temperature change of the server node and the sudden load increase. This coupling characteristic reflects the impact of sudden load increase on the temperature of the hardware resources and is an important basis for optimizing the node screening rule. The node screening rule is a rule for selecting appropriate nodes to allocate service requests, including the weighted coefficients of the temperature threshold and the load threshold.

[0070] Step 142. According to the execution frequency of the migration operation in the correlation analysis result, combine the temperature change curves and load change curves of the available server nodes and standby server nodes during the migration process, calculate the attenuation function parameters, and dynamically adjust the sensitivity of the migration trigger threshold through the attenuation function parameters to dynamically optimize the migration trigger threshold.

[0071] Among them, the attenuation function parameter refers to the parameter used to adjust the sensitivity of the migration trigger threshold. Through this parameter, the frequency of migration operations can be dynamically controlled to avoid resource waste or system instability caused by frequent migrations. The sensitivity of the migration trigger threshold refers to the degree of response of the system to the resource occupancy rate or temperature change of the server node. The higher the sensitivity, the more rapid the response of the system to the resource occupancy rate or temperature change, and the more frequent the migration operations may be; the lower the sensitivity, the more sluggish the response of the system to the change, and the fewer the migration operations may be. The execution frequency of migration operations is the number of migration operations occurring per unit time, reflecting the frequency of the migration strategy.

[0072] Step 143: Based on the optimized node screening rule and migration trigger threshold, obtain the optimized preset load balancing strategy, and apply the optimized preset load balancing strategy to the server cluster to form a closed-loop feedback mechanism.

[0073] Among them, the optimized preset load balancing strategy refers to the load balancing strategy formed based on the optimized node screening rule and migration trigger threshold. Through the closed-loop feedback mechanism, this strategy dynamically adjusts the resource allocation and migration operations of the server cluster to ensure the stability and temperature control of the system in high-concurrency scenarios. The closed-loop feedback mechanism is a control mechanism that adjusts the system input by monitoring the system output in real time and comparing it with the expected value to achieve system stability and optimization. The closed-loop feedback mechanism includes monitoring and feedback, comparison and deviation calculation, adjustment and optimization, closed-loop cycle, optimizing the node screening rule, dynamically adjusting the migration trigger threshold, application and feedback. Through the closed-loop feedback mechanism, the embodiments of the present application can maintain the continuous response of service requests in high-concurrency scenarios and constrain the temperature fluctuation range of hardware resources, thereby improving the stability and efficiency of the server cluster.

[0074] The following is a specific example: When the stock market opens, a certain securities trading system faces tens of thousands of trading requests per second. The CPU temperature of a certain node in the server cluster soars to 92°C, and the load reaches 98%. Extract the temperature fluctuation data of this node (the temperature rises from 60°C to 92°C within 5 minutes after the opening) and the sudden increase in load data (the request volume suddenly increases from 1000 / s to 20000 / s). Through analysis by the long short-term memory network model, it is found that the temperature response delay is 30 seconds after the sudden increase in load, and the temperature rises 10°C for every additional 5000 request volumes. Optimize the node screening rules: Prioritize selecting nodes with a temperature response delay greater than 1 minute and a temperature rise rate per request volume lower than 5°C / 5000 as standby nodes. The migration frequency after the opening is counted as 10 times per hour. Combining with the temperature / load curve fitting, the decay function parameter λ = 0.15 is calculated. Dynamically adjust the migration trigger threshold: The CPU temperature threshold is relaxed from 85°C to 87°C, and the migration operation frequency drops to 6 times per hour. After deploying the optimization strategy, the monitoring shows that the temperature fluctuation range of the standby node drops from ±15°C to ±8°C. Update the decay function parameter every 10 minutes. Finally, λ stabilizes at 0.12, the trigger threshold is adjusted to 86.5°C, and the server cluster reaches a balanced state.

[0075] By executing steps 141 to 143, the embodiments of the present application realize the continuous response of service requests and the temperature stability of hardware resources in high-concurrency scenarios by extracting the coupling characteristics of temperature fluctuations and sudden increases in load, dynamically adjusting the migration trigger threshold, and forming a closed-loop feedback mechanism. By optimizing the node screening rules, it is ensured that the load distribution is more reasonable, avoiding node overload or hardware overheating; by dynamically adjusting the migration trigger threshold, unnecessary migration operations are reduced, and the system overhead is reduced; through the closed-loop feedback mechanism, the load balancing strategy and the migration trigger threshold are continuously optimized to adapt to the changing load and hardware status.

[0076] In a possible embodiment, step b1: Generate migration configuration parameters according to the resource demand characteristics of the service capabilities to be migrated, including: Step c1: Generate initial pre-allocation parameters according to the required number of central processing unit (CPU) cores, the required memory capacity, and the required network bandwidth occupancy rate.

[0077] Among them, the number of CPU cores refers to the number of physical computing cores inside the processor, and each core can execute instructions and tasks independently. For example, an 8-core CPU means there are 8 independent processing units that can handle tasks of 8 threads simultaneously. The number of cores directly affects the parallel computing ability and multitasking processing efficiency, and is one of the core indicators for measuring server performance. The memory capacity is the size of the random access memory used by the server for temporary storage and processing of data, measured in bytes (such as 16GB, 64GB). The memory capacity determines the amount of data that the server can process simultaneously. If the memory is insufficient, it will lead to frequent disk swapping, reducing performance. The initial pre-allocation parameter is a resource allocation scheme preset based on task requirements (number of CPU cores, memory capacity, network bandwidth) and is used to guide the initial deployment of virtual machines or containers. For example, the initial parameter may be "number of CPU cores = 4, memory = 16GB, bandwidth = 100Mbps".

[0078] Step c2: Combine the real-time temperature data and load status data of the standby server node, and dynamically adjust the initial pre-allocation parameter to obtain the adjusted pre-allocation parameter.

[0079] Among them, the adjusted pre-allocation parameter refers to a resource allocation scheme that is dynamically optimized based on the initial resource allocation (number of CPU cores, memory capacity, network bandwidth) and the real-time server status (temperature, load).

[0080] Step c3: Based on the migration strategy constraints, set the corresponding fault tolerance and efficiency parameters, and combine the fault tolerance and efficiency parameters with the adjusted pre-allocation parameter to obtain the migration configuration parameter.

[0081] Among them, the migration strategy constraint refers to the restrictive conditions or rules set to ensure the migration success rate and system stability during the dynamic migration of virtual machines. The fault tolerance and efficiency parameters refer to the parameters set to balance system stability and migration efficiency during the migration process. The migration configuration parameter refers to the configuration information generated according to the resource requirement characteristics of the service to be migrated, the real-time status of the standby server node, and the migration strategy constraints, and is used to guide the migration operation.

[0082] The following is a specific example: In the scenario of a major e-commerce promotion, the server cluster needs to handle a sudden surge in service requests. By collecting real-time temperature data and load status data of each node, it is detected that a certain node has an excessive load due to the surge in requests, and a migration task is immediately started. According to the number of CPU cores, memory capacity, and network bandwidth occupancy rate required for service capabilities, the initial pre-allocation parameters "Number of CPU cores = 4, Memory = 16GB, Bandwidth = 100Mbps" are generated; combining the real-time temperature data and load status data of the standby node, the initial pre-allocation parameters are dynamically adjusted to "Number of CPU cores = 3, Memory = 12GB, Bandwidth = 80Mbps"; based on the migration policy constraints, fault tolerance and efficiency parameters are set, such as the number of retries = 3, migration bandwidth limit = 100Mbps, and the migration configuration parameters "Number of CPU cores = 3, Memory = 12GB, Bandwidth = 80Mbps, Number of retries = 3, Migration bandwidth limit = 100Mbps" are generated; the migration configuration parameters are applied to the migration task to ensure that the migration process is both efficient and reliable.

[0083] By executing steps c1 to c3, the embodiments of the present application realize seamless migration of service capabilities in a high-concurrency scenario by generating initial pre-allocation parameters, dynamically adjusting pre-allocation parameters, and setting fault tolerance and efficiency parameters. The initial pre-allocation parameters ensure that the resource allocation of the migration target node meets the service capability requirements; by dynamically adjusting the pre-allocation parameters, it adapts to the real-time status of the standby node to avoid node overload or hardware overheating; setting fault tolerance and efficiency parameters balances system stability and migration efficiency to ensure that the migration process is both efficient and reliable.

[0084] In a possible embodiment, step 143, based on the optimized node screening rule and migration trigger threshold, obtains the optimized preset load balancing strategy, including: Step d1, adding an edge distance factor corresponding to a specified scenario to the optimized node screening rule, where the specified scenarios include a high-concurrency scenario and a preset high-temperature sensitive scenario; the preset high-temperature sensitive scenario refers to a scenario in the server cluster where the node temperature is likely to exceed the safety threshold due to limited hardware heat dissipation conditions, environmental temperature fluctuations, or high-load operation.

[0085] Among them, the edge distance factor is a parameter used to quantify the physical distance and network topology relationship between nodes. The high-concurrency scenario refers to a scenario where the system needs to process a large number of user requests simultaneously in a short period of time.

[0086] Step d2, based on the node screening rule with the added edge distance factor, the migration trigger threshold, and the abnormal fuse mechanism, uses the federated learning model to generate the optimized preset load balancing strategy.

[0087] Among them, the abnormal fusing mechanism means that if three consecutive migration operations cause the temperature of the standby node to exceed the safe range, the migration is suspended and an artificial intervention alarm is triggered; after the alarm is lifted, the attenuation function parameters are reset and the optimization strategy is reloaded. The edge distance factor is a parameter used to quantify the physical distance between nodes and the network topology relationship. It reflects the physical distance between server nodes (such as the rack position in the data center, the distance across data centers, etc.) and the network topology structure (such as the network hop count, latency, etc.) on the service performance. The smaller the edge distance factor, the closer the physical distance between nodes or the better the network topology relationship, the lower the communication latency, and the higher the data transmission efficiency. The node screening rule with the addition of the edge distance factor is as follows: Exemplarily, in a high-concurrency scenario, the system needs to quickly respond to a large number of user requests, so nodes with a smaller edge distance factor are preferentially selected to reduce communication latency and improve the response speed. In a high-temperature sensitive scenario, nodes with a smaller edge distance factor are preferentially selected to reduce the additional heat generated due to communication between nodes and at the same time reduce the additional load brought by network transmission.

[0088] In the embodiments of the present application, in a high-concurrency scenario, nodes with a smaller edge distance factor can reduce communication latency and improve data transmission efficiency, thereby enhancing the response speed and processing capacity of the system. By preferentially selecting nodes with a smaller edge distance factor, performance bottlenecks caused by network latency can be avoided, ensuring that the system can still operate stably under high load. In a high-temperature sensitive scenario, nodes with a smaller edge distance factor can reduce the additional heat brought by communication between nodes and reduce the heat dissipation pressure. By reducing the network transmission distance and hop count, the additional load caused by network communication can be reduced, thereby reducing the temperature fluctuation of the nodes and preventing the node temperature from exceeding the safety threshold. The introduction of the edge distance factor makes node screening more refined, can better match service requirements with node resources, and avoid resource waste. By optimizing node selection, unnecessary network transmission and heat generation can be reduced, thereby improving the overall resource utilization efficiency. In high-concurrency and high-temperature sensitive scenarios, the introduction of the edge distance factor helps to select better nodes and reduce the system failure risk. By reducing communication latency and heat generation, the stability and reliability of the system can be enhanced, ensuring the continuous availability of service capabilities. The embodiments of the present application generate a dynamic load balancing strategy by fusing node screening rules, migration thresholds, and fusing mechanisms through a federated learning framework.

[0089] By performing steps d1~d3, the embodiment of the present application improves the response efficiency through the multi-dimensional fusion of the edge distance factor and the dynamic policy optimization driven by federated learning. Through the priority scheduling of edge nodes, the cross-regional network latency is reduced from 50 ms to 20 ms, and the efficiency is increased by 40%. In high-temperature sensitive scenarios, the dynamic adjustment of the migration threshold effectively avoids frequent fusing caused by temperature fluctuations, and the system stability is enhanced by 50%. The federated learning model accurately predicts the load distribution, reduces the idle computing resources, and optimizes the resource utilization rate by 30%. In addition, the fusing mechanism and the federated feedback form a double protection, reducing the system crash risk and improving the abnormal fault tolerance ability by 70%. In the e-commerce Double Eleven scenario, the server cluster's bearing capacity is increased by 3 times, and the operation and maintenance cost is reduced by 40%. At the same time, it meets the cross-border data privacy compliance requirements, achieving efficient, stable, and secure load balancing optimization.

[0090] Figure 2 The following is a schematic structural diagram of a service capability invocation system in an application software provided by an embodiment of the present application, as Figure 2 shown. The system includes: An acquisition module 21, which acquires the real-time temperature data and load status data of each available server node in the server cluster where the application software is deployed, and a preset load balancing policy. The load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate.

[0091] A detection and migration module 22, when it detects that the real-time temperature of any available server node exceeds the preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches the capacity upper limit, combines the virtual machine dynamic migration technology to migrate the service capabilities running on the available server node to a standby server node, and the standby server node is another available server node in the server cluster.

[0092] An association analysis module 23, which performs an association analysis on the real-time temperature data, load status data, and execution frequency of migration operations of all available server nodes to obtain an association analysis result.

[0093] An optimization module 24, based on the association analysis result, dynamically optimizes the node screening rules in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing policy maintains the continuous response of service requests in high-concurrency scenarios and restricts the temperature fluctuation range of the server cluster's hardware resources.

[0094] Figure 2 The described service capability invocation system in an application software can execute Figure 1A method for invoking service capabilities in an application software described in the illustrated embodiment, the implementation principle and technical effects thereof will not be elaborated further. For each module and unit in the service capability invocation system in the above embodiment of the application software, the specific manner of performing operations has been described in detail in the embodiment related to the method, and will not be elaborated herein.

[0095] In a possible design, Figure 2 The service capability invocation system of the illustrated embodiment in an application software can be implemented as a computing device, such as Figure 3 shown, the computing device may include a storage component 31 and a processing component 32.

[0096] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0097] The processing component 32 is configured to: obtain the real-time temperature data and load status data of each available server node in the server cluster where the application software is deployed, and a preset load balancing policy, the load status data including the number of service requests currently processed by the available server node and the resource occupancy rate; when it is detected that the real-time temperature of any one of the available server nodes exceeds the preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches the preset capacity upper limit, in combination with the virtual machine dynamic migration technology, migrate the service capabilities running in the available server node to a standby server node, the standby server node being another available server node in the server cluster; perform correlation analysis on the real-time temperature data, the load status data, and the execution frequency of the migration operation of all available server nodes to obtain a correlation analysis result; based on the correlation analysis result, dynamically optimize the node screening rules in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing policy can maintain the continuous response of service requests in a high-concurrency scenario and constrain the temperature fluctuation range of the hardware resources of the server cluster.

[0098] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the above methods.

[0099] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), 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 disks or optical discs.

[0100] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0101] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0102] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0103] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0104] The embodiments of the present application also provide a computer storage medium storing a computer program, which when executed by a computer can implement the above-mentioned Figure 1 service capability invocation method in an application software shown in the embodiments.

[0105] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0106] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0107] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for adjusting service capabilities in an application software, characterized in that, Including: Obtain the real-time temperature data and load status data of each available server node in the server cluster where the application software is deployed, and a preset load balancing policy. The load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate; When it is detected that the real-time temperature of any one of the available server nodes exceeds the preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches the preset capacity upper limit, combine the virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, and the standby server node is another available server node in the server cluster; Perform correlation analysis on the real-time temperature data, the load status data of all available server nodes, and the execution frequency of the migration operation to obtain a correlation analysis result; Based on the correlation analysis result, dynamically optimize the node screening rules in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology, so that the optimized preset load balancing policy can maintain the continuous response of service requests in a high-concurrency scenario and constrain the temperature fluctuation range of the hardware resources of the server cluster.

2. The method according to claim 1, wherein The combining the virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, and the standby server node is another available server node in the server cluster, includes: Determine the priority of the service capabilities running in the available server node, and the priority is dynamically allocated based on the resource requirement characteristics and business types of the service capabilities. The resource requirement characteristics include the number of central processing unit (CPU) cores required, the memory capacity required, and the network bandwidth occupancy rate required; Based on the real-time temperature data and the load status data of all available server nodes in the server cluster, screen candidate server nodes with a temperature lower than the preset temperature threshold and a load not reaching the capacity upper limit; When the number of the candidate server nodes is multiple, sort the candidate server nodes according to the remaining resource capacity to generate a candidate server node list; According to the priority of the service capabilities and the resource requirement characteristics, match the candidate server node with the most adaptable remaining resource capacity and the smallest difference in the candidate server node list as the standby server node; Before the migration execution, perform secondary verification on the real-time temperature data and the load status data of the standby server node to ensure that the temperature fluctuation of the standby server node during the migration process is within the temperature fluctuation range, and the load status data is within the preset load status safety range; Migrate the service capabilities to the standby server node through the virtual machine dynamic migration technology. The migration process of the service capabilities includes migrating each service capability in order of priority and updating the load status data and temperature data of the standby server node in real time after each migration is completed.

3. The method according to claim 2, wherein The migrating the service capabilities to the standby server node through the virtual machine dynamic migration technology, and the migration process of the service capabilities includes migrating each service capability in order of priority, includes: Arrange all service capabilities running on the available server nodes in descending order according to the priority of each service capability to generate a migration task queue; Select the service capabilities to be migrated from the migration task queue in sequence, and based on the resource requirement characteristics of the service capabilities to be migrated, calculate whether the remaining resource capacity of the standby server node meets the resource requirements of the service capabilities to be migrated; If the resource requirements are met, start the virtual machine live migration technology to migrate the service capabilities to be migrated from the available server nodes to the standby server nodes; Alternatively, if the resource requirements are not met, suspend the current migration task and re-screen the standby server nodes from the candidate server node list to update the migration targets of the un-migrated tasks in the migration task queue.

4. The method according to claim 3, wherein If the resource requirements are met, start the virtual machine live migration technology to migrate the service capabilities to be migrated from the available server nodes to the standby server nodes, including: Generate migration configuration parameters according to the resource requirement characteristics of the service capabilities to be migrated, and pre-allocate the resources corresponding to the migration configuration parameters on the standby server nodes; Start the virtual machine live migration technology to synchronize the memory state, central processing unit state, and network connection information of the service capabilities to be migrated from the available server nodes to the standby server nodes; During the synchronization process, monitor the standby server node in real time. If it is detected that the resource occupancy rate of the standby server node exceeds the preset occupancy rate threshold or the temperature exceeds the preset temperature threshold, suspend the synchronization process and trigger the resource re-allocation mechanism to re-evaluate and adjust the pre-allocated resources; After the synchronization is completed, switch the running state of the service capabilities to be migrated from the available server nodes to the standby server nodes, and release the corresponding resources on the available server nodes; After the switching is completed, update the load status data and temperature data of the standby server node in real time. The temperature data is periodically collected at a preset sampling frequency by the temperature sensor deployed on the standby server node.

5. The method according to claim 1, characterized in that, Based on the correlation analysis results, dynamically optimize the node selection rules in the preset load balancing strategy and the trigger threshold of the virtual machine live migration technology, so that the optimized preset load balancing strategy can maintain continuous response to service requests in high-concurrency scenarios and constrain the temperature fluctuation range of the server cluster hardware resources, including: Extract the coupling characteristics of temperature fluctuation and load surge from the correlation analysis results, and based on the coupling characteristics, optimize the node selection rules in the preset load balancing strategy; According to the execution frequency of the migration operation in the correlation analysis results, combine the temperature change curves and load change curves of the available server nodes and the standby server nodes during the migration process, calculate the attenuation function parameters, and dynamically adjust the sensitivity of the migration trigger threshold through the attenuation function parameters to dynamically optimize the migration trigger threshold; Based on the optimized node screening rules and migration trigger thresholds, the optimized preset load balancing policy is obtained, and the optimized preset load balancing policy is applied to the server cluster to form a closed-loop feedback mechanism.

6. The method according to claim 4, characterized in that Generating migration configuration parameters according to the resource requirement characteristics of the service capabilities to be migrated, including: Generating initial pre-allocation parameters according to the required number of CPU cores, required memory capacity, and required network bandwidth occupancy rate; Combining the real-time temperature data of the standby server node and the load status data to dynamically adjust the initial pre-allocation parameters to obtain adjusted pre-allocation parameters; Based on the migration policy constraints, setting corresponding fault tolerance and efficiency parameters, and combining the fault tolerance, the efficiency parameters with the adjusted pre-allocation parameters to obtain migration configuration parameters.

7. The method according to claim 5, characterized in that, The obtaining the optimized preset load balancing policy based on the optimized node screening rules and migration trigger thresholds includes: Adding an edge distance factor corresponding to a specified scenario to the optimized node screening rules, where the specified scenarios include high-concurrency scenarios and preset high-temperature sensitive scenarios; Based on the node screening rules with the added edge distance factor, migration trigger thresholds, and an exception fusing mechanism, using a federated learning model to generate the optimized preset load balancing policy.

8. A service capability invocation system in an application software, characterized in that Including: An acquisition module for acquiring the real-time temperature data and load status data of each available server node in a server cluster deployed with application software, and a preset load balancing policy, where the load status data includes the number of service requests currently processed by the available server node and the resource occupancy rate; A detection and migration module for, when detecting that the real-time temperature of any of the available server nodes exceeds a preset temperature threshold or the resource occupancy rate indicates that the capacity of the server node reaches a preset capacity upper limit, combining virtual machine dynamic migration technology to migrate the service capabilities running in the available server node to a standby server node, where the standby server node is another available server node in the server cluster; An association analysis module for performing an association analysis on the real-time temperature data, the load status data, and the execution frequency of migration operations of all available server nodes to obtain an association analysis result; An optimization module for dynamically optimizing the node screening rules in the preset load balancing policy and the trigger threshold of the virtual machine dynamic migration technology based on the association analysis result, so that the optimized preset load balancing policy maintains continuous response to service requests in high-concurrency scenarios and restricts the temperature fluctuation range of the hardware resources of the server cluster.

9. A computing device, characterized in that, Including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a service capability calling method in an application software as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, it implements a service capability calling method in an application software as described in any one of claims 1 to 7.

Images