Dynamic configuration computing resource management method and system based on power wireless terminal

Abstract

The application provides a kind of dynamic configuration computing resource management method and system based on power wireless terminal, by obtaining the real-time running state data set of power wireless terminal and power service demand data set, the multi-dimensional feature fusion processing is carried out to real-time running state data set and power service demand data set, generates resource adaptation feature set, based on the preset resource adaptation rule base, real-time strategy matching processing is carried out to resource adaptation feature set, generates computing resource configuration strategy set, to adjust operation is carried out to the computing resource of power wireless terminal in this way, generates resource allocation verification result and terminal running efficiency feedback data, terminal running efficiency feedback data is continuously compared with the preset efficiency optimization threshold, and according to the comparison result, resource adaptation rule base is iteratively updated processing. Adopting the application can realize the accurate matching of computing resource allocation and power service demand fluctuation.

Description

Technical Field

[0001] This invention relates to the fields of resource management and data processing, and more specifically, to a method and system for dynamic configuration computing resource management based on a power wireless terminal. Background Technology

[0002] With the increasing intelligence of power systems, dynamic allocation of computing resources for power wireless terminals has become a key technology for ensuring the efficient operation of power services. Existing technologies typically allocate resources based on static load monitoring of terminal devices or independent priority ranking of service demands, such as triggering resource adjustments through fixed thresholds or matching task requirements using predefined rules. However, such methods struggle to cope with the complex operating conditions in power scenarios, including instantaneous fluctuations in network load, sudden surges in service requests, and overlapping environmental interference. This can easily lead to a disconnect between resource allocation strategies and real-time demands, resulting in problems such as unbalanced utilization of computing cores, network bandwidth contention, and delays in high-priority task responses. Furthermore, traditional static rule bases lack dynamic feedback mechanisms and cannot continuously optimize strategies based on terminal performance data. Over long-term operation, this can lead to resource fragmentation and strategy coordination failures, severely restricting the resource scheduling efficiency and service stability of power wireless terminals in scenarios such as intelligent inspection and emergency fault response. Summary of the Invention

[0003] This invention provides a method and system for dynamic configuration computing resource management based on a power wireless terminal.

[0004] In a first aspect, embodiments of the present invention provide a method for dynamically configuring computing resources management based on a power wireless terminal, comprising: acquiring a real-time operating status data set and a power service demand data set of the power wireless terminal, wherein the real-time operating status data set includes the terminal device's operating parameters and network load fluctuation characteristics, and the power service demand data set includes service priority identifiers and real-time task processing requirements; performing multi-dimensional feature fusion processing on the real-time operating status data set and the power service demand data set to generate a resource adaptation feature set, wherein the resource adaptation feature set includes terminal device load balancing characteristics, service demand conflict characteristics, and resource allocation efficiency evaluation indicators; performing real-time policy matching processing on the resource adaptation feature set based on a preset resource adaptation rule base to generate a computing resource configuration strategy set, wherein the computing resource configuration strategy set includes resource allocation priority adjustment strategies, task scheduling optimization strategies, and network bandwidth reallocation strategies; performing adjustment operations on the computing resources of the power wireless terminal according to the computing resource configuration strategy set, generating resource allocation verification results and terminal operating efficiency feedback data; continuously comparing the terminal operating efficiency feedback data with a preset efficiency optimization threshold, and iteratively updating the resource adaptation rule base according to the comparison results.

[0005] Secondly, embodiments of the present invention provide a computer system, including: a memory storing a computer program; and a processor for loading the computer program to implement the dynamic configuration computing resource management method based on a power wireless terminal as described above.

[0006] This invention provides a dynamic configuration computing resource management method based on power wireless terminals. By acquiring dynamic data sets of terminal device operating status and power service demands in real time, a multi-dimensional feature fusion mechanism is established to generate a dynamic resource adaptation feature set including load balancing features, service conflict features, and resource efficiency evaluation indicators. Based on a preset rule base, dynamic policy matching is performed to generate a collaboratively optimized resource allocation strategy set, achieving precise matching between computing resource allocation and fluctuations in power service demands. This method effectively solves the problems of low resource utilization and task timeouts caused by sudden changes in network load or service priority conflicts in traditional static resource allocation models through real-time coupled analysis of device status and service demands. Simultaneously, by combining terminal operating performance feedback data with a continuous iterative update mechanism of the rule base, dynamic adaptive resource management is formed, significantly improving the flexibility and stability of resource allocation strategies in complex power scenarios. Furthermore, a multi-dimensional policy collaborative verification mechanism avoids allocation conflicts across different resource dimensions, ensuring the spatiotemporal consistency of computing cores, memory, and network bandwidth allocation. While ensuring real-time response to high-priority power services, it optimizes resource fragmentation, thereby achieving a dual improvement in computing resource utilization efficiency and service operation reliability of power wireless terminals without manual intervention. Attached Figure Description

[0007] Figure 1 This is a flowchart of a dynamic configuration computing resource management method based on a power wireless terminal provided in an embodiment of the present invention.

[0008] Figure 2 This is a schematic diagram of the composition of a computer system provided in an embodiment of the present invention. Detailed Implementation

[0009] Please see Figure 1 , Figure 1 The flowchart illustrates a dynamic configuration computing resource management method based on a power wireless terminal, provided in an embodiment of the present invention. This method can be executed by a computer system and includes the following steps:

[0010] Step S100: Obtain the real-time operating status data set and the power service demand data set of the power wireless terminal. The real-time operating status data set includes the terminal device's operating parameters and network load fluctuation characteristics, while the power service demand data set includes service priority identifiers and real-time task processing requirements.

[0011] The real-time operational status dataset is a collection of data reflecting the current operational status of the power wireless terminal. Among these, the terminal device's operating parameters are various parameters related to the terminal device's operation, such as CPU utilization, memory usage, and disk I / O rate. These parameters directly reflect the terminal device's operational load. Network load fluctuation characteristics describe the characteristics of network load changes over different time periods, such as the curve of network bandwidth utilization over time and fluctuations in network latency. The power service demand dataset is a collection of demand information related to power services. Service priority indicators are used to distinguish the importance of different power services. For example, critical services ensuring the safe and stable operation of the power grid can have higher priority, while auxiliary services have relatively lower priority. Real-time task processing requirements specify the specific requirements for each power service task in terms of processing time, resource consumption, etc., such as a task requiring completion within a set time or limitations on the use of computing cores and memory.

[0012] When acquiring real-time operational status data sets and power service demand data sets from power wireless terminals, for the real-time operational status data sets, a monitoring program can be installed on the terminal device. This program collects the terminal device's operating parameters at preset time intervals and transmits them to the data processing center via the network. For network load fluctuation characteristics, network monitoring equipment, such as a network traffic analyzer, can be used to monitor and record network bandwidth usage, latency, and other data in real time. For the power service demand data sets, interfaces can be established with various power service systems to obtain service priority identifiers and real-time task processing requirements from these systems. For example, the power dispatching system can transmit the priority and processing requirements of its service tasks to the data processing center through an interface, thereby completing the collection of power service demand data.

[0013] As one implementation, before obtaining the real-time operating status data set and the power service demand data set of the power wireless terminal in step S100, the method provided in this embodiment of the invention may further include the following preparatory steps S101 to S105:

[0014] Step S101: Configure the monitoring agent program of the power wireless terminal. The monitoring agent program is used to collect the terminal device's operating parameters and network load data according to a preset period.

[0015] The monitoring agent is a software program installed on the power wireless terminal to periodically collect the terminal device's operating parameters and network load data. The preset period is a pre-defined time interval for data collection, such as every minute, every five minutes, or every ten minutes. Terminal device operating parameters include the aforementioned CPU utilization, memory usage, and disk I / O rate, which reflect the terminal device's operating status. Network load data includes network bandwidth usage, network latency, and packet loss rate, which reflect the network's operational status.

[0016] When configuring the monitoring agent, for example, a monitoring agent can be developed using programming languages ​​such as Python or Java. This agent utilizes the API interfaces provided by the operating system to obtain the operating parameters and network load data of the terminal devices. Then, the developed monitoring agent is deployed to the power wireless terminal, either through a remote deployment tool or by manual installation. Finally, a preset period is set, which can be achieved by modifying the monitoring agent's configuration file. For example, setting the collection interval to 5 minutes in the configuration file will allow the monitoring agent to collect data periodically according to this cycle. For instance, in a power system with multiple power wireless terminals, a unified deployment tool can be used to deploy the monitoring agent to each terminal, setting the same preset period, thus achieving unified data collection from all terminal devices.

[0017] Step S102: Establish a standardized interface for business requirement data to convert task requests output by different power business systems into power business requirement data in a unified format.

[0018] The standardized business requirement data interface is used to uniformly process task requests output by different power business systems. Different power business systems may use different data formats and protocols to output task requests, which poses challenges to unified data processing. By establishing a standardized interface, these differently formatted task requests can be converted into a unified format of power business requirement data, facilitating subsequent processing and analysis.

[0019] Establishing a standardized interface for business requirement data can be achieved using the following technical methods. First, analyze the data format and protocol of task requests output by different power business systems to determine the fields and data types that need to be standardized. Then, develop an interface program that can use web service technologies, such as RESTful APIs, to communicate with different power business systems. After receiving task requests from different business systems, the interface program parses and transforms them into power business requirement data in a standardized format.

[0020] Step S103: Clean the historical resource allocation records, remove invalid data entries and fill in the missing terminal device identifier field.

[0021] Historical resource allocation records document the resource allocation status of power grid wireless terminals over a past period. These records may contain invalid data entries and missing terminal device identification fields. Invalid data entries may be due to data acquisition errors, transmission failures, or other reasons, and these data will affect subsequent data analysis and processing results. Missing terminal device identification fields will prevent accurate identification of the terminal device corresponding to each resource allocation record, thus affecting the formulation of resource allocation strategies.

[0022] Step S104: Build an initial version of the resource adaptation rule base. The initial version includes default load balancing rules and conflict avoidance rules generated based on historical data statistics.

[0023] The resource adaptation rule base is a set of rules used to guide the allocation of computing resources. The initial version of the resource adaptation rule base, built in the initial stage, includes default load balancing rules and conflict avoidance rules. The default load balancing rules are generated based on historical data statistics and are used to balance the load of various terminal devices, avoiding situations where some terminal devices are overloaded while others are underloaded. Conflict avoidance rules are used to prevent conflicts during resource allocation, such as multiple tasks competing for the same resource simultaneously.

[0024] When building the initial version of the resource adaptation rule base, historical resource allocation data is collected, including the load status of terminal devices and task allocation. Then, this historical data is statistically analyzed to identify load balancing patterns and the causes of conflicts. Based on these analysis results, default load balancing rules and conflict avoidance rules are formulated. For conflict avoidance rules, for example, if it is found that conflicts easily occur when multiple high-priority tasks simultaneously request the same network bandwidth, rules can be formulated to allocate network bandwidth sequentially according to task priority and time order.

[0025] Step S105: Complete the connectivity test of the monitoring agent program and standardized interface to ensure the synchronous collection of real-time operating status data set and power business demand data set.

[0026] Connectivity testing verifies the integrity of the connection between the monitoring agent and the standardized interface. Only by ensuring a stable connection between the monitoring agent and the standardized interface can the synchronous acquisition of real-time operational status data and power business demand data be guaranteed. Synchronous acquisition of these two data sets means collecting them at the same point in time; this ensures the accuracy of subsequent data analysis and processing results.

[0027] The following methods can be used to complete the connectivity testing of the monitoring agent program and standardized interface. First, write a connectivity test program that sends a test request to the monitoring agent program. Upon receiving the request, the monitoring agent program returns a response. If the standardized interface can successfully receive the response from the monitoring agent program, the connection is considered normal. Simultaneously, during the testing process, it is essential to ensure that the timestamps of the collected real-time operational status data set and the power business demand data set are consistent to guarantee synchronized data collection.

[0028] Step S200: Perform multi-dimensional feature fusion processing on the real-time operation status data set and the power business demand data set to generate a resource adaptation feature set. The resource adaptation feature set includes terminal equipment load balancing features, business demand conflict features, and resource allocation efficiency evaluation indicators.

[0029] Multi-dimensional feature fusion processing integrates multiple features from real-time operational status data sets and power business demand data sets to extract more valuable information. These two sets contain information from different aspects; multi-dimensional feature fusion processing combines this information to generate a more comprehensive set of resource adaptation features. Terminal device load balancing features describe the load balance among terminal devices, such as the distribution of CPU utilization and memory usage across different devices. Business demand conflict features identify potential conflicts during resource allocation, such as multiple tasks simultaneously requesting the same resource. Resource allocation efficiency evaluation metrics assess the effectiveness of resource allocation, such as task completion rate and resource utilization.

[0030] The following steps can be used to perform multi-dimensional feature fusion processing on real-time operational status data sets and power business demand data sets. First, both sets are preprocessed, including data cleaning and normalization, to ensure data quality and consistency. Then, a suitable feature fusion algorithm, such as Principal Component Analysis (PCA) or Linear Discriminant Analysis (LDA), is selected to fuse features from the two datasets. Finally, based on the fused features, terminal equipment load balancing characteristics, business demand conflict characteristics, and resource allocation efficiency evaluation indicators are generated.

[0031] As one implementation method, step S200 involves performing multi-dimensional feature fusion processing on the real-time operating status data set and the power business demand data set to generate a resource adaptation feature set. This may specifically include the following steps S210 to S250:

[0032] Step S210: Extract time series features of terminal device operating parameters from the real-time operating status data set. The time series features include parameter fluctuation period, peak duration, and abnormal parameter offset.

[0033] Time series features are the characteristics of data changes over a period of time. Extracting time series features of terminal device operating parameters from real-time operational status datasets can provide a better understanding of the changing patterns of terminal device operating status. Parameter fluctuation period is the length of the period during which terminal device operating parameters (such as CPU utilization, memory usage, etc.) exhibit periodic changes over a period of time. Peak duration is the length of time a parameter remains after reaching its peak value. Abnormal parameter offset is the degree to which a parameter deviates from its normal range at a certain point in time.

[0034] The following methods can be used to extract time-series features of terminal device operating parameters from real-time operational status datasets. First, perform time-series analysis on the terminal device operating parameters in the real-time operational status dataset, for example, using an Autoregressive Integral Moving Average (ARIMA) model or seasonal decomposition. Then, calculate the parameter fluctuation period, peak duration, and outlier parameter offsets based on the analysis results. For example, for CPU utilization data, fit the data using an ARIMA model to obtain the trend of CPU utilization changes, and then calculate the parameter fluctuation period, peak duration, and outlier parameter offsets based on this trend.

[0035] Step S220: Decompose the real-time task processing requirements in the power business demand data set to obtain multiple sub-task resource requirement characteristics, including processing latency constraints, computing core occupancy rate, and memory allocation threshold.

[0036] Task decomposition involves breaking down a complex task into multiple simpler subtasks to facilitate better resource allocation and management. Real-time task processing requirements in power business demand datasets typically involve one or more tasks, each potentially requiring different resources. Task decomposition breaks these tasks down into subtasks and identifies the resource requirements of each subtask. Processing latency constraints define the time required for each subtask to complete. Computational core occupancy represents the proportion of computational cores required by the subtask during execution. Memory allocation thresholds define the maximum amount of memory allowed for a subtask during execution.

[0037] The following steps can be used to decompose real-time task processing requirements from a power business demand dataset. First, analyze the real-time task processing requirements to determine the logical structure and dependencies of the tasks. Then, decompose the tasks into multiple sub-tasks based on their logical structure and dependencies. Finally, determine processing latency constraints, computational core utilization, and memory allocation thresholds for each sub-task. For example, a power business task might be power load forecasting, which can be decomposed into sub-tasks such as data acquisition, data preprocessing, model training, and prediction result output.

[0038] Step S230: Perform correlation matching processing on the time series features and the resource requirement features of the sub-tasks to generate a task-device adaptability index. The task-device adaptability index is used to quantify the degree of matching between the processing capabilities of the terminal device and the requirements of the sub-tasks.

[0039] Correlation matching involves comparing and analyzing time-series features and sub-task resource requirement features to identify the correlations between them. The task-device fit index is a quantitative metric used to measure the degree of matching between the processing capabilities of the terminal device and the resource requirements of the sub-task. By generating the task-device fit index, a basis for resource allocation can be provided, assigning sub-tasks to the most suitable terminal devices.

[0040] The following steps can be used to correlate and match time series features with subtask resource requirement features. First, standardize both the time series features and the subtask resource requirement features to eliminate differences in units of measurement. Then, select a matching algorithm, such as Euclidean distance or cosine similarity, to calculate the similarity between the time series features and the subtask resource requirement features. Finally, generate a task-device fit index based on the similarity. For example, the Euclidean distance algorithm can be used to calculate the Euclidean distance between the CPU utilization time series feature of a terminal device and the computational core utilization requirement feature of a subtask. The smaller the distance, the higher the degree of matching, and the task-device fit index can be generated based on this distance.

[0041] As one implementation method, step S230 involves performing correlation matching processing between the time series features and the resource requirement features of the sub-tasks to generate a task-equipment adaptability index, which may specifically include the following steps S231 to S236:

[0042] Step S231: Extract the duration distribution characteristics of parameter fluctuation period and the equipment load state change curve corresponding to the peak duration from the time series features.

[0043] The duration distribution characteristics of parameter fluctuation cycles refer to the distribution of the duration of parameter fluctuation cycles over a period of time, such as the frequency of parameter fluctuation cycles of different durations. The equipment load state change curve corresponding to the peak duration is the curve of the equipment load state changing over time when the parameter reaches its peak. By extracting these features, we can better understand the load change patterns of terminal devices.

[0044] The following method can be used to extract the duration distribution characteristics of parameter fluctuation periods and the equipment load state change curve corresponding to the peak duration from time series features. First, perform periodic analysis on the time series features, such as using Fourier transform or wavelet transform, to determine the duration of the parameter fluctuation period. Then, statistically analyze the frequency of parameter fluctuation periods of different durations to obtain the duration distribution characteristics of the parameter fluctuation periods. For the equipment load state change curve corresponding to the peak duration, the time point when the parameter reaches its peak can be found in the time series features, and then the equipment load state data for a period of time before and after that time point can be extracted to plot the load state change curve.

[0045] Step S232: Perform task execution window partitioning on the processing latency constraints in the resource requirement characteristics of subtasks to generate the minimum time window requirement and the maximum time window tolerance threshold for each subtask.

[0046] The task execution window segmentation process divides the task execution time into different windows based on the processing latency constraints of subtasks. The minimum time window requirement is the shortest time that a subtask must complete within this window. The maximum time window tolerance threshold is the longest execution time that a subtask can tolerate. By generating the minimum time window requirement and the maximum time window tolerance threshold for each subtask, the execution time of subtasks can be better scheduled.

[0047] The following steps can be used to divide the task execution window based on the processing latency constraints in the resource requirements of subtasks. First, analyze the processing latency constraints of the subtasks to determine the start and end times of the tasks. Then, based on the importance and urgency of the tasks, divide the task execution time into a minimum time window and a maximum time window.

[0048] Step S233: Align and match the equipment load status change curve with the minimum time window requirement on the time axis to identify the set of available time periods in the equipment load status curve that meet the minimum time window requirement of the subtask.

[0049] Time axis alignment and matching involves aligning the device load state change curve with the minimum time window requirement on the time axis to identify the time periods in the device load state curve that satisfy the minimum time window requirement. The available time period set is the set of all time periods in the device load state curve that satisfy the minimum time window requirement of the subtask. By identifying the available time period set, it can be determined which time periods the terminal device has sufficient resources to execute the subtask.

[0050] Aligning device load status change curves with minimum time window requirements on the time axis can be achieved using the following method. First, align the device load status change curves and minimum time window requirements on the time axis, ensuring their starting points and time units are consistent. Then, iterate through the device load status curves to find time periods that meet the minimum time window requirements. For example, for the CPU utilization load status change curve of a terminal device and the minimum time window requirement (30 minutes) of a subtask, after aligning them on the time axis, iterate through the CPU utilization load status curves to find time periods where CPU utilization is below a certain threshold (e.g., 70%) and lasts for at least 30 minutes. These time periods constitute the set of available time periods.

[0051] Step S234: Based on the continuous distribution characteristics of the available time period set and the interval parameter of the peak duration, calculate the dynamic matching coefficient of each subtask within the corresponding available time period. The dynamic matching coefficient is used to quantify the fit between the equipment load fluctuation and the time window requirements of the subtask.

[0052] The dynamic matching coefficient is a quantitative metric used to measure the fit between device load fluctuations and subtask time window requirements. The continuous distribution characteristic of the available time period set refers to the distribution of available time periods along the time axis, such as whether the available time periods are continuous or discrete. The peak duration interval parameter is the time interval between two adjacent peak durations. By calculating the dynamic matching coefficient, the suitability of terminal devices to execute subtasks within available time periods can be better evaluated.

[0053] The following steps can be used to calculate the dynamic matching coefficient for each subtask within its corresponding available time period, based on the continuous distribution characteristics of the available time period set and the interval parameter of peak duration. First, analyze the continuous distribution characteristics of the available time period set, such as calculating the average length of the available time periods and the number of consecutive available time periods. Then, combine the interval parameter of peak duration to determine the stability of equipment load fluctuations. Finally, calculate the dynamic matching coefficient based on the stability of equipment load fluctuations and the time window requirements of the subtask. For example, for a subtask and a terminal device's available time period set, if the available time periods are consecutive and have a long average length, and the interval parameter of peak duration is large, it indicates that the equipment load fluctuations are relatively stable, and therefore the dynamic matching coefficient is high.

[0054] Step S235: Normalize and correct the dynamic matching coefficients according to the maximum time window tolerance threshold to generate a set of corrected dynamic matching coefficients.

[0055] The normalization correction process adjusts the dynamic matching coefficients to a reasonable range. The maximum time window tolerance threshold is the longest execution time that a subtask can tolerate. By normalizing the dynamic matching coefficients based on the maximum time window tolerance threshold, it can be ensured that the dynamic matching coefficients accurately reflect the fit between equipment load fluctuations and the time window requirements of subtasks, while also taking into account the time constraints of subtasks.

[0056] The following method can be used to normalize and correct the dynamic matching coefficients based on the maximum time window tolerance threshold. First, determine the normalization range, for example, normalize the dynamic matching coefficients to the [0,1] interval. Then, adjust the dynamic matching coefficients according to the maximum time window tolerance threshold and the actual time window requirements of the subtask. For example, if the actual time window requirements of the subtask are close to the maximum time window tolerance threshold, the dynamic matching coefficients may need to be appropriately reduced. Assuming the dynamic matching coefficient is 0.8, the maximum time window tolerance threshold is 90 minutes, and the actual time window requirements of the subtask are 80 minutes, the dynamic matching coefficients are corrected to 0.7 according to the preset correction rules, resulting in the corrected set of dynamic matching coefficients.

[0057] Step S236: Combine the modified dynamic matching coefficient set with the computational core occupancy rate and memory allocation threshold in the resource requirement characteristics of the subtask, perform multi-dimensional weighted fusion processing, and generate a task-device adaptation index associated with each subtask and terminal device.

[0058] Multi-dimensional weighted fusion processing comprehensively processes the corrected dynamic matching coefficient set, computational core utilization, and memory allocation threshold, considering their respective weights to generate a comprehensive metric. The task-device adaptability metric is associated with each subtask and terminal device, measuring the degree of matching between the terminal device and the subtask. Through multi-dimensional weighted fusion processing, the resource requirements of subtasks and the load of terminal devices can be considered more comprehensively.

[0059] The process of generating a task-device compatibility index associated with each subtask and terminal device by combining the revised dynamic matching coefficient set with the computational core utilization rate and memory allocation threshold from the subtask resource requirement characteristics and performing multi-dimensional weighted fusion processing can be carried out using the following steps: First, assign weights to the revised dynamic matching coefficient set, computational core utilization rate, and memory allocation threshold, for example, weights of 0.5, 0.3, and 0.2 respectively. Then, sum the revised dynamic matching coefficient set, computational core utilization rate, and memory allocation threshold using weighted summation to obtain the task-device compatibility index.

[0060] Step S240: Based on the preset resource conflict detection model, perform conflict analysis on network load fluctuation characteristics and service priority identifiers to generate resource competition hotspot area identifiers and conflict mitigation suggestion parameters.

[0061] The pre-built resource conflict detection model is a pre-established model used to detect potential conflicts during resource allocation. For example, the resource conflict detection model can be a decision tree model. Its training data can be obtained by periodically extracting data from various data sources (such as network monitoring systems and power business systems) using automated data acquisition tools and storing it in a data warehouse. During training, common feature engineering techniques can be used to perform data cleaning, selection, transformation, and combination, and to divide the data into training and test sets. The initial resource conflict detection model is then trained until it converges.

[0062] Network load fluctuation characteristics describe the changes in network load over different time periods, while service priority indicators distinguish the importance of different services. Conflict analysis involves analyzing network load fluctuation characteristics and service priority indicators to identify areas and situations where resource contention may occur. Resource contention hotspot area indicators identify areas where resource contention may occur, and conflict mitigation suggestion parameters are parameters used to mitigate resource contention conflicts.

[0063] Based on a pre-defined resource conflict detection model, conflict analysis is performed on network load fluctuation characteristics and service priority identifiers to generate resource contention hotspot area identifiers and conflict mitigation suggestion parameters. The steps are as follows: First, the network load fluctuation characteristics and service priority identifiers are input into the pre-defined resource conflict detection model. This model can use machine learning algorithms, such as decision tree algorithms and support vector machine algorithms, to analyze and process the input data. Based on the network load fluctuation characteristics and service priority identifiers, the model identifies time periods and regions where resource contention may occur, generating resource contention hotspot area identifiers. Then, based on the conflict analysis results, conflict mitigation suggestion parameters are generated. For example, for a power network, the pre-defined resource conflict detection model, by analyzing network load fluctuation characteristics and service priority identifiers, discovers that multiple high-priority services simultaneously request the same network bandwidth within a certain time period. This time period and the corresponding network region are identified as resource contention hotspot areas, and conflict mitigation suggestion parameters for adjusting the network bandwidth allocation ratio are generated.

[0064] As one implementation method, step S240, based on a preset resource conflict detection model, performs conflict analysis processing on network load fluctuation characteristics and service priority identifiers to generate resource contention hotspot area identifiers and conflict mitigation suggestion parameters, which may specifically include the following steps S241 to S246:

[0065] Step S241: Based on the bandwidth utilization change curve in the network load fluctuation characteristics, determine the network resource saturation time period and idle time period, and associate them with the corresponding terminal device identifier.

[0066] The bandwidth utilization curve in the network load fluctuation characteristics describes how network bandwidth utilization changes over time. The network resource saturation period is the time when network bandwidth utilization reaches or approaches its maximum value; during this period, network resources are scarce and resource contention is likely to occur. The idle period is the time when network bandwidth utilization is low; during this period, network resources are relatively abundant. The terminal device identifier is used to uniquely identify each terminal device.

[0067] The following method can be used to determine network resource saturation and idle periods based on the bandwidth utilization change curve in network load fluctuation characteristics and associate them with the corresponding terminal device identifiers. First, analyze the bandwidth utilization change curve and set a bandwidth utilization threshold, such as 80%. When the bandwidth utilization exceeds this threshold, the corresponding time period is the network resource saturation period; when the bandwidth utilization is below a certain lower threshold, such as 20%, the corresponding time period is the idle period. Then, record the terminal device identifiers using network resources in each time period using network monitoring equipment, and associate the network resource saturation and idle periods with the corresponding terminal device identifiers.

[0068] Step S242: Extract the urgent task marker and task dependency chain from the business priority identifier to generate a task execution order constraint set.

[0069] The business priority identifier includes an urgent task marker and a task dependency chain. The urgent task marker identifies which tasks are urgent and require priority processing. The task dependency chain describes the dependencies between tasks, meaning that the execution of one task may depend on the completion of other tasks. The task execution order constraint set is a set that specifies the order in which tasks are executed, generated based on the urgent task marker and the task dependency chain.

[0070] Extracting the urgent task markers and task dependency chains from the business priority identifiers to generate a task execution order constraint set can be achieved using the following steps: First, extract the urgent task markers and task dependency chains from the business priority identifiers. Then, prioritize urgent tasks based on their urgent task markers. For tasks with dependencies, determine their execution order based on the task dependency chains.

[0071] Step S243: Perform spatiotemporal overlap analysis on the network resource saturation time period and the task execution order constraint set to identify the resource competition conflict period and the set of affected tasks.

[0072] Spatiotemporal overlap analysis compares and analyzes the network resource saturation period and the task execution order constraint set in time and space to identify the overlapping parts. The resource contention conflict period is the time period within the network resource saturation period when multiple tasks simultaneously request the same resource. The affected task set is the set of tasks affected during the resource contention conflict period.

[0073] To identify resource contention conflict periods and the set of affected tasks by performing spatiotemporal overlap analysis between network resource saturation periods and task execution order constraints, the following method can be used: First, align the network resource saturation periods and the task execution order constraints on the timeline. Then, traverse the task execution order constraints to find the tasks that need to be executed within the network resource saturation period. For these tasks, if they simultaneously request the same resource, the corresponding time period is the resource contention conflict period, and these tasks constitute the set of affected tasks. For example, if the network resource saturation period is 10:00-11:00 AM, and the task execution order constraints are task A (9:30-10:30), task B (10:00-11:00), and task C (10:30-11:30), where task A and task B both need to use the same network bandwidth, then the resource contention conflict period is 10:00-10:30 AM, and the set of affected tasks is task A and task B.

[0074] Step S244: Based on the preset conflict resolution algorithm, simulate the network bandwidth allocation strategy during the resource contention conflict period to generate multiple candidate bandwidth allocation schemes.

[0075] The preset conflict resolution algorithm is a pre-designed algorithm used to resolve resource contention conflicts. The network bandwidth allocation strategy during the resource contention conflict period refers to how network bandwidth should be allocated during that period. Simulation processing involves testing and evaluating different network bandwidth allocation strategies through computer simulation. Candidate bandwidth allocation schemes are multiple possible network bandwidth allocation schemes obtained through simulation processing.

[0076] The following steps can be used to simulate network bandwidth allocation strategies during resource contention conflict periods, based on a pre-defined conflict resolution algorithm, to generate multiple candidate bandwidth allocation schemes. First, determine the pre-defined conflict resolution algorithm, such as a fair allocation algorithm or a priority allocation algorithm. Then, design different network bandwidth allocation strategies based on the network bandwidth resources and the needs of the affected task set during the resource contention conflict period. Input these network bandwidth allocation strategies into the pre-defined conflict resolution algorithm for simulation processing, record the execution results of each strategy, such as task completion rate and bandwidth utilization, and generate multiple candidate bandwidth allocation schemes based on the execution results.

[0077] As one implementation method, step S244 involves simulating the network bandwidth allocation strategy during the resource contention conflict period based on a preset conflict resolution algorithm, generating multiple candidate bandwidth allocation schemes. Specifically, this may include the following steps S2441 to S2446:

[0078] Step S2441: Decompose the tasks in the affected task set into independent execution units and dependent execution units according to the task execution order constraint set, and allocate an initial bandwidth occupancy ratio to each execution unit.

[0079] The tasks in the affected task set may contain multiple execution units, which can be divided into independent execution units and dependent execution units. Independent execution units are task units that can be executed independently without depending on the completion of other task units. Dependent execution units are task units that require the completion of other task units before they can be executed. The initial bandwidth usage ratio is the pre-allocated network bandwidth ratio for each execution unit.

[0080] The following steps can be used to decompose the tasks in the affected task set into independent execution units and dependent execution units according to the task execution order constraint set, and to allocate an initial bandwidth allocation ratio to each execution unit. First, based on the task execution order constraint set, analyze the logical structure and dependencies of each task in the affected task set, and decompose the tasks into independent execution units and dependent execution units. Then, based on the resource requirements and importance of each execution unit, allocate an initial bandwidth allocation ratio to each execution unit. For example, the affected task set contains task A and task B. Task A contains independent execution unit A1 and dependent execution unit A2, and task B contains independent execution unit B1. According to the task execution order constraint set, task A executes first, and task B executes later. Task A has higher resource requirements; therefore, allocate 40% of the initial bandwidth allocation ratio to independent execution unit A1, 30% to dependent execution unit A2, and 30% to independent execution unit B1.

[0081] Step S2442: Based on the duration of the resource contention conflict period and the bandwidth utilization rate change curve during the network resource saturation period, perform time-domain dynamic adjustment processing on the initial bandwidth utilization ratio to generate the first candidate bandwidth allocation scheme; wherein, the time-domain dynamic adjustment processing includes allocating an increasing bandwidth ratio to dependent execution units during the resource contention conflict period, and simultaneously allocating a decreasing bandwidth ratio to independent execution units.

[0082] The duration of the resource contention conflict period is the length of the time during which the resource contention conflict occurs. The bandwidth utilization rate change curve during the network resource saturation period describes how the bandwidth utilization rate changes over time within the network resource saturation period. The time-domain dynamic adjustment process adjusts the initial bandwidth utilization ratio based on changes over time. The first candidate bandwidth allocation scheme is a candidate bandwidth allocation scheme obtained through the time-domain dynamic adjustment process.

[0083] Based on the duration of resource contention conflict periods and the bandwidth utilization rate change curves during network resource saturation periods, the initial bandwidth utilization ratio can be dynamically adjusted in the time domain to generate the first candidate bandwidth allocation scheme. The following steps can be used: First, analyze the duration of resource contention conflict periods and the bandwidth utilization rate change curves during network resource saturation periods to determine the timing and magnitude of bandwidth adjustments. During the resource contention conflict period, as time progresses, allocate an increasing bandwidth ratio to dependent execution units and a decreasing bandwidth ratio to independent execution units.

[0084] Step S2443: Based on the task response timeliness requirements in the task dependency chain, perform priority weighting on the bandwidth allocation ratio of dependent execution units to generate a second candidate bandwidth allocation scheme; wherein, priority weighting includes allocating additional bandwidth redundancy to dependent execution units with strict response timeliness requirements.

[0085] The task response timeliness requirement in the task dependency chain refers to the requirement that the task be completed within a specified time. Priority-weighted processing assigns different weights to different dependent execution units based on the importance and timeliness requirements of the tasks. The second candidate bandwidth allocation scheme is a candidate bandwidth allocation scheme obtained by prioritizing and weighting the bandwidth allocation ratio of the dependent execution units.

[0086] Based on the task response timeliness requirements in the task dependency chain, the bandwidth allocation ratio of dependent execution units is prioritized and weighted to generate a second candidate bandwidth allocation scheme. The following steps can be used: First, analyze the task response timeliness requirements in the task dependency chain to determine the importance and timeliness level of each dependent execution unit. For dependent execution units with strict response timeliness requirements, allocate additional bandwidth redundancy. For example, if dependent execution unit A2 in the task dependency chain has strict response timeliness requirements, and its bandwidth ratio in the first candidate bandwidth allocation scheme is 40%, allocate an additional 10% bandwidth redundancy to it, adjusting its bandwidth ratio to 50%. Simultaneously, adjust the bandwidth ratios of other execution units accordingly to obtain the second candidate bandwidth allocation scheme.

[0087] Step S2444: Extract historical bandwidth adjustment records from the network bandwidth reallocation strategy, perform historical trend fitting processing on the initial bandwidth occupancy ratio, and generate a third candidate bandwidth allocation scheme; wherein, the historical trend fitting processing includes matching the bandwidth allocation pattern of similar conflict periods and copying it to the current resource contention conflict period.

[0088] The historical bandwidth adjustment records in the network bandwidth reallocation strategy are records of bandwidth adjustments made during past periods of resource contention conflict. Historical trend fitting involves analyzing these historical bandwidth adjustment records to identify bandwidth allocation patterns during similar conflict periods and applying them to the current period of resource contention conflict. The third candidate bandwidth allocation scheme is a candidate bandwidth allocation scheme obtained through historical trend fitting.

[0089] The steps to extract historical bandwidth adjustment records from network bandwidth reallocation strategies, perform historical trend fitting on the initial bandwidth occupancy ratio, and generate a third candidate bandwidth allocation scheme are as follows: First, extract historical bandwidth adjustment records from the network bandwidth reallocation strategy, including information such as conflict periods and bandwidth allocation ratios. Then, analyze the characteristics of the current resource contention conflict period, such as duration and network bandwidth resources, and match bandwidth allocation patterns for similar conflict periods in the historical bandwidth adjustment records. Finally, copy the matched bandwidth allocation pattern to the current resource contention conflict period, adjust the initial bandwidth occupancy ratio, and generate a third candidate bandwidth allocation scheme. For example, by analyzing historical bandwidth adjustment records, a past conflict period similar to the current resource contention conflict period was found, with a bandwidth allocation pattern of 30% for independent execution unit A1, 50% for dependent execution unit A2, and 20% for independent execution unit B1. Applying this bandwidth allocation pattern to the current resource contention conflict period and adjusting the initial bandwidth occupancy ratio yields a third candidate bandwidth allocation scheme.

[0090] Step S2445: Load the network load fluctuation data and the set of affected tasks during the resource contention conflict period in the simulated execution environment, inject the first candidate bandwidth allocation scheme, the second candidate bandwidth allocation scheme and the third candidate bandwidth allocation scheme respectively, execute the simulated allocation operation and record the task processing latency change data.

[0091] The simulation execution environment is an environment used to simulate the resource allocation process, which can be implemented using computer software. Network load fluctuation data during resource contention periods describes the changes in network load during these periods. The affected task set is the set of tasks affected during the resource contention period. The simulated allocation operation is the process of allocating network bandwidth in the simulation execution environment according to different candidate bandwidth allocation schemes. Task processing latency variation data shows the changes in task processing latency during the simulated allocation operation.

[0092] To simulate network load fluctuations and the set of affected tasks during resource contention periods in a simulated execution environment, the following steps can be taken: First, construct a simulated execution environment that can simulate network load fluctuations and task execution. Then, load the network load fluctuation data and the set of affected tasks during the resource contention period into the simulated execution environment. Next, inject the first, second, and third candidate bandwidth allocation schemes into the simulated execution environment and execute the simulated allocation operation. During the simulated allocation operation, record the processing latency changes for each task, such as task start time and task end time.

[0093] Step S2446: Based on the task completion rate deviation and bandwidth utilization fluctuation range in the task processing latency change data, select a set of candidate bandwidth allocation schemes that meet the preset conflict resolution conditions.

[0094] The task completion rate deviation in the task processing latency variation data represents the difference between the actual task completion rate and the expected task completion rate. The bandwidth utilization fluctuation range refers to the range of bandwidth utilization changes during the simulated allocation operation. Preset conflict resolution conditions are pre-defined criteria used to screen candidate bandwidth allocation schemes, such as task completion rate deviation being less than a certain threshold or bandwidth utilization fluctuation range being within a certain range. The candidate bandwidth allocation scheme set is the set of candidate bandwidth allocation schemes that meet the preset conflict resolution conditions.

[0095] The following steps can be used to select a set of candidate bandwidth allocation schemes that meet preset conflict resolution conditions based on the task completion rate deviation and bandwidth utilization fluctuation range in the task processing latency variation data. First, extract the task completion rate deviation and bandwidth utilization fluctuation range from the task processing latency variation data. Then, based on the preset conflict resolution conditions, screen the first, second, and third candidate bandwidth allocation schemes. For example, the preset conflict resolution conditions are: task completion rate deviation less than 5%, and bandwidth utilization fluctuation range between 10% and 20%. Analysis of the task processing latency variation data reveals that the first candidate bandwidth allocation scheme has a task completion rate deviation of 3% and a bandwidth utilization fluctuation range of 15%; the second candidate bandwidth allocation scheme has a task completion rate deviation of 6% and a bandwidth utilization fluctuation range of 25%; and the third candidate bandwidth allocation scheme has a task completion rate deviation of 2% and a bandwidth utilization fluctuation range of 12%. Therefore, the set of candidate bandwidth allocation schemes that meet the preset conflict resolution conditions consists of the first and third candidate bandwidth allocation schemes.

[0096] Step S245: Based on the impact of candidate bandwidth allocation schemes on the processing latency of the affected task set, select the target bandwidth allocation scheme with the smallest latency increment, and extract its bandwidth adjustment parameters as conflict mitigation suggestion parameters.

[0097] The impact of candidate bandwidth allocation schemes on the processing latency of the affected task set refers to the change in processing latency when different candidate bandwidth allocation schemes are adopted. The latency increment is the increase in task processing latency relative to the original latency after adopting a certain candidate bandwidth allocation scheme. The target bandwidth allocation scheme is the candidate bandwidth allocation scheme with the smallest latency increment. Conflict mitigation recommendation parameters are the bandwidth adjustment parameters in the target bandwidth allocation scheme, used to mitigate resource contention conflicts.

[0098] The following steps can be used to select the target bandwidth allocation scheme with the smallest latency increment based on the impact of candidate bandwidth allocation schemes on the processing latency of the affected task set, and to extract its bandwidth adjustment parameters as conflict mitigation suggestion parameters. First, analyze the impact of each scheme in the candidate bandwidth allocation scheme set on the processing latency of the affected task set, and calculate the latency increment for each scheme. Then, compare the latency increments of each scheme and select the target bandwidth allocation scheme with the smallest latency increment. Finally, extract bandwidth adjustment parameters from the target bandwidth allocation scheme, such as the bandwidth allocation ratio of each execution unit, as conflict mitigation suggestion parameters. For example, if the candidate bandwidth allocation scheme set consists of the first candidate bandwidth allocation scheme and the third candidate bandwidth allocation scheme, and analysis shows that the latency increment of the first candidate bandwidth allocation scheme is 5 minutes and the latency increment of the third candidate bandwidth allocation scheme is 3 minutes, then the target bandwidth allocation scheme is the third candidate bandwidth allocation scheme. Extract its bandwidth adjustment parameters, such as 30% for independent execution unit A1, 50% for dependent execution unit A2, and 20% for independent execution unit B1, as conflict mitigation suggestion parameters.

[0099] Step S246: Encapsulate the resource competition conflict period and the corresponding terminal device identifier into a resource competition hotspot area identifier.

[0100] A resource contention conflict period is a time when multiple tasks simultaneously request the same resource within a network resource saturation period. A terminal device identifier is used to uniquely identify each terminal device. A resource contention hotspot area identifier is used to identify areas where resource contention may occur; it is obtained by encapsulating the resource contention conflict period and the corresponding terminal device identifier together. The following method can be used to encapsulate the resource contention conflict period and the corresponding terminal device identifier into a resource contention hotspot area identifier: First, determine the start and end times of the resource contention conflict period, as well as the terminal device identifiers using network resources within that period. Then, encapsulate this information according to a predefined format, such as JSON.

[0101] Step S250: Combine the task-device compatibility index, resource competition hotspot area identifier, and conflict mitigation suggestion parameters to construct a resource compatibility feature set and associate it with the terminal device identifier and business task identifier.

[0102] The task-device compatibility metric quantifies the match between the processing capabilities of terminal devices and the requirements of subtasks. Resource contention hotspot area identifiers identify areas where resource contention may occur. Conflict mitigation suggestion parameters mitigate resource contention conflicts. The resource compatibility feature set is a collection containing information such as the task-device compatibility metric, resource contention hotspot area identifiers, and conflict mitigation suggestion parameters, used to guide resource allocation. Terminal device identifiers uniquely identify each terminal device, and business task identifiers uniquely identify each business task.

[0103] The following steps can be used to construct a resource adaptation feature set by integrating task-device compatibility metrics, resource contention hotspot area identifiers, and conflict mitigation suggestion parameters, and then associating it with terminal device identifiers and business task identifiers. First, integrate the task-device compatibility metrics, resource contention hotspot area identifiers, and conflict mitigation suggestion parameters to form a set containing multiple pieces of information. Then, associate this set with the terminal device identifier and business task identifier, so that each terminal device and each business task has corresponding resource adaptation features. For example, for terminal device A and business task 1, the task-device compatibility metric is 80%, the resource contention hotspot area identifier is {"start_time":"10:00","end_time":"11:00","terminal_devices":["A"]}, and the conflict mitigation suggestion parameters are 30% for independent execution units and 50% for dependent execution units. Integrate this information into a resource adaptation feature set and associate it with terminal device A and business task 1.

[0104] Step S300: Based on the preset resource adaptation rule base, perform real-time policy matching processing on the resource adaptation feature set to generate a computing resource configuration policy set, which includes resource allocation priority adjustment policy, task scheduling optimization policy and network bandwidth reallocation policy.

[0105] The preset resource adaptation rule base is a pre-established set of rules used to guide resource allocation. The resource adaptation feature set includes information such as task-device compatibility indicators, resource contention hotspot area identifiers, and conflict mitigation suggestion parameters. Real-time policy matching processing analyzes and matches the resource adaptation feature set according to the preset resource adaptation rule base to find suitable resource allocation strategies. The computing resource configuration strategy set is a collection of various strategies, including resource allocation priority adjustment strategies, task scheduling optimization strategies, and network bandwidth reallocation strategies, used to configure the computing resources of the power wireless terminal.

[0106] Based on a pre-defined resource adaptation rule base, real-time policy matching processing is performed on the resource adaptation feature set to generate a set of computing resource allocation strategies. The steps are as follows: First, the resource adaptation feature set is input into the pre-defined resource adaptation rule base. Rules in the rule base can be in the form of an if-then statement, such as "if task-device compatibility index > 80% and resource contention hotspot area identifier is empty then adopt resource allocation priority adjustment strategy A". Then, the resource adaptation feature set is matched according to the rules in the rule base to find suitable resource allocation strategies. Finally, these resource allocation strategies are integrated into a set of computing resource allocation strategies.

[0107] As one implementation method, step S300 involves performing real-time policy matching processing on the resource adaptation feature set based on a preset resource adaptation rule base to generate a set of computing resource configuration policies. This may specifically include the following steps S310 to S360:

[0108] Step S310: Extract the historical trend of the load balancing characteristics of the terminal device from the resource adaptation feature set and match it to the load balancing rule group in the resource adaptation rule base.

[0109] The historical trend of terminal device load balancing characteristics refers to the changes in terminal device load balancing characteristics over a period of time, such as changes in terminal device CPU utilization and memory usage. The load balancing rule group in the resource adaptation rule base is a set of rules used to balance the load on terminal devices.

[0110] Extracting historical trends of terminal device load balancing characteristics from the resource adaptation feature set and matching them to load balancing rule groups in the resource adaptation rule base can be achieved through the following steps: First, extract historical data on terminal device load balancing characteristics from the resource adaptation feature set, such as CPU utilization data of terminal devices over the past week. Then, analyze this historical data to identify trends, such as whether it shows an upward trend, a downward trend, or periodic changes. Finally, match these trends with load balancing rule groups in the resource adaptation rule base to find suitable rules. For example, if analysis reveals that the CPU utilization of a certain terminal device has shown an upward trend over the past week, and there is a rule in the load balancing rule group of the resource adaptation rule base that states "if the CPU utilization of a terminal device shows an upward trend, then distribute some tasks to other terminal devices with lower loads," then this rule will be matched.

[0111] Step S320: Based on the number of task processing conflicts and the frequency of resource preemption in the business requirement conflict characteristics, activate the conflict avoidance rule group in the resource adaptation rule base.

[0112] The task processing conflict count in the business requirement conflict characteristics refers to the number of times task processing conflicts occur during resource allocation. The resource preemption frequency refers to the frequency at which resources are preempted during resource allocation. The conflict avoidance rule group in the resource adaptation rule base is a set of rules used to avoid resource allocation conflicts.

[0113] Activating conflict avoidance rule groups in the resource adaptation rule base based on the number of task processing conflicts and resource preemption frequency in the business requirement conflict characteristics can be achieved through the following steps: First, extract the number of task processing conflicts and resource preemption frequency from the resource adaptation feature set. Then, set thresholds for the number of task processing conflicts and resource preemption frequency. When the number of task processing conflicts or the resource preemption frequency exceeds the threshold, activate the conflict avoidance rule group in the resource adaptation rule base.

[0114] Step S330: Call the resource utilization rate curve and task completion rate index in the resource allocation efficiency evaluation index, perform weight allocation processing on the load balancing rule group and conflict avoidance rule group, and generate rule combination priority ranking.

[0115] The resource utilization rate curve in the resource allocation efficiency evaluation index describes how resource utilization rate changes over time. The task completion rate index is the ratio of the number of tasks completed within a preset time to the total number of tasks. The load balancing rule group and the conflict avoidance rule group are two sets of rules in the resource adaptation rule base. Weight allocation processing assigns different weights to the load balancing rule group and the conflict avoidance rule group to determine their importance in resource allocation. Rule combination priority ranking sorts the load balancing rule group and the conflict avoidance rule group according to the weight allocation results to determine their execution order.

[0116] The following steps can be used to prioritize load balancing and conflict avoidance rule groups by invoking the resource utilization curve and task completion rate metric from the resource allocation performance evaluation system. First, analyze the resource utilization curve and task completion rate metric to determine their relative importance. For example, if resource utilization has a significant impact on system performance, its weight can be set higher; similarly, if task completion rate has a significant impact on business objectives, its weight can be set higher. Then, assign weights to the load balancing and conflict avoidance rule groups based on their respective weights. Finally, prioritize the load balancing and conflict avoidance rule groups based on their weight allocation results, placing rule groups with higher weights at the top.

[0117] Step S340: Based on the priority ranking of rule combinations, perform joint optimization calculations on the load balancing characteristics of terminal devices and the conflict characteristics of business demands to generate an initial version of the resource allocation priority adjustment strategy.

[0118] The rule combination priority ranking determines the execution order of the load balancing rule group and the conflict avoidance rule group. Terminal device load balancing characteristics reflect the load balance among terminal devices, while service demand conflict characteristics reflect potential conflicts during resource allocation. Joint optimization calculation comprehensively considers both terminal device load balancing characteristics and service demand conflict characteristics, performing optimization calculations based on the rule combination priority ranking to find a more reasonable resource allocation scheme. The initial version of the resource allocation priority adjustment strategy is a preliminary strategy generated based on the joint optimization calculation, used to adjust the priority of resource allocation.

[0119] Based on rule combination priority ranking, the initial version of the resource allocation priority adjustment strategy can be generated by jointly optimizing the load balancing characteristics of terminal devices and the characteristics of business demand conflicts through rule combination priority ranking. The steps are as follows: First, according to the rule combination priority ranking, the rule groups with higher priority are applied for calculation. Assuming the load balancing rule group has a higher priority, the terminal devices that need load adjustment and the direction of adjustment are determined based on the load balancing rule group and the load balancing characteristics of the terminal devices. For example, some tasks may be transferred from high-load terminal devices to low-load terminal devices. Then, considering the characteristics of business demand conflicts, the initial adjustment plan is revised in conjunction with the conflict avoidance rule group to avoid triggering new conflicts during the adjustment process. For example, if adjusting task allocation may lead to the preemption of resources for a high-priority task, corresponding adjustments are made according to the conflict avoidance rule group. Finally, the initial version of the resource allocation priority adjustment strategy is generated by combining the above calculation results.

[0120] Step S350: Verify the policy effectiveness of the initial version by simulating the execution environment. Based on the improvement in resource utilization and conflict resolution efficiency in the verification results, optimize the resource allocation priority adjustment policy and generate the final version.

[0121] The simulated execution environment is a virtual environment used to simulate the resource allocation process. It can simulate the operating status of terminal devices, the execution process of tasks, and the allocation of resources. The policy performance verification process involves running an initial version of the resource allocation priority adjustment policy in the simulated execution environment to observe and evaluate its effects on resource utilization and conflict resolution. The resource utilization improvement is the degree to which resource utilization is improved after adopting the policy compared to the original level, and the conflict resolution efficiency is the policy's effectiveness in resolving resource allocation conflicts. Based on the verification results, the resource allocation priority adjustment policy is optimized. This involves adjusting the parameters and rules in the policy according to the resource utilization improvement and conflict resolution efficiency to improve the policy's performance, ultimately generating the final version of the resource allocation priority adjustment policy.

[0122] As one implementation method, step S350 involves verifying the policy performance of the initial version by simulating an execution environment, which may specifically include the following steps S351 to S356:

[0123] Step S351: Construct a simulated execution environment containing the current terminal device state image and task queue, and load the initial version of the resource allocation priority adjustment strategy.

[0124] The current terminal device state mirror refers to a copy of the actual operating state of the terminal device, including parameters such as CPU utilization, memory usage, and network bandwidth usage. The task queue is a collection of tasks currently awaiting execution, containing information such as task priority and resource requirements. The simulated execution environment is a software-simulated environment that simulates resource allocation and task execution based on the input terminal device state mirror and task queue. Loading the initial version of the resource allocation priority adjustment strategy involves applying the generated initial version strategy to the simulated execution environment to verify its effectiveness.

[0125] The following steps can be used to build a simulated execution environment containing the current terminal device status image and task queues, and load an initial version of the resource allocation priority adjustment policy. First, collect the current terminal device status data through a monitoring agent and organize this data into an image file, for example, using virtual machine snapshot technology to create the terminal device status image. Simultaneously, obtain the current task queue information, including detailed task information, from the power business system. Then, use simulated execution environment software, such as a cloud computing platform-based simulation system, to load the terminal device status image and task queues into the simulated environment. Finally, configure the initial version of the resource allocation priority adjustment policy into the simulated execution environment so that it can run. For example, using the OpenStack cloud platform to build the simulated execution environment, import the collected terminal device status data and task queue information into an OpenStack virtual machine instance, and configure the initial version of the resource allocation priority adjustment policy.

[0126] Step S352: Inject historical network load fluctuation data and sudden task request data into the simulated execution environment to trigger policy execution and record the resource allocation process log.

[0127] Historical network load fluctuation data records changes in network load over a past period, including curves showing changes in parameters such as network bandwidth utilization and network latency over time. Sudden task request data simulates task requests that may suddenly occur during actual operation; these tasks vary in priority and resource requirements. Injecting historical network load fluctuation data and sudden task request data makes the simulation environment more realistic. Triggering policy execution involves initiating resource allocation priority adjustment strategies within the simulation environment. Recording resource allocation process logs is for subsequent analysis of policy execution; the logs may include information such as task allocation, resource usage, and the occurrence and resolution of conflicts.

[0128] The following steps can be used to inject historical network load fluctuation data and sudden task request data into a simulated execution environment to trigger policy execution and log the resource allocation process. First, obtain historical network load fluctuation data from the network monitoring system and organize it into a suitable format, such as a CSV file. For sudden task request data, simulated sudden tasks can be generated based on the statistical patterns of historical task requests. Then, inject the historical network load fluctuation data and sudden task request data into the simulated execution environment, for example, by inputting the data into the system through the simulated execution environment's interface. Next, initiate the execution of the resource allocation priority adjustment policy, allowing the policy to allocate resources according to the injected data. During policy execution, use logging tools such as Log4j to record various information during the resource allocation process, including task allocation decisions, resource allocation amounts, and task start and end times.

[0129] Step S353: Extract the task processing latency change, calculate the core utilization fluctuation range, and memory leak anomaly events from the process log.

[0130] The process log records the execution process information of the resource allocation priority adjustment strategy in the simulated execution environment. Task processing latency variation is the difference between the task's start-to-completion time and the expected time, reflecting the impact of the strategy on task execution time. Computation core utilization fluctuation range is the difference between the maximum and minimum utilization of computation cores (such as CPU cores) during strategy execution, reflecting the stability of computing resource usage. Memory leak exceptions are situations where memory cannot be properly released during task execution, which can affect system performance and stability.

[0131] Extracting task processing latency changes, calculating core utilization fluctuation ranges, and identifying memory leak anomalies from process logs can be achieved using the following steps: First, parse the process logs, categorizing the information according to dimensions such as task, time, and resource usage. Then, for each task, calculate the difference between its actual processing time and expected processing time to obtain the task processing latency change. For core utilization, identify the maximum and minimum values ​​during strategy execution, calculate their difference, and obtain the core utilization fluctuation range. For memory leak anomalies, identify instances where memory cannot be properly released by analyzing memory usage information and error messages in the logs, and record the relevant task and time information. For example, use regular expressions to parse the process logs, extracting task start time, end time, and core utilization data, calculating task processing latency changes and core utilization fluctuation ranges, and identifying memory leak anomalies by searching for keywords such as "insufficient memory" and "memory not released" in the logs.

[0132] Step S354: Compare the change in task processing latency with the preset latency tolerance threshold to generate a latency compliance assessment result.

[0133] The preset latency tolerance threshold is the maximum allowable variation in task processing latency, determined based on business needs and system performance requirements. The latency compliance assessment result determines whether the variation in task processing latency is within the preset latency tolerance threshold range, categorized as compliant or non-compliant.

[0134] The following steps can be used to compare the change in task processing latency with a preset latency tolerance threshold to generate a latency compliance assessment result. First, determine the preset latency tolerance threshold; for example, based on business requirements, set the maximum allowable change in task processing latency to 10 minutes. Then, compare the change in processing latency for each task extracted from the process log with the preset latency tolerance threshold. If the change in task processing latency is less than or equal to the preset latency tolerance threshold, the task is considered compliant; if the change in task processing latency is greater than the preset latency tolerance threshold, the task is considered non-compliant. Finally, count the number of compliant and non-compliant tasks to generate the latency compliance assessment result.

[0135] Step S355: Determine the resource allocation stability score based on the fluctuation range of the core utilization rate, and generate a resource health index by combining the frequency of memory leak anomalies.

[0136] The resource allocation stability score is a quantitative evaluation of the stability of computing resource allocation. It is determined based on the fluctuation range of computing core utilization; the smaller the fluctuation range, the more stable the resource allocation, and the higher the score. The frequency of memory leak events is the ratio of the number of memory leak events occurring during policy execution to the total number of tasks. The resource health index is an indicator that comprehensively considers both the resource allocation stability score and the frequency of memory leak events, used to assess the overall health of system resources.

[0137] The following steps can be used to determine the resource allocation stability score based on the fluctuation range of computing core utilization and to generate a resource health index by combining this score with the frequency of memory leak events. First, set a scoring standard based on the fluctuation range of computing core utilization. For example, a score of 10 is given when the fluctuation range is less than 10%; a score of 7 is given when the fluctuation range is between 10% and 20%; and a score of 3 is given when the fluctuation range is greater than 20%. The resource allocation stability score is determined based on the fluctuation range of computing core utilization. Next, calculate the frequency of memory leak events. For example, if 5 memory leak events occur while processing 100 tasks, the frequency is 5%. Finally, the resource health index is generated by combining the resource allocation stability score and the frequency of memory leak events. A weighted average method can be used, for example, with a weight of 0.7 for the resource allocation stability score and a weight of 0.3 for the frequency of memory leak events. The resource health index = resource allocation stability score × 0.7 + (1 - frequency of memory leak events) × 0.3 × 10. Assuming the resource allocation stability score is 7 points and the frequency of memory leak events is 5%, then the resource health index = 7 × 0.7 + (1 - 0.05) × 0.3 × 10 = 4.9 + 2.85 = 7.75.

[0138] Step S356: Calculate the overall effectiveness score of the strategy by combining the latency compliance assessment results, resource allocation stability score, and resource health index, and perform parameter optimization on the initial version based on the score results.

[0139] The latency compliance assessment results reflect the impact of the strategy on task processing time, the resource allocation stability score reflects the stability of computing resource allocation, and the resource health index comprehensively considers resource allocation stability and memory leaks. The overall strategy performance score is a holistic score obtained by combining these three indicators, used to evaluate the overall performance of the resource allocation priority adjustment strategy. Parameter tuning involves adjusting the parameters in the initial version of the resource allocation priority adjustment strategy based on the overall strategy performance score to improve the strategy's performance.

[0140] The following steps can be used to calculate the overall strategy effectiveness score by combining the latency compliance assessment results, resource allocation stability score, and resource health index, and then fine-tune the parameters of the initial version based on the score. First, assign weights to the latency compliance assessment results, resource allocation stability score, and resource health index, for example, weights of 0.3, 0.3, and 0.4 respectively. Then, convert the latency compliance assessment results into a score; for example, if the compliance task ratio is 80%, the score is 8 points. Next, calculate the overall strategy effectiveness score according to the weights: Overall Strategy Effectiveness Score = Latency Compliance Score × 0.3 + Resource Allocation Stability Score × 0.3 + Resource Health Index × 0.4. Assuming the latency compliance score is 8 points, the resource allocation stability score is 7 points, and the resource health index is 7.75 points, then the overall strategy effectiveness score = 8 × 0.3 + 7 × 0.3 + 7.75 × 0.4 = 2.4 + 2.1 + 3.1 = 7.6. Finally, fine-tune the parameters of the initial version based on the overall strategy effectiveness score. If the score is low, it indicates that the strategy's performance needs improvement. Parameters within the strategy can be adjusted, such as task allocation priority rules and resource allocation ratios. For example, analysis might reveal that the task allocation priority rules are causing excessive latency for some high-priority tasks. Adjusting the priority rules would then increase the allocation priority of high-priority tasks. After multiple adjustments and verifications, until the overall strategy performance score reaches a satisfactory level, the final version of the resource allocation priority adjustment strategy is generated.

[0141] Step S360: Perform policy coordination verification between the final version and the task scheduling optimization policy and the network bandwidth reallocation policy to generate a set of computing resource configuration policies that pass the consistency verification.

[0142] Task scheduling optimization strategies are used to optimize task execution order and allocation, while network bandwidth reallocation strategies are used to adjust network bandwidth allocation. Policy coherence verification checks whether the final version of the resource allocation priority adjustment strategy is coordinated with and does not conflict with the task scheduling optimization strategy and the network bandwidth reallocation strategy. The set of computing resource configuration strategies that passes consistency verification has been verified to ensure that the various strategies can work together to form a unified and effective set of computing resource configuration strategies.

[0143] The final version of the computational resource configuration strategy is validated against the task scheduling optimization strategy and the network bandwidth reallocation strategy to generate a set of strategies that pass consistency verification. The following steps are employed: First, a detailed analysis is performed on the final version of the resource allocation priority adjustment strategy, the task scheduling optimization strategy, and the network bandwidth reallocation strategy to determine the objectives, rules, and execution flow of each strategy. Then, these three strategies are run simultaneously in a simulated execution environment to observe their interaction during resource allocation and task execution. Conflicts are checked, such as the resource allocation priority adjustment strategy assigning a task to one terminal device, but the task scheduling optimization strategy assigning the same task to another terminal device, or the bandwidth allocated by the network bandwidth reallocation strategy not matching the resource requirements of the tasks in the resource allocation priority adjustment strategy. If conflicts are found, the strategies are adjusted, for example, by modifying the rules or parameters within the strategies to ensure coordination. After multiple adjustments and verifications, the three strategies are validated until they work collaboratively in the simulated execution environment without conflict, generating a set of computational resource configuration strategies that pass consistency verification. For example, in a simulated execution environment, a conflict was found between the resource allocation priority adjustment strategy and the task scheduling optimization strategy in task allocation. By adjusting the task sorting rules in the task scheduling optimization strategy to match the resource allocation priority adjustment strategy, a set of computing resource configuration strategies that passed the consistency check was finally generated.

[0144] Step S400: Based on the set of computing resource configuration strategies, perform adjustment operations on the computing resources of the power wireless terminal, and generate resource allocation verification results and terminal operation performance feedback data.

[0145] The set of computing resource configuration strategies includes strategies such as resource allocation priority adjustment, task scheduling optimization, and network bandwidth reallocation. These strategies specify how to configure the computing resources of the power wireless terminal. Performing an adjustment operation on the computing resources of the power wireless terminal involves reallocating and adjusting the terminal device's computing resources (such as CPU, memory, and network bandwidth) according to the rules and parameters in the strategy set. The resource allocation verification result verifies the resource allocation after the adjustment operation, including whether the resources were allocated according to the strategy and whether the allocation was reasonable. Terminal performance feedback data reflects information about the terminal device's performance after the adjustment operation, such as task processing speed and resource utilization.

[0146] As one implementation method, step S400 involves adjusting the computing resources of the power wireless terminal according to the set of computing resource allocation strategies, generating resource allocation verification results and terminal operation performance feedback data. Specifically, this may include the following steps S410 to S450:

[0147] Step S410: Analyze the resource allocation priority adjustment strategy in the set of computing resource configuration strategies, and determine the list of terminal devices to be adjusted and the corresponding computing core reallocation parameters.

[0148] The resource allocation priority adjustment strategy in the set of computing resource allocation strategies specifies how to adjust the priority of resource allocation. Analyzing this strategy involves a detailed analysis of its rules and parameters to understand its intent and specific operational methods. The list of terminal devices to be adjusted is the set of terminal devices that require computing resource adjustments according to the strategy. The computing core reallocation parameters are parameters such as the number or proportion of computing cores allocated to each terminal device to be adjusted.

[0149] The following steps can be used to analyze the resource allocation priority adjustment policies in the set of computing resource allocation policies to determine the list of terminal devices to be adjusted and the corresponding computing core reallocation parameters. First, the resource allocation priority adjustment policies are parsed to extract the rules and conditions. For example, the policy might stipulate that "when the CPU utilization of terminal device A exceeds 80%, 2 computing cores will be allocated to terminal device B." Then, based on the current operating status of the terminal devices, such as CPU utilization and memory usage, it is determined which terminal devices need adjustment and these are added to the list of terminal devices to be adjusted. Finally, according to the rules in the policy, the computing core reallocation parameters are determined for each terminal device to be adjusted. For example, if monitoring reveals that the CPU utilization of terminal device A is 85%, according to the policy rules, 2 computing cores will be allocated from terminal device A to terminal device B. The computing core reallocation parameter for terminal device A is determined to be a reduction of 2 computing cores, and the computing core reallocation parameter for terminal device B is an increase of 2 computing cores.

[0150] Step S420: Based on the task dependency chain in the task scheduling optimization strategy, restructure the current task queue to generate an optimized task execution order table.

[0151] The task dependency chain in the task scheduling optimization strategy describes the dependencies between tasks, meaning that the execution of one task may depend on the completion of other tasks. The current task queue is the set of tasks currently waiting to be executed, and these tasks may be arranged in the order of their submission time. Sequence refactoring involves readjusting the order of tasks in the current task queue based on the task dependency chain to improve the efficiency and rationality of task execution. The optimized task execution order table is a list of task execution orders obtained after sequence refactoring.

[0152] As one implementation method, step S420 involves reconstructing the current task queue according to the task dependency chain in the task scheduling optimization strategy to generate an optimized task execution order table. This may specifically include the following steps S421 to S425:

[0153] Step S421: Identify the critical path task nodes and non-critical path task nodes in the task dependency chain, and add a priority execution flag to the critical path task nodes.

[0154] Task dependency chains can be represented as a graph, where nodes represent tasks and edges represent dependencies between tasks. Critical path task nodes are those that play a crucial role in the overall task completion time. Non-critical path task nodes are those that do not affect the overall task completion time. Priority execution flags are used to mark critical path task nodes so that these tasks are executed first during task scheduling.

[0155] Identifying critical path and non-critical path task nodes in a task dependency chain and adding priority execution markers to critical path task nodes can be achieved through the following steps. First, use graph theory algorithms, such as the Critical Path Method (CPM), to analyze the task dependency chain. By calculating the earliest start time, earliest finish time, latest start time, and latest finish time of each task node, the critical path is identified. Task nodes on the critical path are critical path task nodes, and the remaining task nodes are non-critical path task nodes. Then, add priority execution markers to the critical path task nodes. For example, add a "priority" field to the task information and set the "priority" field of the critical path task nodes to "high". For instance, for a task dependency chain containing tasks A, B, C, and D, CPM analysis determines that task A->B->D is the critical path, and priority execution markers are added to tasks A, B, and D.

[0156] Step S422: Based on the urgent task marker in the business priority identifier, sort the tasks with the same priority in a secondary order according to the processing delay constraint.

[0157] The urgent task marker in the business priority identifier is used to distinguish the urgency of tasks. The processing time constraint is the requirement that tasks must be completed within a specified time. For tasks with the same priority, their execution order cannot be determined solely by priority; a secondary sorting based on the processing time constraint is necessary to ensure that urgent tasks with strict processing time requirements are executed first.

[0158] Based on the urgent task marker in the business priority identifier, the following steps can be used to perform a secondary sorting of tasks with the same priority according to their processing delay constraints. First, extract the urgent task marker and processing delay constraint information from the business priority identifier. Then, filter out tasks with the same priority and sort them according to their processing delay constraints. Ascending order can be used, meaning tasks with shorter processing delay constraints are ranked first. For example, given tasks E, F, and G with the same priority, task E has a processing delay constraint of 30 minutes, task F has a processing delay constraint of 60 minutes, and task G has a processing delay constraint of 20 minutes, they can be sorted in ascending order according to their processing delay constraints as task G, task E, and task F.

[0159] Step S423: Detect circular dependency chains in the task queue and remove circular dependencies by inserting virtual synchronization nodes.

[0160] A circular dependency chain occurs when tasks depend on each other, forming a cycle. This circular dependency can prevent tasks from executing correctly because each task is waiting for the others to complete. A virtual synchronization node is a virtual task node used to break circular dependencies. It does not perform any actual tasks; it simply acts as a synchronization point.

[0161] Detecting circular dependencies in a task queue and breaking them by inserting virtual synchronization nodes can be achieved using the following steps. First, use a topological sorting algorithm from graph theory, such as Kahn's algorithm, to perform a topological sort on the task dependency chain. If the sorting process fails, a circular dependency chain exists. Next, identify the task nodes in the circular dependency chain. For example, if task H depends on task I, and task I depends on task H, this forms a circular dependency chain. Then, insert a virtual synchronization node into the circular dependency chain, for example, inserting virtual synchronization node J between task H and task I. Modify the dependency of task H to depend on virtual synchronization node J, and virtual synchronization node J depends on task I. This breaks the circular dependency, allowing the tasks to execute normally.

[0162] Step S424: Based on the resource requirements of task processing, tasks that occupy the same computing resources are divided into time slices to generate a task time slice distribution map without resource conflicts.

[0163] Task processing resource requirements describe the computing resources (such as CPU, memory, etc.) needed by each task during execution. Time-slicing partitioning separates tasks that consume the same computing resources over time, preventing them from simultaneously occupying the same resources and thus avoiding resource conflicts. A task time-slice distribution map without resource conflicts is a graph showing the time allocation of tasks, ensuring that each task does not conflict with other tasks during execution.

[0164] Based on the resource requirements of tasks, tasks that consume the same computing resources can be time-sliced ​​to generate a task time-slice distribution map without resource conflicts. The following steps can be used: First, analyze the resource requirements of each task to determine the type and quantity of computing resources needed. Then, group tasks that consume the same computing resources. For each group, time-slice tasks based on their processing latency constraints and resource requirements. A greedy algorithm can be used, prioritizing tasks with shorter processing latency constraints. For example, if tasks K, L, and M all require CPU resources, and task K has a processing latency constraint of 20 minutes, task L has a processing latency constraint of 30 minutes, and task M has a processing latency constraint of 40 minutes, then task K should be scheduled to execute from 0-20 minutes, then task L from 20-50 minutes, and finally task M from 50-90 minutes. Finally, plot the task time allocation on a time-slice distribution map to generate a task time-slice distribution map without resource conflicts.

[0165] Step S425: Based on the priority execution identifier, the secondary sorting result, and the time slice distribution map, reconstruct the task queue and generate an optimized task execution order table.

[0166] Priority execution markers indicate critical path task nodes that must be executed first. Secondary sorting determines the execution order of tasks with the same priority. The time slice distribution diagram shows the allocation of tasks over time, avoiding resource conflicts. Restructuring the task queue involves rearranging the current task queue based on this information to generate an optimized task execution order table.

[0167] The following steps can be used to reconstruct the task queue and generate an optimized task execution order table based on priority indicators, secondary sorting results, and time slice distribution diagrams. First, based on priority indicators, prioritize critical path task nodes at the front of the task queue. Then, for tasks with the same priority, arrange them according to the secondary sorting results. Finally, based on the time slice distribution diagram, insert the tasks into the task queue in chronological order. For example, if the critical path task nodes are tasks A, B, and D, and the secondary sorting result for tasks E, F, and G with the same priority is tasks G, E, and F, and the time slice distribution diagram shows that task G executes from 0-20 minutes, task A from 20-30 minutes, task E from 30-50 minutes, task B from 50-60 minutes, task F from 60-90 minutes, and task D from 90-100 minutes, the reconstructed task queue yields the optimized task execution order table as task G, task A, task E, task B, task F, and task D.

[0168] Step S430: Based on the bandwidth adjustment ratio in the network bandwidth reallocation strategy, modify the network interface configuration of the target terminal device in real time and monitor changes in network throughput.

[0169] The bandwidth adjustment ratio in the network bandwidth reallocation strategy specifies how to adjust the network bandwidth allocation for the target terminal device. The target terminal device is the terminal device that requires network bandwidth adjustment. Real-time modification of the target terminal device's network interface configuration involves adjusting parameters such as bandwidth limits for the terminal device's network interface according to the bandwidth adjustment ratio. Monitoring changes in network throughput involves monitoring network throughput (the amount of data passing through the network per unit time) in real time after modifying the network interface configuration to evaluate the effectiveness of the bandwidth adjustment.

[0170] The following steps can be used to modify the network interface configuration of a target terminal device in real time and monitor changes in network throughput based on the bandwidth adjustment ratio in the network bandwidth reallocation policy. First, determine the network interface information of the target terminal device, such as the network interface name and IP address. Then, calculate the new network bandwidth limit for the target terminal device according to the bandwidth adjustment ratio in the network bandwidth reallocation policy. Next, use network management tools, such as the `tc` command in Linux or the `Netsh` command in Windows, to modify the network interface configuration of the target terminal device in real time and set the new network bandwidth limit. Finally, use network monitoring tools, such as Wireshark or Ntopng, to monitor changes in network throughput in real time and record the network throughput data before and after the modification to evaluate the effect of the bandwidth adjustment.

[0171] Step S440: After performing the adjustment operation, collect the real-time resource utilization rate, task processing completion rate and network transmission error rate of the terminal device, and generate resource allocation verification results.

[0172] The adjustment operation refers to the adjustment of computing resources of the power wireless terminal according to a set of computing resource allocation strategies, including computing core reallocation, task scheduling optimization, and network bandwidth reallocation. Real-time resource utilization rate is the real-time CPU utilization, memory utilization, and other resource usage of the terminal device after the adjustment operation. Task completion rate is the ratio of the number of tasks completed within a set time to the total number of tasks. Network transmission error rate is the ratio of the number of data packets that malfunction during network transmission to the total number of data packets. The resource allocation verification result is generated based on the collected real-time resource utilization rate, task completion rate, and network transmission error rate, and is used to verify whether the resource allocation adjustment operation has achieved the expected results.

[0173] After performing the adjustment operation, the real-time resource utilization rate, task completion rate, and network transmission error rate of the terminal device are collected to generate resource allocation verification results. The following steps can be used: First, a monitoring agent program is used to collect real-time resource utilization data of the terminal device, such as CPU utilization and memory utilization. For the task completion rate, the start and end times of tasks can be recorded through a task management system, and the number of completed tasks and the total number of tasks can be counted to calculate the task completion rate. For the network transmission error rate, network monitoring equipment, such as a network analyzer, can be used to count the number of data packets that erroneously occurred during network transmission and the total number of data packets to calculate the network transmission error rate. Then, the collected data is organized and analyzed to determine whether the resource allocation adjustment operation was reasonable. For example, if the real-time resource utilization rate is more balanced, the task completion rate increases, and the network transmission error rate decreases, it indicates that the resource allocation adjustment operation has achieved the expected results; otherwise, further analysis of the reasons and adjustments are needed. Finally, based on the analysis results, a resource allocation verification result is generated, which may include information such as whether the resource allocation is reasonable and which aspects need improvement.

[0174] Step S450: Compare and analyze the resource allocation verification results with the historical performance data before the adjustment, extract the resource utilization improvement rate and task processing acceleration rate, and encapsulate them as terminal operation performance feedback data.

[0175] The resource allocation verification results reflect the resource allocation and operational efficiency of the terminal devices after the adjustment operation. Historical performance data before the adjustment includes data such as resource utilization and task processing time of the terminal devices before the adjustment operation. The resource utilization improvement rate is the percentage increase in resource utilization after the adjustment operation compared to before the adjustment. The task processing speedup rate is the percentage reduction in task processing time after the adjustment operation compared to before the adjustment. Terminal operational efficiency feedback data is data that encapsulates information such as the resource utilization improvement rate and task processing speedup rate, used to provide feedback on changes in the operational efficiency of the terminal devices.

[0176] The following steps can be used to compare and analyze the resource allocation verification results with the historical performance data before adjustment, extract the resource utilization improvement rate and task processing speedup rate, and encapsulate them as terminal operation performance feedback data. First, extract the adjusted resource utilization rate and task processing time data from the resource allocation verification results, and extract the resource utilization rate and task processing time data before adjustment from the historical performance data. Then, calculate the resource utilization improvement rate: Resource utilization improvement rate = (Adjusted resource utilization rate - Unadjusted resource utilization rate) / Unadjusted resource utilization rate × 100%. Calculate the task processing speedup rate: Task processing speedup rate = Unadjusted task processing time / Adjusted task processing time. For example, if the resource utilization rate before adjustment was 60% and the resource utilization rate after adjustment is 70%, then the resource utilization improvement rate = (70% - 60%) / 60% × 100% ≈ 16.7%; if the task processing time before adjustment was 100 minutes and the task processing time after adjustment is 80 minutes, then the task processing speedup rate = 100 / 80 = 1.25. Finally, the resource utilization improvement rate and task processing speedup rate are encapsulated as terminal performance feedback data, for example, in JSON format: {"resource_utilization_improvement_rate":16.7,"task_processing_speedup_ratio":1.25}.

[0177] Step S500: Continuously compare the terminal performance feedback data with the preset performance optimization threshold, and iteratively update the resource adaptation rule base based on the comparison results.

[0178] Terminal performance feedback data includes information such as resource utilization improvement rate and task processing acceleration rate, reflecting the terminal device's performance after resource allocation adjustments. The preset performance optimization thresholds are pre-defined target values ​​for resource utilization improvement and task processing acceleration, used to measure whether the resource allocation adjustments have achieved the expected results. Continuous comparison involves constantly comparing the terminal performance feedback data with the preset performance optimization thresholds. Iterative update processing involves adjusting and updating the rules in the resource adaptation rule base based on the comparison results to improve the effectiveness of resource allocation.

[0179] As one implementation method, step S500 involves continuously comparing the terminal performance feedback data with a preset performance optimization threshold, and iteratively updating the resource adaptation rule base based on the comparison results. Specifically, this may include the following steps S510 to S550:

[0180] Step S510: Extract the resource utilization improvement ratio from the terminal operation performance feedback data and compare it with the lowest improvement threshold in the performance optimization threshold.

[0181] The terminal performance feedback data includes information such as the resource utilization improvement rate and task processing speedup rate. The minimum improvement threshold in the performance optimization threshold is a pre-set minimum target value for resource utilization improvement. Extracting the resource utilization improvement rate from the terminal performance feedback data and comparing it with the minimum improvement threshold is to determine whether the resource allocation adjustment has met the basic performance improvement requirements.

[0182] Extracting the resource utilization improvement rate from terminal performance feedback data and comparing it with the minimum improvement threshold in the performance optimization threshold can be achieved using the following steps: First, parse the terminal performance feedback data and extract the resource utilization improvement rate. For example, if the terminal performance feedback data is {"resource_utilization_improvement_rate":16.7,"task_processing_speedup_ratio":1.25}, the extracted resource utilization improvement rate is 16.7%. Next, obtain the minimum improvement threshold in the performance optimization threshold, for example, a minimum improvement threshold of 15%. Finally, compare the extracted resource utilization improvement rate with the minimum improvement threshold. If the resource utilization improvement rate is greater than or equal to the minimum improvement threshold, it indicates that the resource allocation adjustment has met the basic performance improvement requirements; if the resource utilization improvement rate is less than the minimum improvement threshold, further analysis of the reasons and adjustments are needed.

[0183] Step S520: If the resource utilization improvement rate is lower than the minimum improvement threshold, activate the rule base diagnosis mode to perform anomaly detection on the load balancing rule group in the resource adaptation rule base.

[0184] The rule base diagnostic mode is used to check for anomalies in the rules of the resource adaptation rule base. A load balancing rule group is a set of rules in the resource adaptation rule base used to balance the load on terminal devices. When the resource utilization improvement rate is lower than the minimum improvement threshold, it indicates that the resource allocation adjustment is not ideal, which may be due to a problem with the load balancing rule group. Therefore, it is necessary to activate the rule base diagnostic mode to detect anomalies in the load balancing rule group.

[0185] As one implementation method, in step S520, anomaly detection is performed on the load balancing rule group in the resource adaptation rule base, which may specifically include the following steps S521 to S525:

[0186] Step S521: Count the number of times the load balancing rule group is triggered and the number of times it is successfully executed within the preset time period, and calculate the rule execution success rate.

[0187] The preset time period is a pre-defined time range used to statistically analyze the execution status of load balancing rule groups, such as the past week. The number of times a load balancing rule group is triggered is the number of times the rule group is executed within the preset time period. The number of successful executions is the number of times a rule is successfully executed out of the number of triggers. The rule execution success rate is the ratio of successful executions to triggers, used to measure the performance of the load balancing rule group.

[0188] To calculate the rule execution success rate by statistically analyzing the number of triggers and successful executions of a load balancing rule group within a preset time period, the following steps can be used: First, filter the execution records of the load balancing rule group within the preset time period from the rule execution log. The rule execution log records information such as the rule trigger time and execution result. Then, count the number of triggers and successful executions. For example, in the past week, the load balancing rule group was triggered 100 times, of which 80 were successfully executed. Finally, calculate the rule execution success rate: Rule execution success rate = (Number of successful executions / Number of triggers) × 100%, i.e., Rule execution success rate = 80 / 100 × 100% = 80%.

[0189] Step S522: If the rule execution success rate is lower than the preset success rate threshold, it is determined that there is an anomaly in the load balancing rule group.

[0190] The preset success rate threshold is the minimum required success rate for rule execution, used to determine whether the load balancing rule group is working properly. When the rule execution success rate is lower than the preset success rate threshold, it indicates that the load balancing rule group is experiencing a high number of failures during execution, which may indicate an anomaly requiring further analysis.

[0191] If the rule execution success rate is lower than a preset success rate threshold, the following steps can be used to determine if the load balancing rule group is abnormal. First, determine the preset success rate threshold, for example, 85%. Then, compare the calculated rule execution success rate with the preset success rate threshold. If the rule execution success rate is lower than the preset success rate threshold, such as 80% in the example above, then the load balancing rule group is determined to be abnormal.

[0192] Step S523: Extract terminal device load data and resource allocation logs within the abnormal time period, and identify the rule conditions that were not successfully executed and the corresponding device status characteristics.

[0193] The abnormal time period refers to the time when the rule execution success rate is lower than the preset success rate threshold. Terminal device load data includes CPU utilization, memory usage, and other load information of the terminal devices during the abnormal time period. The resource allocation log records detailed information on resource allocation during the abnormal time period, including task allocation and resource usage. The conditions for unsuccessfully executed rules are those conditions within the load balancing rule group that cause rule execution to fail. The corresponding device status characteristics are the status characteristics of the terminal devices when the rules failed to execute, such as high CPU utilization or insufficient memory.

[0194] To extract terminal device load data and resource allocation logs within an abnormal time period and identify unexecuted rule conditions and corresponding device status characteristics, the following steps can be used: First, obtain terminal device load data within the abnormal time period from the monitoring agent and format it appropriately. Simultaneously, extract resource allocation logs within the abnormal time period from the rule execution logs. Then, analyze the resource allocation logs to identify unexecuted rule records. For each unexecuted rule record, extract the rule conditions and corresponding terminal device status information.

[0195] Step S524: Divide the device status characteristics into multiple abnormal scenario categories through cluster analysis, and generate supplementary rules for each abnormal scenario category.

[0196] Cluster analysis is used to group similar data points into a single category. Device status features are the status information of the terminal device when a rule fails to execute, such as CPU utilization and memory usage. Abnormal scenario categories are different categories of abnormal situations divided based on device status features. Rule supplementary conditions are conditions added to enable the load balancing rule group to better handle abnormal situations. The steps to divide device status features into multiple abnormal scenario categories using cluster analysis and generate rule supplementary conditions for each abnormal scenario category are as follows: First, organize the extracted device status features into a dataset, where each data point contains values ​​for multiple device status features, such as (CPU utilization, memory usage). Then, use a cluster analysis algorithm, such as the K-Means algorithm, to perform cluster analysis on the dataset. Based on the clustering results, divide the device status features into multiple abnormal scenario categories. For example, divide the device status features into three abnormal scenario categories: high CPU utilization and high memory usage, high CPU utilization and low memory usage, and low CPU utilization and high memory usage. For each abnormal scenario category, analyze its characteristics and the possible reasons for rule execution failure, and generate rule supplementary conditions. For example, for the abnormal scenario category of high CPU utilization and high memory usage, the generated rule is supplemented with the condition "when the terminal device's CPU utilization exceeds 80% and memory usage exceeds 90%, release some memory before allocating tasks".

[0197] Step S525: Add the supplementary rules to the original load balancing rule group to form an expanded load balancing rule group.

[0198] Additional rule conditions are generated to handle exceptional situations. Adding them to the existing load balancing rule group makes the rule group more comprehensive and better able to handle various exceptional circumstances. An expanded load balancing rule group is a rule group based on the original load balancing rule group with the added additional rule conditions.

[0199] Adding supplementary rules to an existing load balancing rule group to create an expanded load balancing rule group can be done using the following steps: First, analyze the existing load balancing rule group to determine the structure and format of the rules. Then, add the generated supplementary rules to the existing load balancing rule group according to the rule's structure and format. For example, if the existing load balancing rule group contains a rule that states "When the CPU utilization of a terminal device exceeds 80%, allocate some tasks to other terminal devices," the supplementary rule generated for the abnormal scenario category of high CPU utilization and high memory usage is "When the CPU utilization of a terminal device exceeds 80% and the memory usage exceeds 90%, release some memory before allocating tasks." Adding this supplementary rule to the existing rule creates the expanded rule: "When the CPU utilization of a terminal device exceeds 80%, if the memory usage exceeds 90%, release some memory before allocating some tasks to other terminal devices; otherwise, directly allocate some tasks to other terminal devices." Finally, update the load balancing rule group in the resource adaptation rule base and save the expanded load balancing rule group to the rule base.

[0200] Step S530: Extract the historical records of rule matching failure events and analyze the matching deviation between the rule triggering conditions and the current terminal device status.

[0201] The history of rule matching failure events records events where, during resource allocation, a rule failed to successfully match the terminal device state. The rule triggering condition is the prerequisite for rule execution, and the current terminal device state refers to the terminal device's current operating state, such as CPU utilization and memory usage. The matching deviation is the degree of difference between the rule triggering condition and the current terminal device state, used to evaluate the rule's applicability.

[0202] To extract historical records of rule matching failure events and analyze the matching deviation between the rule triggering conditions and the current terminal device state, the following steps can be taken. First, extract historical records of rule matching failure events from the rule execution log. These records contain the rule triggering conditions and the corresponding terminal device state information. Then, for each rule matching failure event record, compare the rule triggering conditions with the current terminal device state. For example, if the rule triggering condition is "terminal device CPU utilization exceeds 80%", and the current terminal device state is CPU utilization of 75%, then there is a 5% matching deviation. Numerical calculation methods can be used, such as calculating the difference or proportion between the rule triggering condition and the current terminal device state, to quantify the matching deviation. Finally, perform statistical analysis on the matching deviation of all rule matching failure events to identify rules with large matching deviations and their corresponding device state characteristics, so that rules can be adjusted accordingly.

[0203] Step S540: Adaptively calibrate the condition thresholds in the load balancing rule group based on the matching deviation to generate an updated load balancing rule group.

[0204] Match deviation reflects the degree of difference between the rule triggering condition and the current state of the terminal device. The condition threshold is the boundary condition for triggering rules in the load balancing rule group; for example, 80% in "terminal device CPU utilization exceeds 80%" is the condition threshold. Adaptive calibration automatically adjusts the condition threshold based on the match deviation, enabling the rules to better adapt to the actual state of the terminal device. The updated load balancing rule group is the rule group obtained after adaptive calibration.

[0205] The updated load balancing rule group can be generated by adaptively calibrating the condition thresholds in the load balancing rule group based on the matching deviation. First, analyze the statistical results of the matching deviation to determine the condition thresholds that need adjustment. For example, if the rule "Terminal device CPU utilization exceeds 80%" has a large matching deviation, and in most cases, terminal devices need task allocation when CPU utilization reaches 75%, then the condition threshold can be adjusted to 75%. Next, modify the corresponding condition thresholds in the load balancing rule group. For each rule that needs adjustment, replace the condition threshold with the new value. Finally, generate the updated load balancing rule group to ensure that the rules can more accurately match the state of the terminal devices and improve resource allocation efficiency.

[0206] Step S550: Re-inject the updated load balancing rule group into the resource adaptation rule base and disable the rule base diagnostic mode.

[0207] The updated load balancing rule group is the result of anomaly detection, rule supplementation, and condition threshold calibration. Re-injecting it into the resource adaptation rule base makes the load balancing rule group in the resource adaptation rule base more complete and better guides resource allocation. Disabling the rule base diagnostic mode stops the rule base diagnostic checks after the rule update is complete, resuming normal resource allocation operations.

[0208] The steps to re-inject the updated load balancing rule group into the resource adaptation rule base and disable the rule base diagnostic mode are as follows: First, save the updated load balancing rule group to the corresponding location in the rule base, replacing the original load balancing rule group. Database operations or file read / write operations can be used to update the rule group. Then, update the rule base version information, recording the update time and content of the rule group. Finally, disable the rule base diagnostic mode, restoring the resource adaptation rule base to normal operation, enabling it to continue matching and generating resource allocation strategies based on the new rule group. By continuously comparing terminal performance feedback data with preset performance optimization thresholds and iteratively updating the resource adaptation rule base, resource allocation strategies can be continuously optimized, improving the computing resource utilization efficiency and operational performance of power wireless terminals.

[0209] It is understood that the various algorithms involved in the above descriptions of the embodiments of the present invention, such as Euclidean distance, cosine distance, conflict resolution algorithms, etc., can all be obtained from relevant content in the prior art. To save space, they will not be elaborated on in this application embodiment. In addition, those skilled in the art can supplement the details based on common knowledge in the art when implementing the solution of this application. For example, they can use normalization to eliminate dimensional conflicts before feature fusion, use interpolation to eliminate dimensional differences, reasonably set the threshold based on historical data scores or business needs, train the model based on a general model training method, etc. This application will not provide redundant descriptions of the implementation process in excessive detail here.

[0210] Please refer to the following: Figure 2 , Figure 2This is a schematic diagram of a computer system provided in an embodiment of the present invention. The computer system includes at least a processor 101, a communication interface 102, and a memory 103. The processor 101, communication interface 102, and memory 103 can be connected via a bus or other means. The processor 101 (or Central Processing Unit, CPU) is the computing and control core of the computer system, capable of parsing various instructions and processing various data within the computer system. The communication interface 102 may optionally include a standard wired interface or a wireless interface (such as Wi-Fi, mobile communication interface, etc.), and can be used to send and receive data under the control of the processor 101; the communication interface 102 can also be used for data transmission and interaction within the computer system. The memory 103 is a storage device in the computer system used to store programs and data. It is understood that the memory 103 here can include the computer system's built-in memory, or it can include extended memory supported by the computer system. The memory 103 provides storage space, which stores the computer system's operating system; this invention does not limit this storage space.

[0211] In one embodiment, the processor 101 executes the dynamic configuration computing resource management method based on a power wireless terminal provided above in the embodiments of the present invention by running a computer program in the memory 103.

Claims

1. A method for dynamically configuring computing resource management based on a power wireless terminal, characterized in that, include: Acquire a real-time operating status data set and a power service demand data set of the power wireless terminal. The real-time operating status data set includes the terminal device's operating parameters and network load fluctuation characteristics, and the power service demand data set includes service priority identifiers and real-time task processing requirements. The real-time operational status data set and the power business demand data set are subjected to multi-dimensional feature fusion processing to generate a resource adaptation feature set. This resource adaptation feature set includes terminal equipment load balancing features, business demand conflict features, and resource allocation efficiency evaluation indicators. Specifically, this includes: extracting time-series features of terminal equipment operating parameters from the real-time operational status data set, where the time-series features include parameter fluctuation period, peak duration, and abnormal parameter offset; and performing task decomposition processing on the real-time task processing requirements in the power business demand data set to obtain multiple sub-task resource demand features, where the sub-task resource demand features include processing delay constraints, calculation... The system calculates core utilization and memory allocation thresholds; it performs correlation matching processing on the time series features and the sub-task resource requirement features to generate a task-device adaptability index, which quantifies the degree of matching between the terminal device's processing capabilities and the sub-task requirements; based on a preset resource conflict detection model, it performs conflict analysis processing on the network load fluctuation features and the service priority identifier to generate resource contention hotspot area identifiers and conflict mitigation suggestion parameters; and it integrates the task-device adaptability index, resource contention hotspot area identifiers, and conflict mitigation suggestion parameters to construct the resource adaptability feature set and associate it with the terminal device identifier and the service task identifier. The process involves performing correlation matching between the time series features and the sub-task resource requirement features to generate a task-device adaptability index. This includes: extracting the duration distribution features of parameter fluctuation periods and the equipment load state change curve corresponding to the peak duration from the time series features; performing task execution window segmentation on the processing latency constraints in the sub-task resource requirement features to generate the minimum time window requirement and maximum time window tolerance threshold for each sub-task; and aligning the equipment load state change curve with the minimum time window requirement on the time axis to identify the available time in the equipment load state curve that meets the minimum time window requirement of the sub-task. A set of available time periods is used; based on the continuous distribution characteristics of the available time period set and the interval parameter of the peak duration, the dynamic matching coefficient of each subtask within the corresponding available time period is calculated. The dynamic matching coefficient is used to quantify the fit between the device load fluctuation and the time window requirements of the subtask; the dynamic matching coefficient is normalized and corrected according to the maximum time window tolerance threshold to generate a corrected set of dynamic matching coefficients; the corrected set of dynamic matching coefficients is combined with the computing core occupancy rate and memory allocation threshold in the resource requirement characteristics of the subtasks to perform multi-dimensional weighted fusion processing to generate a task-device adaptability index associated with each subtask and terminal device; Based on a preset resource adaptation rule base, the resource adaptation feature set is subjected to real-time policy matching processing to generate a computing resource configuration policy set, which includes resource allocation priority adjustment policy, task scheduling optimization policy and network bandwidth reallocation policy. Based on the set of computing resource configuration strategies, the computing resources of the power wireless terminal are adjusted to generate resource allocation verification results and terminal operation performance feedback data. The terminal's performance feedback data is continuously compared with a preset performance optimization threshold, and the resource adaptation rule base is iteratively updated based on the comparison results.

2. The method according to claim 1, characterized in that, The preset resource conflict detection model performs conflict analysis on the network load fluctuation characteristics and the service priority identifier to generate resource contention hotspot area identifiers and conflict mitigation suggestion parameters, including: Based on the bandwidth utilization change curve in the network load fluctuation characteristics, determine the network resource saturation time period and idle time period, and associate them with the corresponding terminal device identifier; Extract the urgent task marker and task dependency chain from the business priority identifier to generate a task execution order constraint set; By performing a spatiotemporal overlap analysis between the network resource saturation time period and the task execution order constraint set, the resource competition conflict period and the set of affected tasks can be identified. Based on a preset conflict resolution algorithm, the network bandwidth allocation strategy during the resource contention conflict period is simulated to generate multiple candidate bandwidth allocation schemes. Based on the impact of the candidate bandwidth allocation schemes on the processing latency of the affected task set, the target bandwidth allocation scheme with the smallest latency increment is selected, and its bandwidth adjustment parameters are extracted as conflict mitigation suggestion parameters. The resource competition conflict period and the corresponding terminal device identifier are encapsulated as a resource competition hotspot area identifier.

3. The method according to claim 1, characterized in that, The method, based on a preset resource adaptation rule base, performs real-time policy matching processing on the resource adaptation feature set to generate a set of computing resource configuration policies, including: Extract the historical change trend of the terminal device load balancing characteristics from the resource adaptation feature set and match them to the load balancing rule group in the resource adaptation rule base; Based on the number of task processing conflicts and the frequency of resource preemption in the business requirement conflict characteristics, activate the conflict avoidance rule group in the resource adaptation rule base; The resource utilization curve and task completion rate index in the resource allocation efficiency evaluation index are called to perform weight allocation processing on the load balancing rule group and conflict avoidance rule group, and generate a rule combination priority ranking. Based on the priority ranking of the rule combination, the load balancing characteristics of the terminal device and the conflict characteristics of business requirements are jointly optimized and calculated to generate an initial version of the resource allocation priority adjustment strategy. The initial version is subjected to strategy performance verification by simulating the execution environment. Based on the improvement in resource utilization and conflict resolution efficiency in the verification results, the resource allocation priority adjustment strategy is optimized and the final version is generated. The final version is then subjected to a policy coordination verification with the task scheduling optimization strategy and the network bandwidth reallocation strategy to generate a set of computing resource configuration strategies that pass the consistency verification.

4. The method according to claim 3, characterized in that, The process of verifying the policy effectiveness of the initial version through a simulated execution environment includes: Construct a simulated execution environment that includes the current terminal device state image and task queue, and load the initial version of the resource allocation priority adjustment strategy; Historical network load fluctuation data and sudden task request data are injected into the simulated execution environment to trigger policy execution and record resource allocation process logs. Extract task processing latency changes, calculate core utilization fluctuation range, and memory leak anomalies from the process log; The change in task processing latency is compared with a preset latency tolerance threshold to generate a latency compliance assessment result. The resource allocation stability score is determined based on the fluctuation range of the core utilization rate, and a resource health index is generated by combining the occurrence frequency of the memory leak anomaly. Based on the latency compliance assessment results, resource allocation stability score, and resource health index, a comprehensive strategy performance score is calculated, and the parameters of the initial version are optimized according to the score results.

5. The method according to claim 1, characterized in that, The step of adjusting the computing resources of the power wireless terminal according to the set of computing resource configuration strategies, and generating resource allocation verification results and terminal operation performance feedback data, includes: The resource allocation priority adjustment strategy in the set of computing resource configuration strategies is analyzed to determine the list of terminal devices to be adjusted and the corresponding computing core reallocation parameters. Based on the task dependency chain in the task scheduling optimization strategy, the current task queue is restructured to generate an optimized task execution order table. Based on the bandwidth adjustment ratio in the network bandwidth reallocation strategy, the network interface configuration of the target terminal device is modified in real time, and changes in network throughput are monitored. After the adjustment operation is performed, the real-time resource utilization rate, task processing completion rate and network transmission error rate of the terminal device are collected, and resource allocation verification results are generated. The resource allocation verification results are compared and analyzed with the historical performance data before the adjustment. The resource utilization improvement rate and task processing acceleration rate are extracted and packaged into terminal operation performance feedback data.

6. The method according to claim 5, characterized in that, The step of reconstructing the current task queue according to the task dependency chain in the task scheduling optimization strategy to generate an optimized task execution order table includes: Identify the critical path task nodes and non-critical path task nodes in the task dependency relationship chain, and add a priority execution flag to the critical path task nodes; Based on the urgent task marker in the business priority identifier, tasks with the same priority are sorted a second time according to processing delay constraints. Detect circular dependency chains in the task queue and remove circular dependencies by inserting virtual synchronization nodes; Based on the characteristics of task processing resource requirements, tasks that occupy the same computing resources are divided into time slices to generate a task time slice distribution map without resource conflicts. Based on the priority execution identifier, the secondary sorting result, and the time slice distribution map, the task queue is reconstructed and an optimized task execution order table is generated.

7. The method according to claim 1, characterized in that, The step of continuously comparing the terminal performance feedback data with a preset performance optimization threshold and iteratively updating the resource adaptation rule base based on the comparison results includes: Extract the resource utilization improvement ratio from the terminal operation performance feedback data and compare it with the lowest improvement threshold in the performance optimization threshold; If the resource utilization improvement rate is lower than the minimum improvement threshold, the rule base diagnosis mode is activated to perform anomaly detection on the load balancing rule group in the resource adaptation rule base. Extract historical records of rule matching failure events and analyze the degree of mismatch between the rule triggering conditions and the current terminal device status; Based on the matching deviation, the condition thresholds in the load balancing rule group are adaptively calibrated to generate an updated load balancing rule group. Re-inject the updated load balancing rule group into the resource adaptation rule base and disable the rule base diagnostic mode.

8. The method according to claim 7, characterized in that, The anomaly detection of the load balancing rule group in the resource adaptation rule base includes: Statistically analyze the number of times the load balancing rule group is triggered and the number of times it is successfully executed within a preset time period, and calculate the rule execution success rate. If the success rate of the rule execution is lower than the preset success rate threshold, the load balancing rule group is determined to be abnormal. Extract terminal device load data and resource allocation logs during abnormal time periods to identify rule conditions that were not successfully executed and the corresponding device status characteristics; Cluster analysis is used to divide the device status characteristics into multiple abnormal scenario categories, and rules are generated to supplement conditions for each abnormal scenario category. The supplementary conditions of the rules are added to the original load balancing rule group to form an expanded load balancing rule group.

9. A computer system, characterized in that, include: A memory, wherein a computer program is stored; A processor is configured to load the computer program to implement the dynamic configuration computing resource management method based on a power wireless terminal as described in any one of claims 1-8.

Images