An industrial internet workflow task scheduling method based on dependency perception and dynamic priority

Abstract

The application relates to the technical field of computers, in particular to an industrial internet workflow task scheduling method based on dependence perception and dynamic priority. In a preset task time window, the length of a ready task queue, a chip junction temperature, a CPU current frequency, CPU utilization, and the number of unfinished tasks and the total number of tasks of each workflow are synchronously acquired and pretreated; based on the pretreated length of the ready task queue and a preset stable queue length threshold value, a scheduling pressure coefficient is obtained, based on the pretreated number of unfinished tasks and the total number of tasks of each workflow, task dependence tightness is calculated, based on the scheduling pressure coefficient and the task dependence tightness, a scheduling-dependence pressure index is obtained, and based on the pretreated chip junction temperature, CPU current frequency and CPU utilization. The application significantly improves the robustness, stability and individualized adaptation capability of the workflow scheduling system in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of industrial internet technology, and in particular to an industrial internet workflow task scheduling method based on dependency awareness and dynamic priority. Background Technology

[0002] Workflow task scheduling technology is widely used in fields such as Industrial Internet of Things (IIoT), data centers, and cloud computing to manage sets of tasks with data dependencies, ensuring that tasks are executed correctly according to their dependencies and meet real-time requirements. In practical applications, workflow scheduling systems often face the complex situation of high-concurrency task flows and peak heat dissipation of computing resources occurring simultaneously. For example, in real-time device monitoring scenarios in IIoT, when multiple devices trigger alarms simultaneously, the system submits a large number of workflow instances in a very short time, forming a high-concurrency task flow; at the same time, computing resources quickly enter a fully loaded state, chip junction temperatures rise, triggering frequency reduction mechanisms, resulting in a decrease in actual processing capacity. These two problems exacerbate each other, forming a vicious cycle of "the busier, the slower."

[0003] The existing technologies have the following shortcomings: First, the scheduling methods for high-concurrency task flows do not consider the dynamic performance degradation of computing resources due to heat dissipation, resulting in a mismatch between scheduling decisions and actual available computing power; second, the power management methods for heat dissipation do not consider the impact of high-concurrency task flows on resource contention, and the frequency reduction decision is independent of the upper-level scheduling strategy; third, the existing technologies lack systematic modeling and coping strategies for the interaction mechanism of the above two types of problems, and cannot identify vicious cycle trends, leading to the loss of control over workflow completion time. Summary of the Invention

[0004] The main objective of this invention is to provide an industrial internet workflow task scheduling method based on dependency awareness and dynamic prioritization. First, by constructing a scheduling-dependency pressure index and a thermal-efficiency synergistic decay index, the combined risks of scheduling pressure and dependency structure, and thermal stress and performance degradation, are quantified respectively. Then, a system synergistic risk index is used to uniformly characterize the interaction mechanism of these two types of risks. Second, by dividing risk regions using preset thresholds and employing a piecewise nonlinear mapping function, differentiated risk responses are achieved: low-risk situations are handled smoothly, while high-risk situations are rapidly prioritized, balancing sensitivity and stability. Third, a closed-loop feedback mechanism feeds the execution results back to the next cycle, enabling adaptive optimization of the scheduling strategy and solving the problems of lack of intervention effectiveness assessment and inability to form closed-loop optimization. This invention achieves unified quantification and adaptive scheduling of the "congestion-overheating" vicious cycle, effectively breaking the vicious cycle of "the busier, the slower," and significantly improving the robustness, stability, and personalized adaptability of the workflow scheduling system in complex environments.

[0005] The technical solution of the present invention is as follows: Firstly, a task scheduling method for industrial internet workflows based on dependency awareness and dynamic priority is proposed, which includes the following steps: S1: Within the preset task time window, synchronously acquire the length of the ready task queue, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total number of tasks for each workflow, and perform preprocessing. S2: Based on the length of the preprocessed ready task queue and the preset stable queue length threshold, the scheduling pressure coefficient is obtained. Based on the number of unfinished tasks and the total number of tasks in each preprocessed workflow, the task dependency tightness is calculated. Based on the scheduling pressure coefficient and the task dependency tightness, the scheduling-dependency pressure index is obtained. Based on the preprocessed chip junction temperature, the current CPU frequency, and the CPU utilization, the thermal-efficiency synergistic decay index is calculated. Based on the scheduling-dependency pressure index and the thermal-efficiency synergistic decay index, the system synergistic risk index is calculated. S3: Compare the system collaborative risk index with the preset threshold, divide the risk areas according to the comparison results, and use a piecewise nonlinear mapping function to calculate the task scheduling evaluation index; S4: Based on the task scheduling evaluation index, make scheduling decisions on the ready task queue according to the preset decision rules, execute tasks and update the system status, and use the updated system status as the input for the next scheduling cycle, and return to step S1 to realize the adaptive optimization of the closed-loop scheduling strategy.

[0006] A further improvement of the present invention is that step S2 includes the following specific steps: S21: Based on the length of the preprocessed ready task queue With preset stable queue length threshold Calculate the scheduling pressure coefficient The formula is: ; S22: Based on each preprocessed workflow Number of unfinished tasks Total number of tasks Computation task dependency tightness The formula is: ; in, The set of workflows to which the currently ready tasks belong. For set The number of workflows in the middle; S23: Based on scheduling pressure coefficient and task dependency tightness Obtain the scheduling-dependency pressure index The formula is: ; in, This is the first adjustment coefficient.

[0007] A further improvement of the present invention is that step S2 further includes the following specific steps: S24: Based on the pre-processed chip junction temperature CPU current frequency and CPU utilization Calculate the thermal-efficiency synergistic decay index The formula is: ; in, The coefficient of thermal stress, To calculate the performance degradation rate, This is the second adjustment coefficient. It is a constant.

[0008] A further improvement of the present invention is that step S2 further includes the following specific steps: S25: Based on the scheduling-dependency pressure index and thermal-efficiency synergistic decay index Calculation system collaborative risk index The formula is: .

[0009] A further improvement of the present invention is that step S3 includes the following specific steps: S31: Preset thresholds specifically include low-risk thresholds. Medium risk threshold and high-risk threshold ,and ; S32: System Coordination Risk Index Separate and low risk thresholds Medium risk threshold and high-risk threshold Compare them and divide them into corresponding risk areas. At that time, it was classified as a low-risk area. At that time, it was classified as a medium-risk area. At that time, it was classified as a high-risk area. At that time, it was designated as a critical zone.

[0010] A further improvement of the present invention is that step S3 further includes the following specific steps: S33: Based on the divided risk areas, the task scheduling evaluation index is calculated using a piecewise nonlinear mapping function. The formula is as follows: ; in, Based on the priority offset, , as well as For mapping coefficients, This is the highest priority limit.

[0011] A further improvement of the present invention is that step S4 includes the following specific steps: S41: The preset decision rule is to allocate tasks in the ready task queue according to the task scheduling evaluation index. Sort in descending order, select The task with the highest value is designated as the task to be executed in the current scheduling cycle. S42: Dispatch the selected task to an idle computing core for execution, and update the workflow to which the task belongs after the task is completed. Number of unfinished tasks And add the satisfied subsequent tasks to the ready task queue; S43: Use the updated ready task queue length, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total number of tasks for each workflow as input for the next scheduling cycle, and return to step S1.

[0012] A further improvement of the present invention is that the adaptive adjustment step in S4 includes: adjusting according to the system collaborative risk index. The first adjustment coefficient for the short-term rate of change is dynamically adjusted. Second adjustment coefficient The value is obtained by performing linear regression to calculate the slope; when the short-term rate of change exceeds a preset rate of change threshold... At that time, and Simultaneously multiplied by a preset increment coefficient ,in , The scheduling period is [number].

[0013] Secondly, a computer-readable storage medium is proposed, on which a computer program is stored. When the computer program is executed by a processor, it implements the aforementioned industrial internet workflow task scheduling method based on dependency awareness and dynamic priority.

[0014] Thirdly, an electronic device is proposed, including a memory for storing instructions and a processor for executing the instructions, so that the device executes an industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described above.

[0015] The technical effects of this invention are as follows: This paper proposes an industrial internet workflow task scheduling method based on dependency awareness and dynamic prioritization. First, it quantifies the combined risks of scheduling pressure and dependency structure, and thermal stress and performance degradation, respectively, by constructing a scheduling-dependency pressure index and a thermal-efficiency synergistic decay index. Then, a system synergistic risk index is used to uniformly characterize the interaction mechanism of these two types of risks. Second, risk regions are divided by preset thresholds and a piecewise nonlinear mapping function is used to achieve differentiated risk responses: low-risk situations are handled smoothly, while high-risk situations are rapidly prioritized, balancing sensitivity and stability. Third, a closed-loop feedback mechanism feeds the execution results back to the next cycle, enabling adaptive optimization of the scheduling strategy and solving the problems of lack of intervention effectiveness assessment and inability to form closed-loop optimization. This invention achieves unified quantification and adaptive scheduling of the "congestion-overheating" vicious cycle, effectively breaking the vicious cycle of "the busier, the slower," and significantly improving the robustness, stability, and personalized adaptability of the workflow scheduling system in complex environments. Attached Figure Description

[0016] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating an industrial internet workflow task scheduling method based on dependency awareness and dynamic priority, according to Embodiment 1 of the present invention. Detailed Implementation

[0017] Example 1 This embodiment constructs an industrial internet workflow task scheduling method based on dependency awareness and dynamic prioritization. First, by constructing a scheduling-dependency pressure index and a thermal-efficiency synergistic decay index, the combined risks of scheduling pressure and dependency structure, and thermal stress and performance degradation are quantified respectively. Then, a system synergistic risk index is used to uniformly characterize the interaction mechanism of these two types of risks. Second, risk regions are divided by preset thresholds and a piecewise nonlinear mapping function is used to achieve differentiated risk responses: low-risk situations are handled smoothly, while high-risk situations are rapidly prioritized, balancing sensitivity and stability. Third, a closed-loop feedback mechanism feeds the execution results back to the next cycle, achieving adaptive optimization of the scheduling strategy and solving the problems of lack of intervention effectiveness assessment and inability to form closed-loop optimization. This invention achieves unified quantification and adaptive scheduling of the "congestion-overheating" vicious cycle, effectively breaking the vicious cycle of "the busier, the slower," and significantly improving the robustness, stability, and personalized adaptability of the workflow scheduling system in complex environments.

[0018] An industrial internet workflow task scheduling method based on dependency awareness and dynamic priority, such as... Figure 1 As shown, the specific steps include the following: S1: Within the preset task time window, synchronously acquire the length of the ready task queue, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total tasks for each workflow, and perform preprocessing.

[0019] In this embodiment, a preset task time window is used to limit the time range for parameter collection within each scheduling cycle, ensuring that the collection behavior is periodic and synchronized with the scheduling decision. The start point of the time window is the end time of the previous scheduling cycle, and the end point is the start time of the current scheduling cycle decision. The window duration can be dynamically adjusted according to the system load characteristics, or pre-calibrated during system initialization based on hardware performance and business requirements.

[0020] Specifically, the preprocessing involves normalizing the length of the ready task queue, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and the total number of tasks for each workflow.

[0021] S2: Based on the length of the preprocessed ready task queue and the preset stable queue length threshold, the scheduling pressure coefficient is obtained. Based on the number of unfinished tasks and the total number of tasks for each preprocessed workflow, the task dependency tightness is calculated. Based on the scheduling pressure coefficient and the task dependency tightness, the scheduling-dependency pressure index is obtained. Based on the preprocessed chip junction temperature, current CPU frequency, and CPU utilization, the thermal-efficiency synergistic decay index is calculated. Based on the scheduling-dependency pressure index and the thermal-efficiency synergistic decay index, the system synergistic risk index is calculated.

[0022] In this embodiment, S2 includes the following specific steps: S21: Based on the length of the preprocessed ready task queue With preset stable queue length threshold Calculate the scheduling pressure coefficient The formula is: ; S22: Based on each preprocessed workflow Number of unfinished tasks Total number of tasks Computation task dependency tightness The formula is: ; in, The set of workflows to which the currently ready tasks belong. For set The number of workflows in the middle; S23: Based on scheduling pressure coefficient and task dependency tightness Obtain the scheduling-dependency pressure index The formula is: ; in, This is the first adjustment coefficient; S24: Based on the pre-processed chip junction temperature CPU current frequency and CPU utilization The thermal-efficiency synergistic decay index is calculated using the following formula: ; in, The coefficient of thermal stress, To calculate the performance degradation rate, This is the second adjustment coefficient. It is a constant; S25: Based on the scheduling-dependency pressure index and thermal-efficiency synergistic decay index Calculation system collaborative risk index The formula is: .

[0023] In this embodiment, a preset stable queue length threshold is used. This is the maximum allowable length of the ready task queue under normal system operation. After the system has been running for a period of time, historical ready queue length data is collected, and the 95th percentile or mean of the queue length under normal operation is added to three times the standard deviation as the value. The scheduling pressure coefficient quantifies the congestion level of the ready task queue within the current scheduling cycle, reflecting the intensity of the high-concurrency task flow impact on the system. It is read in real time through the scheduler's internal counter, recording the number of tasks in the current ready queue. The task dependency tightness quantifies the degree of distribution concentration of currently ready tasks in the incomplete workflow, reflecting the strength of the impact of the task dependency structure on the global completion time.

[0024] In this embodiment, the first adjustment coefficient is used to adjust the exponential amplification effect of the product of the scheduling pressure coefficient and the task dependency tightness, and determines the system's sensitivity to the recognition of the superposition state of "congestion-dependency depth". By simulating typical disturbance scenarios (such as combinations of different concurrency intensities and workflow depths), with the average workflow completion time and system stability as optimization objectives, the optimal value of the first adjustment coefficient is determined by methods such as grid search or Bayesian optimization.

[0025] In this embodiment, the thermal stress coefficient is used to quantify the relative intensity of the current thermal stress on the chip, reflecting the proportion of the chip junction temperature deviating from the ambient temperature relative to the safe operating margin. The calculation formula is as follows: ,in, For chip junction temperature, For ambient reference temperature, This is the highest safe operating junction temperature allowed for the chip.

[0026] In this embodiment, the efficiency degradation rate is used to quantify the actual processing capacity loss caused by thermal throttling or dynamic voltage frequency adjustment. The calculation formula is as follows: ,in For CPU utilization, for The current CPU frequency at any given moment.

[0027] In this embodiment, the second adjustment coefficient is used to adjust the synergistic sensitivity between the thermal stress coefficient and the calculated efficiency decay rate in the process of constructing the thermal-efficiency synergistic decay index, and its initial value is determined through experience or experiment.

[0028] S3: Compare the system collaborative risk index with the preset threshold, divide the risk areas according to the comparison results, and use a piecewise nonlinear mapping function to calculate the task scheduling evaluation index.

[0029] In this embodiment, S3 includes the following specific steps: S31: Preset thresholds specifically include low-risk thresholds. Medium risk threshold and high-risk threshold ,and ; S32: System Coordination Risk Index Separate and low risk thresholds Medium risk threshold and high-risk threshold Compare them and divide them into corresponding risk areas. At that time, it was classified as a low-risk area. At that time, it was classified as a medium-risk area. At that time, it was classified as a high-risk area. At that time, it was designated as a critical zone; S33: Based on the divided risk areas, the task scheduling evaluation index is calculated using a piecewise nonlinear mapping function. The formula is as follows: ; in, Based on the priority offset, , as well as For mapping coefficients, This is the highest priority limit.

[0030] In this embodiment, , as well as Before system deployment, the mapping coefficients are tested by simulating typical disturbance scenarios. The optimal combination of mapping coefficients is determined by using methods such as grid search or Bayesian optimization with the optimization objectives of task completion time, system stability, and trigger frequency.

[0031] S4: Based on the task scheduling evaluation index, make scheduling decisions on the ready task queue according to the preset decision rules, execute tasks and update the system status, and use the updated system status as the input for the next scheduling cycle, and return to step S1 to realize the adaptive optimization of the closed-loop scheduling strategy.

[0032] In this embodiment, S4 includes the following specific steps: S41: The preset decision rule is to allocate tasks in the ready task queue according to the task scheduling evaluation index. Sort in descending order, select The task with the highest value is designated as the task to be executed in the current scheduling cycle. S42: Dispatch the selected task to an idle computing core for execution, and update the workflow to which the task belongs after the task is completed. Number of unfinished tasks And add the satisfied subsequent tasks to the ready task queue; S43: Use the updated ready task queue length, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total number of tasks for each workflow as input for the next scheduling cycle, and return to step S1.

[0033] Specifically, the adaptive adjustment steps in S4 include: based on the system collaborative risk index The first adjustment coefficient for the short-term rate of change is dynamically adjusted. Second adjustment coefficient The value is obtained by performing linear regression to calculate the slope; when the short-term rate of change exceeds the preset rate of change threshold... At that time, and Simultaneously multiplied by a preset increment coefficient ,in , The scheduling period is [number].

[0034] In this embodiment, the preset decision rule specifically involves weighted fusion of the task scheduling evaluation index and the original priority of the task itself to obtain a comprehensive priority. The calculation formula is as follows: ,in For overall priority, The original priority of the task itself. and These are the weighting coefficients.

[0035] In this embodiment, when a task has a deadline constraint, the system adjusts the task scheduling evaluation index based on the remaining time of the task. The adjustment formula is as follows: ,in This is the revised overall task evaluation index. Remaining time for the task For the current moment, The maximum allowed execution time for the task. This represents the urgency coefficient.

[0036] Specifically, the urgency coefficient is used to adjust the amplification of the remaining time percentage on the task scheduling evaluation index. An initial value is preset based on the system's sensitivity to real-time performance. Then, by simulating task loads with different deadline constraints, the optimal urgency coefficient value is determined using methods such as grid search or Bayesian optimization, with the on-time completion rate of tasks and system throughput as optimization objectives.

[0037] Example 2 This embodiment provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-described industrial internet workflow task scheduling method based on dependency awareness and dynamic priority by calling the computer program stored in the memory.

[0038] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the dependency-aware and dynamic priority-based industrial internet workflow task scheduling method provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0039] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this disclosure can be embodied in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0040] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0041] This invention is described with reference to flowchart illustrations and block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and block diagrams, as well as combinations of blocks in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0042] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0043] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A workflow task scheduling method for the Industrial Internet based on dependency awareness and dynamic priority, characterized in that: The specific steps include the following: S1: Within the preset task time window, synchronously acquire the length of the ready task queue, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total number of tasks for each workflow, and perform preprocessing. S2: Based on the length of the preprocessed ready task queue and the preset stable queue length threshold, the scheduling pressure coefficient is obtained. Based on the number of unfinished tasks and the total number of tasks in each preprocessed workflow, the task dependency tightness is calculated. Based on the scheduling pressure coefficient and the task dependency tightness, the scheduling-dependency pressure index is obtained. Based on the preprocessed chip junction temperature, the current CPU frequency, and the CPU utilization, the thermal-efficiency synergistic decay index is calculated. Based on the scheduling-dependency pressure index and the thermal-efficiency synergistic decay index, the system synergistic risk index is calculated. S3: Compare the system collaborative risk index with the preset threshold, divide the risk areas according to the comparison results, and use a piecewise nonlinear mapping function to calculate the task scheduling evaluation index; S4: Based on the task scheduling evaluation index, make scheduling decisions on the ready task queue according to the preset decision rules, execute tasks and update the system status, and use the updated system status as the input for the next scheduling cycle, and return to step S1 to realize the adaptive optimization of the closed-loop scheduling strategy.

2. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 1, characterized in that: S2 includes the following specific steps: S21: Based on the length of the preprocessed ready task queue With preset stable queue length threshold Calculate the scheduling pressure coefficient The formula is: ； S22: Based on each preprocessed workflow Number of unfinished tasks Total number of tasks Computation task dependency tightness The formula is: ； in, The set of workflows to which the currently ready tasks belong. For set The number of workflows in the middle; S23: Based on scheduling pressure coefficient and task dependency tightness Obtain the scheduling-dependency pressure index The formula is: ； in, This is the first adjustment coefficient.

3. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 2, characterized in that: S2 further includes the following specific steps: S24: Based on the pre-processed chip junction temperature CPU current frequency and CPU utilization Calculate the thermal-efficiency synergistic decay index The formula is: ； in, The coefficient of thermal stress, To calculate the efficiency decay rate, This is the second adjustment coefficient. It is a constant.

4. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 3, characterized in that: S2 further includes the following specific steps: S25: Based on the scheduling-dependency pressure index and thermal-efficiency synergistic decay index Computational system collaborative risk index The formula is: 。 5. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 4, characterized in that: S3 includes the following specific steps: S31: The preset thresholds specifically include low-risk thresholds. Medium risk threshold and high-risk threshold ,and ; S32: System Coordination Risk Index Separate and low risk thresholds Medium risk threshold and high-risk threshold Compare them and divide them into corresponding risk areas. At that time, it was classified as a low-risk area. At that time, it was classified as a medium-risk area. At that time, it was classified as a high-risk area. At that time, it was designated as a critical zone.

6. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 5, characterized in that: S3 further includes the following specific steps: S33: Based on the divided risk areas, the task scheduling evaluation index is calculated using a piecewise nonlinear mapping function. The formula is as follows: ； in, Based on the priority offset, , as well as For mapping coefficients, This is the highest priority limit.

7. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 6, characterized in that: S4 includes the following specific steps: S41: The preset decision rule is to allocate tasks in the ready task queue according to the task scheduling evaluation index. Sort in descending order, select The task with the highest value is designated as the task to be executed in the current scheduling cycle. S42: Dispatch the selected task to an idle computing core for execution, and update the workflow to which the task belongs after the task is completed. Number of unfinished tasks And add the satisfied subsequent tasks to the ready task queue; S43: Use the updated ready task queue length, chip junction temperature, current CPU frequency, CPU utilization, and the number of unfinished tasks and total number of tasks for each workflow as input for the next scheduling cycle, and return to step S1.

8. The industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in claim 7, characterized in that: The adaptive adjustment step in S4 includes: based on the system collaborative risk index. The first adjustment coefficient for the short-term rate of change is dynamically adjusted. Second adjustment coefficient The value is obtained by performing linear regression to calculate the slope; when the short-term rate of change exceeds a preset rate of change threshold... At that time, and Simultaneously multiplied by a preset increment coefficient ,in , The scheduling period is [number].

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements an industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in any one of claims 1-8.

10. An electronic device, characterized in that, It includes a memory for storing instructions; and a processor for executing the instructions, causing the device to perform an industrial internet workflow task scheduling method based on dependency awareness and dynamic priority as described in any one of claims 1 to 8.

Images