Self-adaptive software and hardware collaborative scheduling system and method based on distributed integration

Abstract

The invention relates to the technical field of computers, and discloses a distributed integration-based adaptive software and hardware collaborative scheduling system and method.Working data is acquired through a sensing acquisition module, and system states are distinguished and task requirements are predicted through a sensing processing strategy; the resource management module dynamically schedules task resources of software and hardware according to the system state and the task requirement and generates a scheduling result of the task resources, and through refined resource management, the scheduling system can dynamically allocate and schedule the resources according to the actual task requirement and the system state, the resource utilization rate is increased, and waste is reduced; the task scheduling module determines a task scheduling strategy according to the scheduling result of the task resources; the control execution module converts the task scheduling strategy into a control signal and drives the automation equipment to execute the task; and the feedback adjustment module obtains a task execution result and a system state of the automation equipment, generates a feedback signal through a signal generation strategy, and dynamically adjusts a task scheduling strategy.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a distributed integrated adaptive software and hardware collaborative scheduling system and method. Background Art

[0002] With the development of science and technology, some researchers have migrated the adaptive scheduling system to hardware acceleration through algorithms, reducing the occupation of CPU resources and the consumption of computing resources, and successfully coordinated the scheduling of external systems and automation equipment. For distributed environments such as data centers and cloud computing platforms, effective resource scheduling management is crucial to improving the performance and efficiency of the system. For external systems and automation equipment, intelligent scheduling is used to optimize resource usage, ensure service quality, and reduce energy consumption. Through the design of software and hardware collaborative scheduling systems, high-precision positioning and low power consumption are achieved, and system responsiveness and data throughput are improved. However, most of them have not solved how to obtain work data to distinguish system status and predict task requirements to dynamically schedule software and hardware task resources, determine task scheduling strategies to drive automation equipment to perform tasks, and dynamically adjust task scheduling strategies based on feedback signals.

[0003] For example, a Chinese patent with publication number CN117271087A discloses a software-hardware collaborative scheduling system. It includes: a general processor, a task dependency synchronization device, a hardware scheduling device, a packaged command scheduling execution device, and a data processing device; the general processor is used to send each dependent subtask with a dependency relationship to the task dependency synchronization device through the built-in scheduling software; the task dependency synchronization device is used to transmit the to-be-executed subtask that is in a dependency release state and meets the hardware scheduling conditions to the hardware scheduling device; the hardware scheduling device is used to send the task start command matching each to-be-executed subtask to the packaged command scheduling execution device; the packaged command scheduling execution device is used to generate a start operation sequence matching the task start command of the to-be-executed task, and send the start subcommand matching each start operation in the start operation sequence to the data processing device in sequence. The above technical solution can effectively improve the performance of task scheduling.

[0004] For example, a Chinese patent with the authorization announcement number CN117389712B discloses a GPU multi-thread scheduling management system, which belongs to the field of computing technology and includes a thread scheduling module, a thread configuration module, a software configuration module and a breakpoint storage module; the thread scheduling module is configured in the GPU chip, and the thread scheduling module includes a first thread scheduling module and a second thread scheduling module; the thread configuration module is connected to the first thread scheduling module, and the software configuration module and the breakpoint storage module are respectively connected to the second thread scheduling module. Through the scheduling mechanism of the GPU chip's hardware and software collaboration, an efficient multi-thread scheduling implementation scheme is proposed. This scheme only requires the software to initialize and configure each thread, and the GPU chip automatically implements scheduling and thread switching, which greatly reduces the time of multi-thread scheduling and improves the execution efficiency of multi-thread scheduling. It can be widely used in system multi-thread management and scheduling.

[0005] The above patents have the problems raised by the background technology: the above-mentioned software and hardware collaborative scheduling system sends each dependent subtask with a dependency relationship to the task dependency synchronization device through a general processor, and transmits the to-be-executed subtasks that are in the dependency release state and meet the hardware scheduling conditions to the hardware scheduling device through the task dependency synchronization device, and sends the task start command matching each to-be-executed subtask to the package command scheduling execution transposition through the hardware scheduling device, and generates a start operation sequence matching the task start command of the task to be executed through the package command scheduling execution transposition, and sends the start subcommand matching each start operation in the start operation sequence to the data processing device in sequence; the above-mentioned GPU multi-thread scheduling management system is configured in the GPU chip through a thread scheduling module, and the thread scheduling module includes a first thread scheduling module and a second thread scheduling module, and the thread configuration module is connected to the first thread scheduling module, and the software configuration module and the breakpoint storage module are respectively connected to the second thread scheduling module, so that the GPU chip automatically realizes scheduling and thread switching. The above two patents do not solve the problem of how to distinguish the system status and predict the task requirements by obtaining working data to dynamically schedule the task resources of software and hardware, and determine the task scheduling strategy to drive the automation equipment to perform tasks, and dynamically adjust the task scheduling strategy according to the feedback signal. To solve this problem, the present invention proposes an adaptive software and hardware collaborative scheduling system and method based on distributed integration. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above problems existing in the existing distributed integration-based adaptive software and hardware collaborative scheduling system and method, the present invention is proposed.

[0008] Therefore, the object of the present invention is to provide a distributed integrated adaptive software and hardware collaborative scheduling system and method.

[0009] In order to solve the above technical problems, the present invention provides an adaptive software and hardware collaborative scheduling system based on distributed integration: a sensing acquisition module, a resource management module, a task scheduling module, a control execution module and a feedback adjustment module; The sensor acquisition module is used to acquire working data, distinguish system status and predict task requirements through sensor processing strategies; The resource management module is used to dynamically schedule task resources of software and hardware according to system status and task requirements, and generate scheduling results of task resources; The task scheduling module is used to determine the task scheduling strategy according to the scheduling result of the task resources; The control execution module is used to convert the task scheduling strategy into a control signal to drive the automation equipment to execute the task; The feedback adjustment module is used to obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust the task scheduling strategies.

[0010] As a preferred solution of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention, wherein: the working data is obtained, the system state is distinguished and the task demand is predicted by the sensor processing strategy, the sensor processing strategy includes a state classification strategy and a demand prediction strategy, the state classification strategy is used to distinguish the system state according to the characteristics of the working data, and the demand prediction strategy is used to predict the task demand according to the characteristics of the working data; The state classification strategy includes: Determine the status of the task execution result. If the task is completed, determine the CPU usage status and memory occupancy status, configure the basic usage threshold and the extreme usage threshold, configure the basic occupancy threshold and the extreme occupancy threshold. If the CPU usage is less than or equal to the basic usage threshold and the memory occupancy is less than or equal to the basic occupancy threshold, it means that the system is in an idle state. If the CPU usage is greater than the basic usage threshold or less than the extreme usage threshold and the memory usage is less than the extreme usage threshold, the system status is normal; If the task execution is not completed, the device status is judged. If the device is abnormal, it means that the system status is abnormal; If the device is normal, the CPU usage status and memory usage status are determined. If the CPU usage is greater than or equal to the usage extreme threshold and the memory usage is greater than or equal to the usage extreme threshold, it indicates that the system is in an overloaded state.

[0011] As a preferred solution of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention, the demand forecasting strategy includes: Predict task requirements based on the characteristics of work data, and comprehensively judge the CPU usage status and memory occupancy status. If the CPU usage rate is greater than or equal to the usage extreme threshold and the memory usage is less than or equal to the occupancy basic threshold, the task requirement is judged to be data processing type; If the CPU usage is greater than the basic usage threshold and less than the extreme usage threshold, and the memory usage is greater than the basic usage threshold and less than the extreme usage threshold, the task demand is judged to be balanced; If the CPU usage is less than or equal to the basic usage threshold and the memory usage is greater than or equal to the extreme usage threshold, the task requirement is determined to be data storage type.

[0012] As a preferred solution of the distributed integrated adaptive software and hardware collaborative scheduling system described in the present invention, wherein: dynamically scheduling the task resources of the software and hardware according to the system state and the task requirements, generating the scheduling result of the task resources, configuring the scheduling base value, scheduling median value and scheduling extreme value, if the system state is in an idle state, then judging the task requirements, if the task requirements are data processing type and data storage type, then the scheduling result of the task resources is the scheduling extreme value; If the task demand is balanced, the scheduling result of the task resources is the scheduling median; If the system status is normal, the task requirement is determined. If the task requirement is data processing and data storage, the scheduling result of the task resources is the scheduling median; If the task demand is balanced, the scheduling result of the task resources is the scheduling extreme value; If the system status is in an abnormal state, the scheduling result of the task resources is the scheduling base value; If the system is in an overload state, the task demand is determined. If the task demand is data processing type, the scheduling result of the task resources is the scheduling median; If the task requirements are balanced and data storage type, the scheduling result of the task resources is the scheduling base value.

[0013] As a preferred solution of the distributed integrated adaptive hardware and software collaborative scheduling system of the present invention, the task scheduling strategy is determined according to the scheduling result of the task resources, and the task scheduling strategy includes: Divide the system's task resources into resource nodes and reserve spare resource nodes. If the scheduling result of the task resources is the scheduling base value, suspend the task and release the resource nodes. If the scheduling result of the task resource is the scheduling median, the task is migrated to the backup resource node; If the scheduling result of the task resource is a scheduling extreme value, the task will be assigned to the resource node for execution.

[0014] As a preferred solution of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention, the strategy of releasing resource nodes includes: The set of resource nodes in the system is , , Indicates resource nodes, of which Indicates the total number of resource nodes and the set of spare resource nodes , determine the resource nodes to be released, and select a node subset , , , as the resource node to be released, Indicates resource nodes to be released, notify the tasks running on each resource node to be released to suspend task execution, recycle the task resources allocated to the resource node to be released, and update the set of resource nodes to ,in Represents the set of available resource nodes after releasing the resource nodes.

[0015] As a preferred solution of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention, wherein: the task scheduling strategy is converted into a control signal, and if the task scheduling strategy indicates that the task is assigned to a resource node for execution, the control signal is a start signal; If the task scheduling strategy indicates that the task is to be migrated to a backup resource node, the control signal is a migration signal; If the task scheduling strategy indicates pausing the task and releasing the resource node, the control signal is a stop signal.

[0016] An adaptive software-hardware collaborative scheduling method based on distributed integration includes: S1, obtaining work data, distinguishing system states and predicting task requirements through sensor processing strategies; S2. Dynamically schedule the task resources of software and hardware according to the system status and task requirements, and generate the scheduling results of task resources; S3. Determine the task scheduling strategy according to the scheduling result of the task resources; S4, converting the task scheduling strategy into a control signal to drive the automation equipment to perform the task; S5. Obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust task scheduling strategies.

[0017] A computer device comprises a memory for storing instructions and a processor for executing the instructions, so that the computer device implements a distributed integrated adaptive software and hardware collaborative scheduling method.

[0018] A computer-readable storage medium stores a computer program, which, when executed, implements a distributed integrated adaptive software and hardware collaborative scheduling method.

[0019] Beneficial effects of the present invention: The present invention obtains working data through the sensor acquisition module, distinguishes the system status and predicts the task requirements through the sensor processing strategy, not only provides real-time data, but also analyzes historical trends and predicts future task resource requirements, so as to prepare in advance and avoid resource bottlenecks and system overloads; the resource management module dynamically schedules the task resources of software and hardware according to the system status and task requirements, and generates the scheduling results of task resources. Through refined resource management, the scheduling system can dynamically allocate and schedule resources according to actual task requirements and system status, improve resource utilization and reduce waste; the task scheduling module determines the task scheduling strategy according to the scheduling results of task resources, and optimizes the task execution order , shorten execution time, improve throughput, promote the rational matching and use of software and hardware resources, and reduce the overall energy consumption of system operation; the control execution module converts the task scheduling strategy into a control signal to drive the automation equipment to perform tasks. It can respond quickly when hardware failures or software anomalies occur, and reallocate tasks and resources to ensure the robustness and reliability of the scheduling system; the feedback adjustment module obtains the task execution results and system status of the automation equipment, generates feedback signals through signal generation strategies, and dynamically adjusts the task scheduling strategy to adapt to the ever-changing working conditions and environment. The collaborative work of the entire scheduling system can provide more stable and efficient services, and ultimately improve user satisfaction and experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 The system structure diagram of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention; Figure 2It is a flow chart of the resource node release strategy of the distributed integrated adaptive software and hardware collaborative scheduling system of the present invention; Figure 3 The signal generation strategy flow chart of the adaptive software and hardware collaborative scheduling system based on distributed integration of the present invention; Figure 4 The present invention is a method flow chart of the distributed integrated adaptive software and hardware collaborative scheduling method. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0024] Example 1 In this embodiment, a system structure diagram of a distributed integrated adaptive software and hardware collaborative scheduling system is provided, such as Figure 1 As shown, the distributed integrated adaptive software and hardware collaborative scheduling system includes a sensor acquisition module, a resource management module, a task scheduling module, a control execution module and a feedback adjustment module.

[0025] The sensor acquisition module is used to obtain working data, distinguish system status and predict task requirements through sensor processing strategies.

[0026] Acquire working data, distinguish system status and predict task requirements through sensor processing strategy. Sensor processing strategy includes status classification strategy and demand prediction strategy. Status classification strategy is used to distinguish system status according to the characteristics of working data, and demand prediction strategy is used to predict task requirements according to the characteristics of working data.

[0027] The working data includes system performance data, task execution data and environmental data. The features of the working data are extracted to obtain CPU usage, memory usage, task execution results and device status.

[0028] Status classification strategies include: Determine the status of the task execution result. If the task is completed, determine the CPU usage status and memory occupancy status, configure the basic usage threshold and the extreme usage threshold, configure the basic occupancy threshold and the extreme occupancy threshold. If the CPU usage is less than or equal to the basic usage threshold and the memory occupancy is less than or equal to the basic occupancy threshold, it means that the system is in an idle state. If the CPU usage is greater than the basic usage threshold or less than the extreme usage threshold and the memory usage is less than the extreme usage threshold, the system status is normal; If the task execution is not completed, the device status is judged. If the device is abnormal, it means that the system status is abnormal; If the device is normal, the CPU usage status and memory usage status are determined. If the CPU usage is greater than or equal to the usage extreme threshold and the memory usage is greater than or equal to the usage extreme threshold, it indicates that the system is in an overloaded state.

[0029] The above system performance data include CPU usage, memory usage, disk I / O rate and network bandwidth usage, etc. These data can reflect the system's operating status and resource usage. The above task execution data include the execution time, required resources and execution results of each task, etc. The above environmental data include temperature, humidity and pressure data, etc. These data are of great significance for predicting the status and potential failures of automation equipment. The important features of the above system performance data, task execution data and environmental data are extracted for distinguishing the system status and predicting task requirements, respectively. The important features here are CPU usage, memory usage, task execution results and equipment status.

[0030] First, determine whether the task is completed. When the task is completed, you need to determine the CPU usage status and memory occupancy status. The basic usage threshold and extreme usage threshold are the thresholds of CPU usage, and the basic occupancy threshold and extreme occupancy threshold are the thresholds of memory occupancy. When the task is not completed, you need to determine the device status. There are two major types of device abnormalities. If the sensor detects hardware fault indications, including hardware overtemperature and unstable power supply, the device will be immediately judged as abnormal. If the system log records information about abnormal software exit, program crash or execution failure, the device abnormality can also be determined. According to the above status classification strategy, the system status is divided into normal status, idle status, overload status and abnormal status.

[0031] Demand forecasting strategies include: Predict task requirements based on the characteristics of work data, and comprehensively judge the CPU usage status and memory occupancy status. If the CPU usage rate is greater than or equal to the usage extreme threshold and the memory usage is less than or equal to the occupancy basic threshold, the task requirement is judged to be data processing type; If the CPU usage is greater than the basic usage threshold and less than the extreme usage threshold, and the memory usage is greater than the basic usage threshold and less than the extreme usage threshold, the task demand is judged to be balanced; If the CPU usage is less than or equal to the basic usage threshold and the memory usage is greater than or equal to the extreme usage threshold, the task requirement is determined to be data storage type.

[0032] Compare the resource consumption of completed and incomplete tasks in the historical work data to find out whether the incomplete task is caused by insufficient resources, and avoid it in future resource allocation. At the same time, monitor the records of device status during task execution in the historical work data to see if there are any reports of hardware failure or performance degradation. If a specific device problem is found to be related to task failure, it is necessary to check and ensure that the device is in good condition before scheduling a new task. The above-mentioned data processing tasks mostly have high requirements for CPU resources but relatively low and stable memory requirements. The data storage tasks mostly have low requirements for CPU resources but relatively high memory requirements. The balanced tasks mostly have relatively balanced requirements for CPU resources and memory, and there are no obvious resource bottlenecks. Therefore, any form of resource surplus or shortage should be avoided in subsequent task resource scheduling.

[0033] Demand forecasting strategies also include: Analyze historical work data and predict task requirements through the prediction model. The function expression of the prediction model is as follows: ; In the formula, express The task requirements at all times, represents a constant term, represents the coefficient of the autoregressive part, express The observation results of the task requirements at each moment, represents the coefficient of the moving average part, Respectively The error term at time, express The error term at time.

[0034] By analyzing historical work data and using time series forecasting methods to predict task requirements, it is possible to capture the linear relationship of work data without the need for other variables, thereby obtaining prediction results for task requirements.

[0035] The resource management module is used to dynamically schedule the task resources of software and hardware according to the system status and task requirements, and generate the scheduling results of task resources.

[0036] Dynamically schedule the task resources of software and hardware according to the system status and task requirements, generate the scheduling results of task resources, configure the scheduling base value, scheduling median value and scheduling extreme value. If the system status is in an idle state, judge the task requirements. If the task requirements are data processing and data storage, the scheduling result of task resources is the scheduling extreme value. If the task demand is balanced, the scheduling result of the task resources is the scheduling median; If the system status is normal, the task requirement is determined. If the task requirement is data processing and data storage, the scheduling result of the task resources is the scheduling median; If the task demand is balanced, the scheduling result of the task resources is the scheduling extreme value; If the system status is in an abnormal state, the scheduling result of the task resources is the scheduling base value; If the system is in an overload state, the task demand is determined. If the task demand is data processing type, the scheduling result of the task resources is the scheduling median; If the task requirements are balanced and data storage type, the scheduling result of the task resources is the scheduling base value.

[0037] In specific applications, in the idle state, the system resources are sufficient, and data processing and data storage tasks are given priority. For balanced tasks, they are scheduled according to the current specific task requirements and task priorities. In the normal state, system resources are reasonably utilized, but resource competition needs to be avoided, so it is necessary to focus on scheduling resources to complete balanced tasks. In the abnormal state, equipment abnormalities may cause some tasks to be unable to execute, and relevant measures need to be taken immediately. In the overloaded state, system resources reach the limit and need to be scheduled carefully to avoid system crashes. Therefore, for high-priority tasks, it is necessary to ensure that they have sufficient resources, and based on real-time monitoring of the system status and task requirements, the task resources of software and hardware can be scheduled in time. At the same time, in the event of abnormal conditions, it is necessary to respond in time to restore the abnormality of the equipment.

[0038] The task scheduling module is used to determine the task scheduling strategy according to the scheduling results of task resources.

[0039] The task scheduling strategy is determined based on the scheduling results of the task resources. The task scheduling strategy includes: Divide the system's task resources into resource nodes and reserve spare resource nodes. If the scheduling result of the task resources is the scheduling base value, suspend the task and release the resource nodes. If the scheduling result of the task resource is the scheduling median, the task is migrated to the backup resource node; If the scheduling result of the task resource is a scheduling extreme value, the task will be assigned to the resource node for execution.

[0040] The strategy for releasing resource nodes is as follows: Figure 2 As shown, specifically including: The set of resource nodes in the system is , , Indicates resource nodes, of which Indicates the total number of resource nodes and the set of spare resource nodes , determine the resource nodes to be released, and select a node subset , , , as the resource node to be released, Indicates resource nodes to be released, notify the tasks running on each resource node to be released to suspend task execution, recycle the task resources allocated to the resource node to be released, and update the set of resource nodes to ,in Represents the set of available resource nodes after releasing the resource nodes.

[0041] In specific applications, the system's task resources are divided into countless resource nodes, and spare resource nodes with sufficient resources are reserved or sufficient task resources are reserved through the interconnection of multiple virtual machines. When the scheduling result of the task resources is the scheduling base value, it is possible that some tasks cannot be executed due to equipment abnormalities. At this time, it is necessary to immediately isolate the abnormal resources or equipment to prevent the spread of the abnormality. It is also possible that in the overload state, for the two non-critical tasks of balanced and data storage, it is necessary to suspend the tasks and release the resource nodes to reduce the burden on the system; when the scheduling result of the task resources is the scheduling median, it may be in the idle state. For balanced tasks, it is necessary to wait for processing according to the utilization of resource nodes. It may be in the normal state. For data processing and data storage, these two types of tasks need to be processed. For these two types of storage tasks, it is necessary to comprehensively consider whether the resource nodes meet the resource requirements of the two types of tasks to avoid overloading of a single resource node. At the same time, spare resource nodes are used to process tasks with higher priority and higher resource requirements. It is also possible that in an overloaded state, for key tasks such as data processing, it is necessary to ensure that they obtain sufficient resources. It may be necessary to jointly execute such key tasks through some spare resource nodes and spare resource nodes; when the scheduling result of task resources is a scheduling extreme value, it is possible that in an idle state, tasks with larger task resource requirements are executed first, including data processing and data storage tasks. It is also possible that in a normal state, for tasks such as balanced tasks, system resources can be better utilized reasonably to avoid competition for resource nodes by other tasks.

[0042] The control execution module is used to convert the task scheduling strategy into a control signal to drive the automation equipment to execute the task.

[0043] The task scheduling strategy is converted into a control signal. If the task scheduling strategy indicates that the task is assigned to a resource node for execution, the control signal is a start signal. If the task scheduling policy indicates that the task is to be migrated to a backup resource node, the control signal is a migration signal; If the task scheduling policy indicates pausing the task and releasing the resource node, the control signal is a stop signal.

[0044] According to the corresponding task scheduling strategy, it is converted into a corresponding control signal. When the generated control signal is a start signal, it is necessary to judge the system state to drive the automation equipment to perform tasks. When the system state is in an idle state, it is necessary to give priority to driving the automation equipment to perform data processing and data storage tasks, and suspend balancing tasks. When the system state is in a normal state, it is necessary to give priority to driving the automation equipment to perform balancing tasks, and suspend data processing and data storage tasks. When the generated control signal is a migration signal, it is necessary to judge the system state to drive the automation equipment to perform tasks. When the system state is in an idle state, the balancing tasks are migrated to the backup resource node for processing. When the system state is in a normal state, the data processing and data storage tasks are migrated to the backup resource node for processing. When the system state is in an overloaded state, the data processing tasks are migrated to the backup resource node for processing. When the generated control signal is a stop signal, it is necessary to judge the system state to drive the automation equipment to perform tasks. When the system state is in an overloaded state, stop the two non-critical tasks of balancing and data storage. When the system state is in an abnormal state, stop the abnormal equipment and isolate the abnormal resource node.

[0045] The feedback adjustment module is used to obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust the task scheduling strategies.

[0046] Obtain the task execution results and system status of the automation equipment, and generate feedback signals through signal generation strategies. The signal generation strategies are as follows: Figure 3 As shown, specifically including: If the automation equipment completes its task, there will be no feedback signal; If the task execution of the automation equipment is not completed, the system status is judged. If the system status is in an overload state, the feedback signal is to enable the backup resource node; If the system status is in an abnormal state, the feedback signal is to replace the automation equipment.

[0047] The task scheduling strategy is dynamically adjusted according to the corresponding feedback signal. When there is no feedback signal, the system executes normally according to the task scheduling strategy. When the feedback signal is to enable the backup resource node, the number of backup resource nodes is expanded according to the actual situation. When the feedback signal is to replace the automation equipment, choose whether to suspend the task according to the task type and release the resource nodes of non-critical tasks.

[0048] Example 2 In this embodiment, a method flow chart of an adaptive software and hardware collaborative scheduling method based on distributed integration is provided, such as Figure 4 As shown, the adaptive software and hardware collaborative scheduling method based on distributed integration includes: S1. Obtain working data, distinguish system status and predict task requirements through sensor processing strategies.

[0049] Acquire working data, distinguish system status and predict task requirements through sensor processing strategy. Sensor processing strategy includes status classification strategy and demand prediction strategy. Status classification strategy is used to distinguish system status according to the characteristics of working data, and demand prediction strategy is used to predict task requirements according to the characteristics of working data.

[0050] Status classification strategies include: Determine the status of the task execution result. If the task is completed, determine the CPU usage status and memory occupancy status, configure the basic usage threshold and the extreme usage threshold, configure the basic occupancy threshold and the extreme occupancy threshold. If the CPU usage is less than or equal to the basic usage threshold and the memory occupancy is less than or equal to the basic occupancy threshold, it means that the system is in an idle state. If the CPU usage is greater than the basic usage threshold or less than the extreme usage threshold and the memory usage is less than the extreme usage threshold, the system status is normal; If the task execution is not completed, the device status is judged. If the device is abnormal, it means that the system status is abnormal; If the device is normal, the CPU usage status and memory usage status are determined. If the CPU usage is greater than or equal to the usage extreme threshold and the memory usage is greater than or equal to the usage extreme threshold, it indicates that the system is in an overloaded state.

[0051] S2. Dynamically schedule the task resources of software and hardware according to the system status and task requirements, and generate the scheduling results of task resources.

[0052] Dynamically schedule the task resources of software and hardware according to the system status and task requirements, generate the scheduling results of task resources, configure the scheduling base value, scheduling median value and scheduling extreme value. If the system status is in an idle state, judge the task requirements. If the task requirements are data processing and data storage, the scheduling result of task resources is the scheduling extreme value. If the task demand is balanced, the scheduling result of the task resources is the scheduling median; If the system status is normal, the task requirement is determined. If the task requirement is data processing and data storage, the scheduling result of the task resources is the scheduling median; If the task demand is balanced, the scheduling result of the task resources is the scheduling extreme value; If the system status is in an abnormal state, the scheduling result of the task resources is the scheduling base value; If the system is in an overload state, the task demand is determined. If the task demand is data processing type, the scheduling result of the task resources is the scheduling median; If the task requirements are balanced and data storage type, the scheduling result of the task resources is the scheduling base value.

[0053] S3. Determine the task scheduling strategy according to the scheduling results of task resources.

[0054] The task scheduling strategy is determined based on the scheduling results of the task resources. The task scheduling strategy includes: Divide the system's task resources into resource nodes and reserve spare resource nodes. If the scheduling result of the task resources is the scheduling base value, suspend the task and release the resource nodes. If the scheduling result of the task resource is the scheduling median, the task is migrated to the backup resource node; If the scheduling result of the task resource is a scheduling extreme value, the task will be assigned to the resource node for execution.

[0055] S4. Convert the task scheduling strategy into a control signal to drive the automation equipment to perform the task.

[0056] The task scheduling strategy is converted into a control signal. If the task scheduling strategy indicates that the task is assigned to a resource node for execution, the control signal is a start signal. If the task scheduling policy indicates that the task is to be migrated to a backup resource node, the control signal is a migration signal; If the task scheduling policy indicates pausing the task and releasing the resource node, the control signal is a stop signal.

[0057] S5. Obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust task scheduling strategies.

[0058] Obtain the task execution results and system status of the automation equipment, and generate feedback signals through signal generation strategies. The signal generation strategies include: If the automation equipment completes its task, there will be no feedback signal; If the task execution of the automation equipment is not completed, the system status is judged. If the system status is in an overload state, the feedback signal is to enable the backup resource node; If the system status is in an abnormal state, the feedback signal is to replace the automation equipment.

[0059] Example 3 In this embodiment, a computer device is provided, including a memory and a processor, the memory is used to store instructions, and the processor is used to execute the instructions, so that the computer device executes the steps of implementing the above-mentioned distributed integration-based adaptive software and hardware collaborative scheduling method.

[0060] Example 4 In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the steps of the above-mentioned distributed integration-based adaptive software and hardware collaborative scheduling method are implemented.

[0061] The computer-readable storage medium includes: a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and other media for storing program codes.

[0062] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. Adaptive software and hardware collaborative scheduling system based on distributed integration, characterized by: include: Sensor acquisition module, resource management module, task scheduling module, control execution module and feedback adjustment module; The sensor acquisition module is used to acquire working data, distinguish system status and predict task requirements through sensor processing strategies; The resource management module is used to dynamically schedule task resources of software and hardware according to system status and task requirements, and generate scheduling results of task resources; The task scheduling module is used to determine the task scheduling strategy according to the scheduling result of the task resources; The control execution module is used to convert the task scheduling strategy into a control signal to drive the automation equipment to execute the task; The feedback adjustment module is used to obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust the task scheduling strategies.

2. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 1, characterized in that: Acquire the working data, distinguish the system state and predict the task demand through the sensor processing strategy, the sensor processing strategy includes a state classification strategy and a demand prediction strategy, the state classification strategy is used to distinguish the system state according to the characteristics of the working data, and the demand prediction strategy is used to predict the task demand according to the characteristics of the working data; The state classification strategy includes: Determine the status of the task execution result. If the task is completed, determine the CPU usage status and memory occupancy status, configure the basic usage threshold and the extreme usage threshold, configure the basic occupancy threshold and the extreme occupancy threshold. If the CPU usage is less than or equal to the basic usage threshold and the memory occupancy is less than or equal to the basic occupancy threshold, it means that the system is in an idle state. If the CPU usage is greater than the basic usage threshold or less than the extreme usage threshold and the memory usage is less than the extreme usage threshold, the system status is normal; If the task execution is not completed, the device status is judged. If the device is abnormal, it means that the system status is abnormal; If the device is normal, the CPU usage status and memory usage status are determined. If the CPU usage is greater than or equal to the usage extreme threshold and the memory usage is greater than or equal to the usage extreme threshold, it indicates that the system is in an overloaded state.

3. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 2, characterized in that: The demand forecasting strategy includes: Predict task requirements based on the characteristics of work data, and comprehensively judge the CPU usage status and memory occupancy status. If the CPU usage rate is greater than or equal to the usage extreme threshold and the memory usage is less than or equal to the occupancy basic threshold, the task requirement is judged to be data processing type; If the CPU usage is greater than the basic usage threshold and less than the extreme usage threshold, and the memory usage is greater than the basic usage threshold and less than the extreme usage threshold, the task demand is judged to be balanced; If the CPU usage is less than or equal to the basic usage threshold and the memory usage is greater than or equal to the extreme usage threshold, the task requirement is determined to be data storage type.

4. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 3, characterized in that: Dynamically schedule the task resources of software and hardware according to the system state and the task requirements, generate the scheduling result of the task resources, configure the scheduling base value, scheduling median value and scheduling extreme value, if the system state is in an idle state, judge the task requirements, if the task requirements are data processing type and data storage type, the scheduling result of the task resources is the scheduling extreme value; If the task demand is balanced, the scheduling result of the task resources is the scheduling median; If the system status is normal, the task requirement is determined. If the task requirement is data processing and data storage, the scheduling result of the task resources is the scheduling median; If the task demand is balanced, the scheduling result of the task resources is the scheduling extreme value; If the system status is in an abnormal state, the scheduling result of the task resources is the scheduling base value; If the system is in an overload state, the task demand is determined. If the task demand is data processing type, the scheduling result of the task resources is the scheduling median; If the task requirements are balanced and data storage type, the scheduling result of the task resources is the scheduling base value.

5. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 4, characterized in that: Determine a task scheduling strategy according to the scheduling result of the task resources, and the task scheduling strategy includes: Divide the system's task resources into resource nodes and reserve spare resource nodes. If the scheduling result of the task resources is the scheduling base value, suspend the task and release the resource nodes. If the scheduling result of the task resource is the scheduling median, the task is migrated to the backup resource node; If the scheduling result of the task resource is a scheduling extreme value, the task will be assigned to the resource node for execution.

6. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 5, characterized in that: The strategy for releasing resource nodes includes: The set of resource nodes in the system is , , Indicates resource nodes, of which Indicates the total number of resource nodes and the set of spare resource nodes , determine the resource nodes to be released, and select a node subset , , , as the resource node to be released, Indicates resource nodes to be released, notify the tasks running on each resource node to be released to suspend task execution, recycle the task resources allocated to the resource node to be released, and update the set of resource nodes to ,in Represents the set of available resource nodes after releasing the resource nodes.

7. The distributed integrated adaptive software and hardware collaborative scheduling system according to claim 6, characterized in that: Converting the task scheduling strategy into a control signal, if the task scheduling strategy indicates that the task is assigned to a resource node for execution, the control signal is a start signal; If the task scheduling strategy indicates that the task is to be migrated to a backup resource node, the control signal is a migration signal; If the task scheduling strategy indicates pausing the task and releasing the resource node, the control signal is a stop signal.

8. The distributed integration-based adaptive software and hardware collaborative scheduling method is implemented based on the distributed integration-based adaptive software and hardware collaborative scheduling system according to any one of claims 1 to 7, characterized in that: include: S1, obtain working data, distinguish system status and predict task requirements through sensor processing strategy; S2. Dynamically schedule the task resources of software and hardware according to the system status and task requirements, and generate the scheduling results of task resources; S3. Determine the task scheduling strategy according to the scheduling result of the task resources; S4, converting the task scheduling strategy into a control signal to drive the automation equipment to perform the task; S5. Obtain the task execution results and system status of the automation equipment, generate feedback signals through signal generation strategies, and dynamically adjust task scheduling strategies.

9. A computer device, characterized in that: include: A memory for storing instructions; The processor is used to execute the instructions so that the computer device implements the distributed integrated adaptive software and hardware collaborative scheduling method as described in claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the distributed integration-based adaptive software and hardware collaborative scheduling method according to claim 8 is implemented.

Images