Task scheduling method, task scheduling device, electronic equipment and readable storage medium

Abstract

The invention discloses a task scheduling method, a task scheduling device, electronic equipment and a readable storage medium. The method comprises the steps of obtaining a priority and a task feature corresponding to a target task; based on the task features, determining task requirements for executing the target task, the task requirements including a predicted execution time length and a predicted resource demand quantity; determining at least one candidate execution node for executing the target task based on the priority; determining a target execution node for executing the target task from at least one candidate execution node according to the task demand and the load level corresponding to each candidate execution node; and scheduling the target execution node to execute the target task.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and in particular, relates to a task scheduling method, a task scheduling device, an electronic device, and a readable storage medium. Background Art

[0002] With the rapid development of information technology, especially the popularization of cloud computing and big data technology, distributed computing environments have become more complex and changeable. These environments usually involve a large number of computing nodes and different types of tasks, requiring efficient scheduling mechanisms to optimize resource allocation, reduce computing costs, and improve the overall efficiency and response speed of the system. However, traditional task scheduling technology ignores the impact of task priority on node scheduling, which leads to increased task execution delay. Summary of the invention

[0003] The embodiments of the present application provide a task scheduling method, a task scheduling device, an electronic device and a readable storage medium, which can solve the problem in the related art that the task execution delay increases only due to unreasonable node scheduling strategy.

[0004] In a first aspect, an embodiment of the present application provides a task scheduling method, the method comprising: obtaining a priority and task characteristics corresponding to a target task; based on the task characteristics, determining the task requirements for executing the target task, wherein the task requirements include an estimated execution time length and an estimated resource requirement; based on the priority, determining at least one candidate execution node for executing the target task; determining a target execution node for executing the target task from at least one of the candidate execution nodes according to the task requirements and the load level corresponding to each of the candidate execution nodes; and scheduling the target execution node to execute the target task.

[0005] In the second aspect, an embodiment of the present application provides a task scheduling device, which includes: an acquisition module, used to acquire the priority and task characteristics corresponding to the target task; a first determination module, used to determine the task requirements for executing the target task based on the task characteristics, wherein the task requirements include an estimated execution time length and an estimated resource requirement; a second determination module, used to determine at least one candidate execution node for executing the target task based on the priority; a third determination module, used to determine a target execution node for executing the target task from at least one of the candidate execution nodes based on the task requirements and the load level corresponding to each of the candidate execution nodes; and a scheduling module, used to schedule the target execution node to execute the target task. In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0006] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0007] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect.

[0008] In a sixth aspect, a computer program product is provided, the computer program product comprising at least one computer program, and when the computer program is loaded and executed by a processor, the method as described in the first aspect is implemented.

[0009] In an embodiment of the present application, the priority and task characteristics corresponding to the target task are obtained, and then based on the task characteristics, the task requirements for executing the target task are determined, wherein the task requirements include an estimated execution time length and an estimated resource requirement. Based on the priority, at least one candidate execution node for executing the target task is determined, and then according to the task requirements and the load level corresponding to each candidate execution node, a target execution node for executing the target task is determined from at least one candidate execution node, and finally the target execution node is scheduled to execute the target task. In this way, by first determining the candidate execution nodes based on the priority and then selecting the target execution node therefrom, the scheduling strategy is more flexible, the efficiency of determining the target execution node can be improved, and unnecessary delays can be avoided, so that the target task can be processed in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a flowchart of a task scheduling method provided by an exemplary embodiment of the present application; Figure 2 is another flowchart of a task scheduling method provided by an exemplary embodiment of the present application; Figure 3 is a structural diagram of a task scheduling device provided by an exemplary embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0012] In conjunction with the accompanying drawings, a task scheduling method, a task scheduling device, an electronic device and a readable storage medium provided in an embodiment of the present application are described in detail below through specific embodiments and their application scenarios.

[0013] Figure 1 A task scheduling method provided by an embodiment of the present application is shown, and the method can be executed by an electronic device, and the electronic device may include: a server and / or a terminal device. In other words, the method can be executed by software or hardware installed in the electronic device, and the method includes the following steps: S110: Obtain the priority and task characteristics corresponding to the target task.

[0014] The priority of the target task is used to indicate the importance and urgency of the task. For example, if the priority of the target task is high, it means that the target task needs to be processed immediately. If the priority of the target task is low, it means that the target task does not need to be processed urgently. The task characteristics of the target task may include but are not limited to the task type, the estimated data volume, the resource usage characteristics, etc. These characteristics will be used to determine the execution time and resource requirements of the target task.

[0015] Optionally, the priority corresponding to the target task may be determined by a priority identifier corresponding to the target task, or the priority corresponding to the target task may be determined based on a task feature of the target task.

[0016] S120: Determine task requirements for executing the target task based on the task characteristics.

[0017] The task requirements include an estimated execution time and an estimated resource requirement.

[0018] It is understandable that after the task characteristics are acquired, the task characteristics may be analyzed to determine the estimated execution time length and estimated resource requirements for executing the target task.

[0019] S130: Based on the priority, determine at least one candidate execution node for executing the target task.

[0020] It is understandable that in distributed computing and task scheduling systems, there are multiple nodes that can execute tasks independently, but not all nodes are suitable for executing the target task. Based on priority, suitable candidate execution nodes can be screened out, and nodes that are not suitable for executing the target task can be excluded. In this way, the selection range of candidate execution nodes for the target task is narrowed, which can improve the efficiency of subsequent selection of target nodes.

[0021] S140: Determine a target execution node for executing the target task from at least one of the candidate execution nodes according to the task requirement and the load level corresponding to each of the candidate execution nodes.

[0022] It is understandable that the task requirements of different tasks are different, and the loads of different execution nodes are also different. Therefore, it is necessary to select a suitable target execution node from at least one of the candidate execution nodes based on the task requirements of the target task and the load level corresponding to each candidate execution node.

[0023] S150: Scheduling the target execution node to execute the target task.

[0024] In an embodiment of the present application, the priority and task characteristics corresponding to the target task are obtained, and then based on the task characteristics, the task requirements for executing the target task are determined, wherein the task requirements include an estimated execution time length and an estimated resource requirement. Based on the priority, at least one candidate execution node for executing the target task is determined, and then according to the task requirements and the load level corresponding to each candidate execution node, a target execution node for executing the target task is determined from at least one candidate execution node, and finally the target execution node is scheduled to execute the target task. In this way, by first determining the candidate execution nodes based on the priority and then selecting the target execution node therefrom, the scheduling strategy is more flexible, the efficiency of determining the target execution node can be improved, and unnecessary delays can be avoided, so that the target task can be processed in a timely manner.

[0025] In one implementation, determining at least one candidate execution node for executing the target task based on the priority includes: when the priority is a first priority, determining that at least one of the candidate execution nodes includes at least one edge execution node; when the priority is a second priority, determining that at least one of the candidate execution nodes includes at least one central execution node, wherein the second priority is lower than the first priority.

[0026] It is understandable that in distributed computing and task scheduling systems, the distance between nodes has an impact on the task execution time. If the task is scheduled to a node with a shorter physical distance, the network delay is lower and the data transmission speed is faster, thereby reducing the task execution time. If the task is scheduled to a node with a longer physical distance, the network delay is higher and the data transmission speed is slower, thereby increasing the task execution time. Therefore, it is necessary to first determine at least one candidate execution node based on the priority of the target task, that is, to determine the node that is closer to the initiator of the target task as the candidate execution node.

[0027] Among them, if the priority is the first priority, it is determined that at least one of the candidate execution nodes includes at least one edge execution node to reduce delays and improve response speed. If the priority is the second priority, it is determined that at least one of the candidate execution nodes includes at least one central execution node to improve resource utilization and optimize load. That is to say, if the priority is the first priority, the node that is closer to the initiator of the target task, that is, the edge execution node, is preferentially selected as the candidate execution node. If a node that is farther away is selected, the communication delay and data transmission time will increase, resulting in an increase in the completion time of the target task. If the second priority is lower than the first priority, it means that the target task is not a high priority task. In order to improve the resource utilization of the central node and optimize the load of all nodes, the central execution node can be selected as the candidate execution node.

[0028] Optionally, when the priority is the first priority, the distance between at least one of the candidate execution nodes and the initiator of the target task is less than a first threshold; when the priority is the second priority, the distance between at least one of the candidate execution nodes and the initiator of the target task is greater than a second threshold, wherein the first threshold and the second threshold may be equal or different.

[0029] In one implementation, the target execution node for executing the target task is determined from at least one of the candidate execution nodes based on the task requirement and the load level corresponding to each of the candidate execution nodes, including: inputting the task requirement and the load level corresponding to each of the candidate execution nodes into a task scheduling model to obtain the target execution node output by the task scheduling model, wherein the task scheduling model is trained based on a hybrid scheduling strategy.

[0030] Among them, the hybrid scheduling strategy can be expressed by the following formula:

[0031] Where S represents the selected target node and N is the set of all candidate execution nodes. i Indicates the current waiting time queue length of candidate execution node i, is the predicted time to execute the task on candidate execution node i, is the predicted resource requirement of the task to be executed on candidate execution node i. α, β, γ are weight parameters that determine the relative importance of waiting time, predicted execution time, and resource requirement in the scheduling decision.

[0032] In addition, since the task scheduling model takes into account multiple performance indicators such as the waiting time queue length of the node, the predicted execution time and resource requirements, in order to optimize multiple performance indicators at the same time, a multi-objective optimization method, such as Pareto optimization, can be used to balance different scheduling objectives:

[0033] in: Represents the scheduling decision vector of all nodes; It is the objective function that needs to be optimized, such as minimizing maximum delay, balancing load, maximizing resource utilization, etc.

[0034] Optionally, the task scheduling model may be constructed based on a Large Language Model (LLM), so that multiple scheduling targets may be considered simultaneously in the process of determining the target execution node.

[0035] Optionally, the hybrid scheduling strategy may include a shortest job first strategy, a predictive scheduling strategy, a flexible scheduling strategy, etc.

[0036] In this embodiment, the processing efficiency of the task scheduling model is effectively improved through multi-objective optimization, and a better scheduling decision is achieved through a hybrid scheduling strategy.

[0037] In another implementation, determining a target execution node for executing the target task from multiple candidate execution nodes based on the task requirement and the load level of each candidate execution node includes: obtaining a current network load; and determining a target execution node for executing the target task from at least one of the candidate execution nodes based on the current network load, the task requirement and the load level corresponding to each of the candidate execution nodes.

[0038] In one implementation, determining the task requirements for executing the target task based on the task characteristics includes: inputting the task characteristics into a first prediction model and a second prediction model respectively, obtaining an estimated execution time length output by the first prediction model and an estimated resource requirement output by the second prediction model, wherein the first prediction model and the second prediction model are constructed based on a regression model.

[0039] The first prediction model is used to predict the estimated execution time length according to the characteristics of the task. The algorithm corresponding to the first prediction model can be expressed by the following formula:

[0040] in, is the estimated execution time length of the prediction, is the intercept term, It is a feature The coefficient of is the j-th feature of the task, and n is the total number of features.

[0041] The second prediction model is used to predict the expected resource requirements, such as CPU, memory, or disk space, based on the characteristics of the task. The algorithm corresponding to the second prediction model can be expressed by the following formula:

[0042] in, It indicates the estimated resource demand. is the intercept term, Is related to task characteristics The coefficient of correlation.

[0043] Optionally, the algorithm corresponding to the first prediction model can be expressed by the following formula:

[0044] in, represents the predicted task execution time or resource requirements, is a vector sequence containing all task features, represents a deep neural network, such as a convolutional neural network (CNN) or a recurrent neural network (RNN) or a long short-term memory network (LSTM), Represents network parameters.

[0045] Optionally, the algorithm corresponding to the second prediction model can be expressed by the following formula:

[0046] in, represents the predicted task execution time or resource requirements, is a vector sequence containing all task features, represents a deep neural network, such as a convolutional neural network (CNN) or a recurrent neural network (RNN) or a long short-term memory network (LSTM), Represents network parameters.

[0047] In the above implementation, predicting the execution time and resource requirements of the task by using the first prediction model and the second prediction model constructed based on the regression model helps to optimize the decision-making process, thereby improving decision-making efficiency and accuracy.

[0048] In one implementation, the method further includes: dynamically adjusting the load level of each of the candidate execution nodes based on a load balancing strategy.

[0049] Among them, the load balancing strategy is expressed by the feedback control algorithm based on control theory:

[0050] in: Represents a candidate execution node resource error, which is the difference between actual resource usage and forecasted demand, , , are the gains of the proportional, integral, and derivative controllers.

[0051] Optionally, the load balancing strategy can be expressed by the following formula:

[0052] in, Represents the load level of node i. This load balancing strategy is used to optimize the load difference between the heaviest loaded node and the lightest loaded node, which helps to avoid the situation where some nodes are overloaded while other nodes are idle, and improves the efficiency and stability of the entire task scheduling.

[0053] In this implementation, dynamically adjusting the load can ensure that when selecting a target execution node, the resource usage of the node is more even, thereby avoiding resource waste.

[0054] In one implementation, after scheduling the target execution node to execute the target task, the method further includes: obtaining feedback analysis results by monitoring the execution process; and dynamically adjusting the resource configuration of the target execution node and / or optimizing the task scheduling model based on the feedback analysis results.

[0055] That is, when the task is executed, the task execution status and node resource status are monitored, wherein the task execution status may include the start time of the task, the end time, whether errors occurred during the execution process, the efficiency of the task, etc. The node resource status may include CPU usage, memory usage, disk usage, network status, health status at runtime, etc. Then, the feedback analysis results are determined based on the task execution status and node resource status, and then analyzed according to the feedback analysis results, which are used as feedback to optimize and improve the task scheduling model and the target execution node resource configuration.

[0056] Optionally, in the embodiment of the present application, resource allocation may also be dynamically adjusted according to the real-time monitored system load and resource usage, wherein the resource adjustment strategy may be expressed as:

[0057] in, is the adjusted resource amount of node i, is the current resource amount of node i. It is the amount of resource adjustment based on the current load and forecast information, which may be positive (increasing resources) or negative (reducing resources). This dynamic adjustment mechanism allows resources to be scaled according to actual needs, optimize resource utilization, prevent resource waste, and adapt to load changes.

[0058] Optionally, in an embodiment of the present application, when there are multiple target tasks, the optimal path algorithm of the graph may be used to determine the order of scheduling and resource allocation of each task:

[0059] in: It depends on the node The task set, It's a task To the task The dependency weight of is a parameter that adjusts the dependency weights.

[0060] Based on the above embodiment, the present application also provides another flowchart of a task scheduling method, such as Figure 2 As shown, the following steps may be included: S210: Data collection and analysis.

[0061] System administrators or automated tools regularly collect historical task data, including task type, size, execution time, resource consumption, etc. Use large language models to deeply analyze the collected data and automatically extract features.

[0062] S220: Model training and verification.

[0063] Using the collected data, the prediction model is trained through LLM. The model is retrained and validated regularly with new data to maintain the accuracy and adaptability of the model.

[0064] S230: Extracting task features.

[0065] When a new task arrives, the system automatically extracts the characteristics of the task, such as task type, estimated data volume, etc. These characteristics will be used to predict execution time and resource requirements.

[0066] S240: Execution time and resource requirement prediction.

[0067] The system uses the trained model to predict the execution time and resource requirements of new tasks.

[0068] S250: Scheduling decision.

[0069] The system will use the prediction results and the current node status to determine the most suitable node for executing the task through a scheduling algorithm.

[0070] S260: Resource adjustment.

[0071] If the resources of the current node are insufficient to meet the task requirements, the system will dynamically allocate resources based on the resource adjustment strategy.

[0072] S270: Task execution.

[0073] The task starts executing on the selected node. The system monitors the task execution and the resource usage status of the node.

[0074] S280: Monitoring and feedback.

[0075] The system monitors task execution and system status in real time, and collects execution data as feedback for continuous improvement of models and strategies.

[0076] Figure 3 A schematic diagram of the structure of a task scheduling device provided by an embodiment of the present application is shown as follows: Figure 3 As shown, the task scheduling device 300 may include: an acquisition module 310 , a first determination module 320 , a second determination module 330 , a third determination module 340 and a scheduling module 350 .

[0077] In this embodiment, an acquisition module 310 is used to acquire the priority and task characteristics corresponding to the target task; a first determination module 320 is used to determine the task requirements for executing the target task based on the task characteristics, wherein the task requirements include an estimated execution time length and an estimated resource requirement; a second determination module 330 is used to determine at least one candidate execution node for executing the target task based on the priority; a third determination module 340 is used to determine a target execution node for executing the target task from at least one of the candidate execution nodes based on the task requirements and the load level corresponding to each of the candidate execution nodes; and a scheduling module 350 is used to schedule the target execution node to execute the target task.

[0078] In one implementation, the second determination module 330 is specifically used to: when the priority is the first priority, determine that at least one of the candidate execution nodes includes at least one edge execution node; when the priority is the second priority, determine that at least one of the candidate execution nodes includes at least one central execution node, wherein the second priority is lower than the first priority.

[0079] In one implementation, the third determination module 340 is specifically used to: input the task requirements and the load levels corresponding to each of the candidate execution nodes into a task scheduling model to obtain the target execution node output by the task scheduling model, wherein the task scheduling model is trained based on a hybrid scheduling strategy.

[0080] In one implementation, the first determination module 320 is specifically used to: input the task characteristics into a first prediction model and a second prediction model respectively, to obtain an estimated execution time length output by the first prediction model and an estimated resource requirement output by the second prediction model, wherein the first prediction model and the second prediction model are constructed based on a regression model.

[0081] In one implementation, it also includes: an adjustment module, which is used to dynamically adjust the load level of each candidate execution node based on a load balancing strategy.

[0082] In one implementation, it also includes: a feedback module, which is used to obtain feedback analysis results by monitoring the execution process; based on the feedback analysis results, dynamically adjust the resource configuration of the target execution node and / or optimize the task scheduling model.

[0083] A task scheduling device provided in an embodiment of the present application can implement each process implemented in the task scheduling method embodiment, and will not be described again here to avoid repetition.

[0084] The task scheduling device in the embodiment of the present application may be a device, or a component, an integrated circuit, or a chip in an electronic device. The embodiment of the present application is not specifically limited.

[0085] The task scheduling device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0086] Optional, such as Figure 4As shown, an embodiment of the present application also provides an electronic device 400, including a processor 410, a memory 420, and a program or instruction stored in the memory 420 and executable on the processor 410. When the program or instruction is executed by the processor 410, each process of the above-mentioned task scheduling method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0087] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned task scheduling method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0088] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0089] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned task scheduling method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0090] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0091] An embodiment of the present application also provides a computer program product, which includes at least one computer program. When the computer program is loaded and executed by a processor, it implements the various processes of the above-mentioned task scheduling method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0092] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0094] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A task scheduling method, characterized in that: The method comprises: Obtain the priority and task characteristics corresponding to the target task; Based on the task characteristics, determining the task requirements for executing the target task, wherein the task requirements include an estimated execution time and an estimated resource requirement; Based on the priority, determining at least one candidate execution node for executing the target task; Determining a target execution node for executing the target task from at least one of the candidate execution nodes according to the task requirement and the load level corresponding to each of the candidate execution nodes; The target execution node is scheduled to execute the target task.

2. The method according to claim 1, characterized in that The determining, based on the priority, at least one candidate execution node for executing the target task includes: When the priority is the first priority, determining that at least one of the candidate execution nodes includes at least one edge execution node; In a case where the priority is a second priority, determining that at least one of the candidate execution nodes includes at least one central execution node, wherein the second priority is lower than the first priority.

3. The method according to claim 1, characterized in that The step of determining a target execution node for executing the target task from at least one of the candidate execution nodes according to the task requirement and a load level corresponding to each of the candidate execution nodes includes: The task requirement and the load level corresponding to each of the candidate execution nodes are input into a task scheduling model to obtain the target execution node output by the task scheduling model, wherein the task scheduling model is trained based on a hybrid scheduling strategy.

4. The method according to claim 1, characterized in that: The determining, based on the task characteristics, the task requirements for executing the target task includes: The task characteristics are input into a first prediction model and a second prediction model respectively to obtain an estimated execution time length output by the first prediction model and an estimated resource requirement output by the second prediction model, wherein the first prediction model and the second prediction model are constructed based on a regression model.

5. The method according to claim 1, characterized in that The method further comprises: Based on the load balancing strategy, the load level of each candidate execution node is dynamically adjusted.

6. The method according to claim 3, characterized in that After scheduling the target execution node to execute the target task, the method further includes: Obtain feedback and analysis results by monitoring the execution process; Based on the feedback analysis result, the resource configuration of the target execution node is dynamically adjusted and / or the task scheduling model is optimized.

7. A task scheduling device, characterized in that: include: An acquisition module is used to obtain the priority and task characteristics corresponding to the target task; A first determination module is used to determine the task requirements for executing the target task based on the task characteristics, wherein the task requirements include an estimated execution time and an estimated resource requirement; A second determination module, configured to determine, based on the priority, at least one candidate execution node for executing the target task; A third determination module, configured to determine a target execution node for executing the target task from at least one of the candidate execution nodes according to the task requirement and a load level corresponding to each of the candidate execution nodes; A scheduling module is used to schedule the target execution node to execute the target task.

8. An electronic device, characterized in that: include: A processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the task scheduling method according to any one of claims 1 to 6.

9. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the task scheduling method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises at least one computer program, and when the computer program is loaded and executed by a processor, the steps of the task scheduling method according to any one of claims 1 to 6 are implemented.

Images