Intelligent resource scheduling and supplying method and system in heterogeneous computing environment

Abstract

The invention relates to the technical field of resource scheduling, and discloses an intelligent resource scheduling supply method and system in a heterogeneous computing environment, and the method comprises the steps: obtaining core heterogeneous data, carrying out the exponential smoothing denoising, obtaining a smooth state value, combining an HAT extension tuple, and carrying out the analysis to obtain a comprehensive capability score; the contribution degree is automatically adjusted through the similarity between the static features and the dynamic features; constructing a coupling correction function, calculating a mixing precision matching degree, and reflecting nonlinear correlation between targets in combination with an exponential product; a dynamic penalty function is constructed, a quantum genetic algorithm fitness function is optimized, and multi-objective collaborative optimization is carried out; constructing a complexity coupling and dynamic adaptation mechanism, and dynamically adjusting a task segmentation proportion; optimizing FPGA bit stream preloading according to the index ratio of the design frequency to the reconfiguration time; constructing a synchronous delay coupling model, and performing state synchronization optimization; balancing multi-dimensional feedback through a combined reward function; the influence of decision factors is quantified through contribution degree entropy, and interpretability is improved.

Description

Technical Field

[0001] This invention relates to the field of resource scheduling technology, specifically to an intelligent resource scheduling and supply method and system in a heterogeneous computing environment. Background Technology

[0002] In existing technologies and patents, unified resource pools mostly focus on logical aggregation, failing to address the core issue of differences in physical resource characteristics. For example, Kubernetes uses DevicePlugin to access GPUs and FPGAs, supporting only whole-card / shard allocation. It cannot convert CPU computing cores, GPU CUDA cores, and FPGA logic unit LUTs into a unified computing power unit, making it impossible for the scheduler to recognize the differences in computing power cost-effectiveness between different hardware. While the intermediate instruction layer masks underlying differences, it does not optimize the compilation efficiency of virtual instructions to hardware native instructions. In industrial vision inspection scenarios, the instruction conversion time accounts for a higher proportion of the total task time than homogeneous scheduling. There is a time lag between computing power resource status acquisition and scheduling decisions, making it difficult for prediction models based on historical data to capture the resource demands of sudden tasks, resulting in insufficient or excessive resource reservation. Some dynamic scheduling strategies improve resource utilization but do not incorporate energy consumption into the objective function, leading to a decrease in energy efficiency under high load scenarios. Real-time tasks are sensitive to latency, but existing patents do not clearly distinguish task priorities, potentially causing critical tasks to be preempted by low-priority tasks. Therefore, there is a need to provide intelligent resource scheduling and supply methods and systems for heterogeneous computing environments. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent resource scheduling and supply method and system in a heterogeneous computing environment. To solve the above-mentioned problems in the prior art, this invention achieves this through the following technical solution: Firstly, the intelligent resource scheduling and supply method in a heterogeneous computing environment provided by embodiments of the present invention specifically includes the following steps: Step 1: Collect heterogeneous data from the heterogeneous computing cluster to obtain core heterogeneous data, perform exponential smoothing and denoising to obtain smoothed state values, combine HAT extended tuples, use geometric mean fusion index to analyze and obtain comprehensive capability score, and determine whether the communication performance of the heterogeneous computing cluster meets the requirements. Step 2: For heterogeneous computing clusters with compatible communication performance, the contribution is automatically adjusted by using the similarity between static and dynamic features through attention fusion and feature coupling mechanisms; a coupling correction function is constructed to calculate the mixed precision matching degree and determine the suitability between the task and the hardware precision. Step 3: For heterogeneous computing clusters that are adapted to the task and hardware precision, construct a multi-objective coupling penalty mechanism, and combine exponential product to reflect the nonlinear correlation between objectives; construct a dynamic penalty function, optimize the fitness function of the quantum genetic algorithm, and perform multi-objective collaborative optimization; Step 4: Based on the completion of collaborative optimization, construct a complexity coupling and dynamic adaptation mechanism, and dynamically adjust the task partitioning ratio using the nonlinear product of data volume and latency sensitivity; design the frequency and reconfiguration time exponential ratio to optimize FPGA bitstream preloading; construct a synchronous delay coupling model and perform state synchronization optimization. Step 5: After the state synchronization optimization is completed, construct a reward coordination and loss linkage mechanism, balance multi-dimensional feedback by combining reward functions, and quantify the impact of decision factors by contribution entropy value to improve interpretability.

[0004] A further invention provides that the method for obtaining the comprehensive ability score is as follows: Exponential smoothing denoising is performed on the core heterogeneous data to preserve real-time performance while suppressing high-frequency noise. The original core heterogeneous data at time t and the original core heterogeneous data at time t-1 are fused according to preset weights to calculate the smoothed state value and sort it. Based on the obtained smooth state values, construct HAT tuples, assign architecture identifier codes to different hardware types, and generate globally unique resource IDs by combining HAT tuples. For any smooth state value sequence, calculate the lower quartile and upper quartile values ​​of the smooth state value sequence, and calculate the difference between the upper quartile and lower quartile values ​​to obtain the anomaly threshold; If the smoothed state value is less than or greater than the outlier threshold, the median of the sequence is used instead of the smoothed state value. Combining HAT extended tuples and utilizing the geometric mean fusion index, the formula is: Calculate the comprehensive ability score ,in, To smooth the total number of dimensions of the state values, For dimensional indexing, The k-th dimension smoothed state value after preprocessing. To find the theoretical minimum extreme value, This is the maximum extremum of the standard theory; A further invention, wherein the method for automatically adjusting the contribution level is as follows: For static features, obtain task submission information and extract static features; For dynamic features, monitor and acquire dynamic features during task execution; Dependency features are obtained through task DAG graph analysis; By employing an attention fusion and feature coupling mechanism, and using attention similarity fusion, key features are automatically highlighted through the formula: Calculate the task profile ,in, For static feature vectors, The cosine similarity between static and dynamic features ranges from [0,1]. The average value of the dynamic characteristics. For dependent feature vectors, Due to the complexity of the dependency, and ; If the similarity between static and dynamic features is high, then the contribution ratio of static features is high; if the dependency complexity is high, then the contribution ratio of dependency feature vectors is low. A further invention, wherein the method for calculating the mixed precision matching degree is as follows: Obtain the hardware-supported accuracy and the task's theoretical accuracy, quantify the compatibility between the task and hardware accuracy through mixed-precision matching degree calculation, and reflect the non-linear correlation between hardware-supported accuracy and task-theoretical accuracy, using the formula: Calculate the mixed precision matching degree ,in, For the theoretical accuracy of the task, To ensure hardware accuracy, Let be the accuracy loss rate of task i on hardware j; If the hardware-supported accuracy perfectly matches the theoretical accuracy of the task, then The greater the accuracy deviation or the higher the accuracy loss rate, the smaller the mixed accuracy matching degree. The obtained mixed precision matching degree is compared with the preset matching degree threshold to analyze and determine the heterogeneous computing cluster that is compatible with the task and hardware precision. A further invention, the method for constructing the multi-objective coupling penalty mechanism is as follows: Construct a multi-objective coupled penalty mechanism: Objective constraints: extract the total latency upper limit and power budget from the task SLA; obtain the resource utilization safety threshold from the system configuration; Predictive data: A random forest model is trained using historical scheduling records to predict the latency of task i on resource j. and energy consumption ; Decision variables: decision Where 1 indicates that task i is assigned to resource j, and 0 indicates that task i is not assigned to resource j, the maximum utilization rate of the resource pool is obtained. ; A further invention, the method for optimizing the fitness function of the quantum genetic algorithm is as follows: By integrating latency, energy consumption, overload, and accuracy loss, a dynamic loss function is calculated to avoid subjective weighting, using the formula: The dynamic loss was calculated. ,in, For the total prediction delay, and , For the total SLA latency budget, For total predicted energy consumption, and , For the power budget of the resource pool, As an overload penalty, and , For accuracy penalty, and ; Based on the obtained dynamic loss, the reciprocal of the dynamic loss is used as the fitness to construct an optimized quantum genetic algorithm fitness function, guiding the optimized quantum genetic algorithm to quickly converge to the optimal solution, through the formula: The fitness of the quantum genetic algorithm is calculated. The quantum genetic algorithm searches for the optimal allocation scheme through qubit encoding; A further invention, the method for optimizing FPGA bitstream preloading is as follows: Dynamically adjust the edge processing ratio to balance latency and transmission cost, using the formula: The proportion of edge tasks was calculated. ,in, To transmit data volume, For dynamic data thresholds; Optimize the preloading strategy to reduce reconfiguration time, obtain the historical usage frequency of bitstream b and combine it with the exponential ratio of cluster reconfiguration time to minimum cluster reconfiguration time to obtain bitstream priority; Bitstreams that are used frequently and reconfigured quickly have higher priority and are preloaded into the FPGA cache to reduce reconfiguration time. A further invention, wherein the method for performing state synchronization optimization is as follows: To balance synchronization latency and data reliability, optimize state synchronization: in, To verify the strength, To synchronize time, Error rate; A further invention provides that the method for quantifying the influence of decision factors is as follows: A combined reward function is constructed to balance multi-dimensional feedback, comprehensively considering feedback rewards based on latency, energy consumption, reliability, and accuracy to reflect a balance relationship. The reward value is then calculated using a formula. ; Calculate the entropy value of the contribution of each factor to the scheduling decision, quantify the influence weight, and use the formula: The contribution entropy value was calculated. ,in, The contribution percentage of factor f in decision-making is calculated using attention weights. The lower the contribution entropy value, the more likely the factor is to dominate the decision-making process, thus improving interpretability. By using the capability index and task profile output by resource perception as scheduling input, a multi-objective decision-making process is used to generate an allocation scheme. After cross-domain collaborative execution, feedback data is collected, and the strategy is updated through a reward function to achieve intelligent adaptive scheduling of resources in heterogeneous environments.

[0005] Secondly, the intelligent resource scheduling and supply system in a heterogeneous computing environment provided in the embodiments of the present invention specifically includes the following modules: Data Acquisition and Fusion Module: Collects heterogeneous data from the heterogeneous computing cluster, obtains core heterogeneous data, performs exponential smoothing and noise reduction to obtain smoothed state values, combines HAT extended tuples, uses geometric mean fusion index to analyze and obtain comprehensive capability scores, and judges whether the communication performance of the heterogeneous computing cluster meets the requirements. Adjustment and correction module: For heterogeneous computing clusters with compatible communication performance, the contribution is automatically adjusted by using the similarity between static and dynamic features through attention fusion and feature coupling mechanism; a coupling correction function is constructed to calculate the mixed precision matching degree and determine the suitability between task and hardware precision. Collaborative optimization module: For heterogeneous computing clusters adapted to task and hardware precision, a multi-objective coupling penalty mechanism is constructed, which combines exponential product to reflect the nonlinear correlation between objectives; a dynamic penalty function is constructed, the fitness function of the quantum genetic algorithm is optimized, and multi-objective collaborative optimization is performed; Coupling and Adaptation Module: Based on the completion of collaborative optimization, a complexity coupling and dynamic adaptation mechanism is constructed. The task segmentation ratio is dynamically adjusted using the nonlinear product of data volume and latency sensitivity. The frequency and reconfiguration time exponential ratio are designed to optimize FPGA bitstream preloading. A synchronous delay coupling model is constructed to optimize state synchronization. Feedback Decision Module: Based on the completion of state synchronization optimization, a reward collaboration and loss linkage mechanism is constructed, and multi-dimensional feedback is balanced by combining reward functions; the influence of decision factors is quantified by contribution entropy value to improve interpretability.

[0006] The beneficial effects of this invention are: 1. Dynamically time-warped and linearly interpolated multidimensional heterogeneous data are applied, and the comprehensive hardware capability is quantified using a geometric mean fusion index, highlighting the impact of bottleneck indicators such as communication latency. A HAT extended tuple is designed to generate globally unique resource IDs, enabling fast cross-platform resource retrieval. An attention similarity fusion model is proposed, dynamically adjusting the contribution ratio of static and dynamic features through cosine similarity to avoid redundant calculations. A nonlinear matching function is constructed to convert the deviation between hardware-supported accuracy and task theoretical accuracy into a quantifiable loss rate, identifying the suitability for accuracy-sensitive tasks. Multi-objective collaborative optimization of tasks is performed, and dependent features are extracted from the DAG multi-objective collaborative optimization graph. Combined with dynamic complexity indicators, the scheduling strategy is dynamically adjusted to adapt to changes in task structure and reduce scheduling latency. 2. A dynamic loss function is designed to fuse latency, energy consumption, overload, and accuracy losses through exponential product, avoiding subjective weight allocation. A fitness function based on the reciprocal of the dynamic loss is proposed, combined with qubit encoding and rotating gate operation, to accelerate convergence while maintaining population diversity. An overload penalty term is automatically triggered to balance task allocation and hardware stability. The task partitioning ratio is dynamically adjusted based on the nonlinear product of data volume and latency sensitivity. A frequency-reconfiguration time exponential ratio optimization preloading strategy is proposed to prioritize high-frequency bitstreams with fast reconfiguration. A correlation function between verification strength, synchronization time, and error rate is constructed to ensure data reliability. A reward signal is designed by comprehensively considering task latency, energy consumption, success rate, and accuracy losses to achieve multi-objective balance. The influence weight of each factor on decision-making is quantified through entropy analysis. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a flowchart of the steps of the intelligent resource scheduling and supply method in a heterogeneous computing environment provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the structure of the intelligent resource scheduling and supply system in a heterogeneous computing environment provided in Embodiment 2 of the present invention. Detailed Implementation

[0009] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0010] Example 1: As Figure 1 As shown in the embodiments of the present invention, the intelligent resource scheduling and supply method and system in a heterogeneous computing environment specifically includes the following steps: Step 1: Collect heterogeneous data from the heterogeneous computing cluster to obtain core heterogeneous data, perform exponential smoothing and noise reduction to obtain smoothed state values, combine HAT extended tuples, use geometric mean fusion index to analyze and obtain a comprehensive capability score to determine whether the communication performance of the heterogeneous computing cluster meets the requirements. In a specific embodiment, heterogeneous data is collected from CPUs, GPUs, FPGAs and edge MCUs in a heterogeneous computing cluster. A distributed perception agent is deployed to cover the heterogeneous resources of CPUs, GPUs, FPGAs and edge MCUs, and N types of core heterogeneous data are collected. The core heterogeneous data includes, but is not limited to: computing power, memory bandwidth, communication latency, power consumption, utilization rate and temperature. Real-time data transmission is achieved through Kafka message queues, ensuring that timestamp errors meet the preset error range; Linear interpolation is performed on core heterogeneous data of different frequencies to unify the granularity, using the formula: The core heterogeneous data at time t were calculated. ,in, for, for, for, for, for, To address the issue of time synchronization problems in multi-source data; Exponential smoothing denoising is performed on the core heterogeneous data to preserve real-time performance while suppressing high-frequency noise. The original core heterogeneous data at time t and the original core heterogeneous data at time t-1 are fused according to preset weights to calculate the smoothed state value and sort it. Based on the obtained smooth state values, HAT tuples are constructed to assign architecture identifier codes to different hardware types. For example, CPU=001, GPU=002, FPGA=003. Combined with HAT tuples, globally unique resource IDs are generated to realize cross-platform resource indexing. The interquartile range is used to replace the 3σ method to adapt to smooth state values ​​that are not normally distributed. For any smooth state value sequence, the lower quartile and upper quartile values ​​of the smooth state value sequence are calculated. The difference between the upper quartile and lower quartile values ​​is calculated to obtain the anomaly threshold. If the smoothed state value is less than or greater than the outlier threshold, the median of the sequence is used instead of the smoothed state value. Combining HAT extended tuples and utilizing the geometric mean fusion index, the weakest link effect is reflected through the formula: Calculate the comprehensive ability score ,in, To smooth the total number of dimensions of the state values, For dimensional indexing, The k-th dimension smoothed state value after preprocessing. To find the theoretical minimum extreme value, This is the maximum extremum of the standard theory; For example, if the GPU's 10 normalized smoothed state values ​​are [0.8, 0.7, 0.9, 0.85, 0.75, 0.82, 0.95, 0.6, 0.88, 0.78], then the overall capability score is obtained by rounding down to one decimal place. If the normalized value of the communication delay drops to 0.3, then It keenly reflects the deterioration of communication performance; Step 2: For heterogeneous computing clusters with compatible communication performance, the contribution is automatically adjusted by using the similarity between static and dynamic features through attention fusion and feature coupling mechanisms; a coupling correction function is constructed to calculate the mixed precision matching degree and determine the suitability between the task and the hardware precision. In a specific embodiment, for static features, task submission information is obtained and static features are extracted. Static features include: task type, amount of input data, instruction set dependency, and theoretical accuracy requirement. For dynamic features, dynamic features are monitored and obtained during task runtime. Dynamic features include: real-time CPU utilization, memory consumption, and I / O request frequency. Dependency features are obtained through task DAG graph analysis. These features include: the number of subtasks extracted, the total amount of data dependencies, and the level of parallelism. By employing an attention fusion and feature coupling mechanism, and using attention similarity fusion, key features are automatically highlighted through the formula: Calculate the task profile ,in, For static feature vectors, The cosine similarity between static and dynamic features ranges from [0,1]. The average value of the dynamic characteristics. For dependent feature vectors, Due to the complexity of the dependency, and ; If the similarity between static and dynamic features is high, then the contribution ratio of static features is high; if the dependency complexity is high, then the contribution ratio of dependency feature vectors is low, so as to avoid excessive interference from dependency features. For example, if the similarity between static and dynamic features in an AI training task is 0.85, then the contribution of static features is 85%. If the complexity depends on the dynamic feature, then the contribution of static features is 85%. Calculating the value to three decimal places, the contribution ratio of the dependent feature vector is obtained from the analysis. ; Obtain the hardware-supported accuracy and the task's theoretical accuracy, quantify the compatibility between the task and hardware accuracy through mixed-precision matching degree calculation, and reflect the non-linear correlation between hardware-supported accuracy and task-theoretical accuracy, using the formula: Calculate the mixed precision matching degree ,in, For the theoretical accuracy of the task, To ensure hardware accuracy, Let be the accuracy loss rate of task i on hardware j; If the hardware-supported accuracy perfectly matches the theoretical accuracy of the task, then The greater the accuracy deviation or the higher the accuracy loss rate, the smaller the mixed accuracy matching degree. The obtained mixed precision matching degree is compared with the preset matching degree threshold to analyze and determine the heterogeneous computing cluster that is compatible with the task and hardware precision. If the mixed precision matching degree is greater than or equal to the preset matching degree threshold, it indicates that the task of the heterogeneous computing cluster is compatible with the hardware precision; if the mixed precision matching degree is less than the preset matching degree threshold, it indicates that the task of the heterogeneous computing cluster is not compatible with the hardware precision. Step 3: For heterogeneous computing clusters that are adapted to the task and hardware precision, construct a multi-objective coupling penalty mechanism, and combine exponential product to reflect the nonlinear correlation between objectives; construct a dynamic penalty function to avoid rigidity of fixed thresholds; optimize the fitness function of quantum genetic algorithm to perform multi-objective collaborative optimization; In a specific embodiment, a multi-objective coupled penalty mechanism is constructed: objective constraints: extract the total latency upper limit and power budget from the task SLA; obtain the resource utilization safety threshold from the system configuration; Predictive data: A random forest model is trained using historical scheduling records to predict the latency of task i on resource j. and energy consumption ; Decision variables: decision Where 1 indicates that task i is assigned to resource j, and 0 indicates that task i is not assigned to resource j, the maximum utilization rate of the resource pool is obtained. ; By integrating latency, energy consumption, overload, and accuracy loss, a dynamic loss function is calculated to avoid subjective weighting, using the formula: The dynamic loss was calculated. ,in, For the total prediction delay, and , For the total SLA latency budget, For total predicted energy consumption, and , For the power budget of the resource pool, As an overload penalty, and , For accuracy penalty, and ; For example, total latency exceeding the total latency budget by 20% means... Energy consumption exceeding the power budget by 10% The exponent term is then rounded to two decimal places. If both are overloaded and Total dynamic loss In this case, latency and energy consumption should be optimized first. Based on the obtained dynamic loss, the reciprocal of the dynamic loss is used as the fitness to construct an optimized quantum genetic algorithm fitness function, guiding the optimized quantum genetic algorithm to quickly converge to the optimal solution, through the formula: The fitness of the quantum genetic algorithm is calculated. The smaller the dynamic loss DLF, the higher the fitness. The quantum genetic algorithm searches for the optimal allocation scheme through qubit encoding, thus stabilizing the decision delay. Step 4: Construct a complexity coupling and dynamic adaptation mechanism, dynamically adjust the task partitioning ratio using the nonlinear product of data volume and latency sensitivity; design the frequency and reconfiguration time exponential ratio to optimize FPGA bitstream preloading; construct a synchronous delay coupling model and perform state synchronization optimization. In a specific embodiment, the edge processing ratio is dynamically adjusted to balance latency and transmission cost, using the formula: The proportion of edge tasks was calculated. ,in, To transmit data volume, The dynamic data threshold is determined by the level of latency sensitivity; the higher the latency sensitivity, the smaller the dynamic data threshold. Optimize the preloading strategy to reduce reconfiguration time, obtain the historical usage frequency of bitstream b and combine it with the exponential ratio of cluster reconfiguration time to minimum cluster reconfiguration time to obtain bitstream priority; Bitstreams that are used frequently and reconfigured quickly have higher priority and are preloaded into the FPGA cache to reduce reconfiguration time. To balance synchronization latency and data reliability, optimize state synchronization: in, To verify the strength, To synchronize time, The longer the synchronization time or the higher the error rate, the stronger the verification strength. Step 5: Based on the completion of state synchronization optimization, construct a reward coordination and loss linkage mechanism, balance multi-dimensional feedback by combining reward functions, and quantify the impact of decision factors by contribution entropy value to improve interpretability; In a specific embodiment, feedback indicators are extracted from the execution log in real time. These feedback indicators include: actual task execution latency, power overrun ratio, task success rate, and total accuracy loss. A combined reward function is constructed to balance multi-dimensional feedback, comprehensively considering feedback rewards based on latency, energy consumption, reliability, and accuracy, reflecting a balance relationship. This is achieved through the formula: Calculate the reward value ,in, The actual execution delay of the task. The delay specified by the SLA, The percentage of power overrun. To improve the success rate of the task, This represents the total loss of accuracy. Calculate the entropy value of the contribution of each factor to the scheduling decision, quantify the influence weight, and use the formula: The contribution entropy value was calculated. ,in, The contribution percentage of factor f in decision-making is calculated using attention weights. The lower the contribution entropy value, the more likely the factor is to dominate the decision-making process, thus improving interpretability. By using the capability index and task profile output by resource perception as scheduling input, a multi-objective decision-making process is used to generate an allocation scheme. After cross-domain collaborative execution, feedback data is collected, and the strategy is updated through a reward function to achieve intelligent adaptive scheduling of resources in heterogeneous environments.

[0011] Example 2: Figure 2 As shown in the embodiment of the present invention, the intelligent resource scheduling and supply system in a heterogeneous computing environment specifically includes the following modules: Data Acquisition and Fusion Module: Collects heterogeneous data from the heterogeneous computing cluster, obtains core heterogeneous data, performs exponential smoothing and noise reduction to obtain smoothed state values, combines HAT extended tuples, uses geometric mean fusion index to analyze and obtain comprehensive capability scores, and judges whether the communication performance of the heterogeneous computing cluster meets the requirements. Adjustment and correction module: For heterogeneous computing clusters with compatible communication performance, the contribution is automatically adjusted by using the similarity between static and dynamic features through attention fusion and feature coupling mechanism; a coupling correction function is constructed to calculate the mixed precision matching degree and determine the suitability between task and hardware precision. Collaborative optimization module: For heterogeneous computing clusters adapted to task and hardware precision, a multi-objective coupling penalty mechanism is constructed, which combines exponential product to reflect the nonlinear correlation between objectives; a dynamic penalty function is constructed, the fitness function of the quantum genetic algorithm is optimized, and multi-objective collaborative optimization is performed; Coupling and Adaptation Module: Based on the completion of collaborative optimization, a complexity coupling and dynamic adaptation mechanism is constructed. The task segmentation ratio is dynamically adjusted using the nonlinear product of data volume and latency sensitivity. The frequency and reconfiguration time exponential ratio are designed to optimize FPGA bitstream preloading. A synchronous delay coupling model is constructed to optimize state synchronization. Feedback Decision Module: Based on the completion of state synchronization optimization, a reward collaboration and loss linkage mechanism is constructed, and multi-dimensional feedback is balanced by combining reward functions; the influence of decision factors is quantified by contribution entropy value to improve interpretability.

[0012] The above provides a detailed description of one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. The above formulas are all dimensionless numerical calculations, and the formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world situation. The preset parameters in the formulas are set by those skilled in the art based on actual conditions and historical experience, and can be adjusted according to actual conditions. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A method for intelligent resource scheduling and provisioning in a heterogeneous computing environment, the method comprising: The method comprises the following steps: ​ Heterogeneous data collection is performed on a heterogeneous computing cluster to obtain core heterogeneous data, and exponential smoothing denoising is performed to obtain a smoothed state value, a HAT extended tuple is combined, a geometric mean fusion index is used, and a comprehensive capability score is obtained through analysis to determine whether the communication performance of the heterogeneous computing cluster meets the requirements; For a heterogeneous computing cluster with communication performance meeting the requirements, the contribution degree is automatically adjusted by using the similarity of static features and dynamic features through an attention fusion and feature coupling mechanism; a coupling correction function is constructed to calculate the mixed precision matching degree and determine the adaptation of the task and hardware precision; For a heterogeneous computing cluster with task and hardware precision adaptation, a multi-objective coupling penalty mechanism is constructed, and the non-linear correlation between objectives is reflected by an exponential product; a dynamic penalty function is constructed to optimize the fitness function of the quantum genetic algorithm for multi-objective collaborative optimization; Based on the end of collaborative optimization, a complexity coupling and dynamic adaptation mechanism is constructed to dynamically adjust the task segmentation ratio by using the non-linear product of data volume and delay sensitivity; an FPGA bit stream preloading is designed by optimizing the exponential ratio of frequency and reconfiguration time; a synchronous delay coupling model is constructed for state synchronization optimization; Based on the end of state synchronization optimization, a reward collaboration and loss linkage mechanism is constructed to balance multi-dimensional feedback by combining a reward function; the influence of decision factors is quantified by the contribution degree entropy to improve the explainability. 2.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment according to claim 1, wherein, The method for obtaining the comprehensive capability score is: Exponential smoothing denoising is performed on the core heterogeneous data to retain real-time performance while suppressing high-frequency noise; the smoothed state value is calculated by fusing the original core heterogeneous data at time t with the original core heterogeneous data at time t-1 according to a preset weight, and then sorted; Based on the obtained smoothed state value, a HAT tuple is constructed to assign an architecture identification code to different hardware types, and a globally unique resource ID is generated by combining the HAT tuple; For any smoothed state value sequence, the lower quartile value and the upper quartile value of the smoothed state value sequence are calculated, and the difference between the upper quartile value and the lower quartile value is calculated to obtain an abnormal threshold value; If the smoothed state value is less than or greater than the abnormal threshold value, the median of the sequence is used to replace the smoothed state value; By combining the HAT extended tuple and using the geometric mean fusion index, the comprehensive capability score is obtained through the formula: The comprehensive ability score is calculated wherein, is the total number of dimensions of smoothed state values, is the dimension index, is the kth dimension smoothed state value after preprocessing, is the minimum extreme value of the tag theory, is the maximum extreme value of the tag theory. 3.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for automatically adjusting the contribution degree is: For static features, task submission information is obtained and static features are extracted; For dynamic features, dynamic features are obtained by monitoring during task execution; Dependency features are obtained by analyzing the task DAG graph; Through the attention fusion and feature coupling mechanism, the key features are automatically highlighted by using attention similarity fusion through the formula: Computing the task profile wherein, is the static feature vector, is the cosine similarity between static and dynamic features, ranging from [0, 1], is the average of the dynamic features, is the dependency feature vector, is the dependency complexity, and ; If the similarity of static and dynamic features is high, the contribution proportion of static features is high; If the dependency complexity is high, the contribution proportion of the dependency feature vector is low. 4.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for calculating the mixed precision matching degree is: The hardware support precision and the task theory precision are obtained, the adaptability of the quantitative task and the hardware precision is calculated through the mixed precision matching degree, the nonlinear correlation of the hardware support precision and the task theory precision is embodied, and the mixed precision matching degree is calculated through the formula: , wherein, is the task theory precision, is the hardware support precision, is the precision loss rate of the task i on the hardware j.​ If the hardware support precision and the task theory precision are completely matched, then The greater the precision deviation or the higher the precision loss rate, the smaller the mixed precision matching degree. The obtained mixed precision matching degree is compared with a preset matching degree threshold to determine the heterogeneous computing cluster with task and hardware precision adaptation. 5.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for constructing the multi-objective coupling penalty mechanism is: A multi-objective coupling penalty mechanism is constructed: target constraints: the total time delay upper limit and power budget are extracted from the task SLA; the resource utilization safety threshold is obtained from the system configuration; Predicted data: train random forest model with historical scheduling records to predict the latency of task i on resource j and energy consumption ; Decision variable: decision where 1 indicates that task i is assigned to resource j and 0 indicates that task i is not assigned to resource j, to obtain the maximum utilization of the resource pool . 6.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for optimizing the fitness function of the quantum genetic algorithm is: Fusion time delay, energy consumption, overload, precision loss, calculate dynamic loss function, avoid subjective weight, through formula: The dynamic loss is calculated wherein, is the total predicted latency, and , is the SLA total latency budget, is the total predicted energy consumption, and , is the power budget of the resource pool, is the overload penalty, and , is the accuracy penalty, and ; Based on the obtained dynamic loss, the reciprocal of the dynamic loss is used as the fitness to construct an optimized quantum genetic algorithm fitness function, guiding the optimized quantum genetic algorithm to quickly converge to the optimal solution, through the formula: The fitness of the quantum genetic algorithm is calculated. The quantum genetic algorithm searches for the optimal allocation scheme by encoding qubits. 7.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for optimizing the FPGA bit stream preloading is: Dynamic adjustment of edge processing ratio, balance delay and transmission cost, through formula: The edge task proportion is calculated wherein, is the transmission data amount, is the dynamic data threshold value; Optimize the preloading strategy, reduce the reconfiguration time, get the historical usage frequency of the bit stream b and combine the exponential ratio of the cluster reconfiguration time and the cluster minimum reconfiguration time to get the bit stream priority; High frequency and fast reconfiguration bit stream priority, preloaded to FPGA cache, reduce the reconfiguration time. 8.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for optimizing state synchronization is: Balancing synchronization delay and data reliability, state synchronization optimization: wherein, is a check strength, is a synchronization time, is an error rate. 9.The intelligent resource scheduling and provisioning method in a heterogeneous computing environment of claim 1, wherein, The method for quantifying the influence of decision factors is: The combined reward function balances multi-dimensional feedback, and comprehensively reflects the feedback rewards of time delay, energy consumption, reliability and accuracy, embodies the balance relationship, and obtains a reward value through formula calculation ; Calculate the contribution entropy value of each factor to the scheduling decision, quantify the influence weight, through formula: The contribution entropy value is calculated wherein, is the contribution proportion of the factor f in the decision, calculated by the attention weight, is the total number of factors, and the lower the contribution entropy value is, the more the factor dominates the decision, thereby improving the explainability. Through the ability index output by resource perception and the task image as the scheduling input, generate the allocation scheme through multi-objective decision, collect feedback data after cross-domain collaborative execution, update the strategy through the reward function, realize intelligent adaptive scheduling of resources in heterogeneous environment.

10. An intelligent resource scheduling provisioning system in a heterogeneous computing environment, the provisioning system configured to perform the provisioning method of any one of claims 1-9, wherein: It includes: Acquisition fusion module: collect heterogeneous data of the heterogeneous computing cluster, obtain core heterogeneous data, and perform exponential smoothing denoising to obtain smooth state value, combine HAT extended tuple, use geometric mean fusion index to analyze and obtain comprehensive ability score, and judge whether the communication performance of the heterogeneous computing cluster meets the requirements; Adjustment correction module: for the heterogeneous computing cluster with communication performance meeting the requirements, automatically adjust the contribution degree through attention fusion and feature coupling mechanism by using the similarity of static features and dynamic features; construct a coupling correction function to calculate the mixed precision matching degree and judge the adaptation of the task and hardware precision; Collaborative optimization module: for the heterogeneous computing cluster with task and hardware precision adaptation, construct a multi-objective coupling penalty mechanism, combine the exponential product to reflect the nonlinear correlation between targets; construct a dynamic penalty function, optimize the fitness function of the quantum genetic algorithm, and perform multi-objective collaborative optimization; Coupling adaptation module: based on the end of collaborative optimization, construct a complexity coupling and dynamic adaptation mechanism to dynamically adjust the task segmentation ratio with the nonlinear product of data volume and delay sensitivity; design the frequency and reconfiguration time exponential ratio to optimize the FPGA bit stream preloading; construct a synchronization delay coupling model to optimize state synchronization; Feedback decision module: based on the end of state synchronization optimization, construct a reward collaboration and loss linkage mechanism to balance multi-dimensional feedback through combined reward function; quantify the influence of decision factors through contribution entropy value to improve explainability.

Images