AI-based efficient task arrangement system and arrangement method thereof

Abstract

The invention provides an AI-based efficient task arrangement system and an arrangement method thereof. The system is mainly composed of a task management module, a resource monitoring module, a scheduling engine module, an execution module and a data analysis module. The task management module is used for creating, distributing and managing tasks; the resource monitoring module is used for monitoring the resource use condition of each computing node in real time; and the scheduling engine module performs task scheduling and resource allocation based on an AI algorithm. According to the AI-based efficient task arrangement system and the arrangement method thereof, through intelligent and self-adaptive task arrangement, the task processing efficiency and speed can be remarkably improved, the intellectualization and the self-adaption of task arrangement are realized by introducing an artificial intelligence technology, and the defects of the prior art in the aspects of flexibility, intellectualization and the like are overcome. Meanwhile, the system can adjust a scheduling strategy in real time in a dynamically changing computing environment, optimizes resource allocation, and improves the overall performance and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of task scheduling, and in particular to an AI-based efficient task scheduling system and a scheduling method thereof. Background Art

[0002] In modern computing environments, task scheduling and resource management are key technologies to ensure efficient operation of the system. With the development of cloud computing, big data processing, distributed computing, and the Internet of Things, the importance of task scheduling technology has become increasingly prominent. Although existing task scheduling technologies have solved the problems of task scheduling and resource allocation to a certain extent, they still have many shortcomings. Traditional static task scheduling methods: Traditional task scheduling methods are usually based on predefined rules or static algorithms, such as polling scheduling, shortest task priority scheduling, etc. These methods perform well when the task type and system load are relatively fixed, but in dynamic and complex computing environments, they are often difficult to cope with sudden task requirements and resource changes. Task scheduling based on heuristic algorithms: Some existing technologies use heuristic algorithms (such as genetic algorithms, ant colony algorithms, etc.) for task scheduling, trying to improve scheduling efficiency by simulating the optimization process of nature. Although these methods have improved the flexibility of scheduling to a certain extent, due to the high computational complexity of heuristic algorithms, they are often difficult to apply in scenarios with high real-time requirements. Rule-based task scheduling: Other task scheduling systems are based on predefined rules for scheduling, such as allocating tasks according to factors such as task priority and resource requirements. This method has poor flexibility and adaptability and is difficult to cope with complex and changing computing environments.

[0003] Lack of flexibility and adaptability: Traditional static scheduling methods and rule-based scheduling methods often show insufficient flexibility and adaptability when faced with dynamically changing task requirements and system loads, making it difficult to adjust scheduling strategies in real time.

[0004] High real-time performance and computational complexity: Although task scheduling based on heuristic algorithms improves scheduling efficiency to a certain extent, its computational complexity is high and it is difficult to meet application scenarios with high real-time requirements.

[0005] Lack of intelligence and adaptability: Most existing task scheduling systems lack intelligence and adaptability and are unable to perform dynamic optimization based on actual operating conditions, resulting in low resource utilization and difficulty in fully utilizing system performance.

[0006] Existing task orchestration technologies have problems such as insufficient flexibility, high real-time and computational complexity, centralized architecture bottlenecks and risks, and lack of intelligence and adaptability. Summary of the invention

[0007] In view of the above defects or deficiencies in the prior art, the present application aims to provide an efficient task scheduling system based on AI, which is mainly composed of the following task management module, resource monitoring module, scheduling engine module, execution module and data analysis module;

[0008] The task management module is responsible for the creation, allocation and management of tasks;

[0009] The resource monitoring module is used to monitor the resource usage of each computing node in real time;

[0010] The scheduling engine module performs task scheduling and resource allocation based on AI algorithm;

[0011] The execution module is responsible for the execution and feedback of specific tasks;

[0012] The data analysis module is used to analyze historical data and real-time data and optimize scheduling strategies.

[0013] An efficient task scheduling method based on AI, the efficient task scheduling method is used in an efficient task scheduling system based on AI, and the efficient task scheduling method based on AI comprises the following steps:

[0014] Step 1: Task creation: Users create new tasks through the task management module and specify the basic information and requirements of the task;

[0015] Step 2: Resource status monitoring. The resource monitoring module monitors the resource usage of each computing node in real time, including CPU, memory, and network.

[0016] Step 3: Scheduling strategy generation: The scheduling engine module generates the optimal scheduling strategy using AI algorithms based on real-time resource status and historical data.

[0017] Step 4: Task allocation: assign tasks to appropriate computing nodes according to the generated scheduling strategy;

[0018] Step 5: Task execution: The execution module executes the task on the specified computing node and monitors the execution status of the task in real time;

[0019] Step 6: Feedback of execution results: After the task is completed, the execution results are fed back to the task management module;

[0020] Step 7: Data analysis: The data analysis module analyzes the execution results and historical data to further optimize the scheduling strategy and improve the overall performance of the system.

[0021] Preferably, the principle of scheduling strategy generation includes the following steps:

[0022] S1. Environment modeling: Model the task scheduling problem as a reinforcement learning environment and define the state, action, and reward functions;

[0023] S2. Strategy training: The scheduling strategy is trained using reinforcement learning algorithms. The model optimizes the scheduling strategy through continuous trial and error and feedback.

[0024] S3, Strategy testing and verification: Test and verify the trained strategy model to ensure the effectiveness of the strategy;

[0025] S4, real-time input of current status: during the task execution, the current status is input in real time;

[0026] S5, Scheduling decision: Use the trained strategy model to make scheduling decisions;

[0027] S6, Task allocation: Assign tasks to appropriate computing nodes based on the decision of the reinforcement learning model;

[0028] S7, Task execution and feedback: Execute tasks and collect execution results and feedback data;

[0029] S8, reward feedback: According to the task execution results, the reward value is calculated and fed back to the strategy model;

[0030] S9, Strategy update: Update the strategy model in real time or periodically to adapt to new tasks and resource status.

[0031] Preferably, the reinforcement learning algorithm includes Q-learning or DQN algorithm.

[0032] Preferably, the real-time input current status is task characteristics and resource status.

[0033] Preferably, the AI ​​algorithm comprises a deep learning or reinforcement learning algorithm.

[0034] Beneficial effects:

[0035] 1. This AI-based dynamic task scheduling method uses deep learning and reinforcement learning algorithms to dynamically adjust scheduling strategies based on real-time monitoring of system status and task requirements.

[0036] Efficient AI algorithm: Introducing reinforcement learning algorithms and deep learning technologies, analyzing historical data and real-time feedback, and optimizing task allocation and resource utilization.

[0037] Distributed task scheduling architecture: With a distributed architecture, each computing node works together to complete task scheduling and resource management, eliminating the performance bottleneck and single point failure risk of a centralized scheduler.

[0038] Intelligent and adaptive capabilities: Use machine learning and data analysis capabilities to achieve intelligent and adaptive task scheduling, and autonomously adjust and optimize scheduling strategies;

[0039] 2. Through intelligent and adaptive task scheduling, the present invention can significantly improve the efficiency and speed of task processing. By introducing artificial intelligence technology, the intelligent and adaptive task scheduling is realized, solving the deficiencies of the existing technology in terms of flexibility and intelligence. At the same time, the system can adjust the scheduling strategy in real time in a dynamically changing computing environment, optimize resource allocation, and improve overall performance and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0041] Figure 1 This is a system architecture diagram of the present invention;

[0042] Figure 2 It is the task scheduling flow chart of the present invention;

[0043] Figure 3 Schematic diagram of the scheduling strategy generated by the present invention. DETAILED DESCRIPTION

[0044] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0045] The accompanying drawings in the embodiments of the present invention: different types of section lines in the drawings are not marked according to national standards, nor do they impose requirements on the materials of the components. Instead, they are used to distinguish the cross-sectional views of the components in the drawings.

[0046] See also Figure 1-3 , an efficient task scheduling system based on AI, which is mainly composed of the following task management module, resource monitoring module, scheduling engine module, execution module and data analysis module;

[0047] The task management module is responsible for the creation, allocation and management of tasks;

[0048] The resource monitoring module is used to monitor the resource usage of each computing node in real time;

[0049] The scheduling engine module performs task scheduling and resource allocation based on AI algorithm;

[0050] The execution module is responsible for the execution and feedback of specific tasks;

[0051] The data analysis module is used to analyze historical data and real-time data and optimize scheduling strategies.

[0052] An efficient task scheduling method based on AI, the efficient task scheduling method is used in an efficient task scheduling system based on AI, and the efficient task scheduling method based on AI comprises the following steps:

[0053] Step 1: Task creation: Users create new tasks through the task management module and specify the basic information and requirements of the task;

[0054] Step 2: Resource status monitoring. The resource monitoring module monitors the resource usage of each computing node in real time, including CPU, memory, and network.

[0055] Step 3: Scheduling strategy generation: The scheduling engine module generates the optimal scheduling strategy using AI algorithms based on real-time resource status and historical data.

[0056] Step 4: Task allocation: assign tasks to appropriate computing nodes according to the generated scheduling strategy;

[0057] Step 5: Task execution: The execution module executes the task on the specified computing node and monitors the execution status of the task in real time;

[0058] Step 6: Feedback of execution results: After the task is completed, the execution results are fed back to the task management module;

[0059] Step 7: Data analysis: The data analysis module analyzes the execution results and historical data to further optimize the scheduling strategy and improve the overall performance of the system.

[0060] The principle of scheduling strategy generation includes the following steps:

[0061] S1. Environment modeling: Model the task scheduling problem as a reinforcement learning environment and define the state, action, and reward functions;

[0062] S2. Strategy training: The scheduling strategy is trained using reinforcement learning algorithms. The model optimizes the scheduling strategy through continuous trial and error and feedback.

[0063] S3, Strategy testing and verification: Test and verify the trained strategy model to ensure the effectiveness of the strategy;

[0064] S4, real-time input of current status: during the task execution, the current status is input in real time;

[0065] S5, Scheduling decision: Use the trained strategy model to make scheduling decisions;

[0066] S6, Task allocation: Assign tasks to appropriate computing nodes based on the decision of the reinforcement learning model;

[0067] S7, Task execution and feedback: Execute tasks and collect execution results and feedback data;

[0068] S8, reward feedback: According to the task execution results, the reward value is calculated and fed back to the strategy model;

[0069] S9, Strategy update: Update the strategy model in real time or periodically to adapt to new tasks and resource status.

[0070] Among them, the reinforcement learning algorithm includes Q-learning or DQN algorithm.

[0071] Among them, the real-time input current status is task characteristics and resource status.

[0072] Among them, AI algorithms include deep learning or reinforcement learning algorithms.

[0073] Example 1

[0074] Task scheduling based on deep learning, in this embodiment, the scheduling engine module uses deep learning algorithm to schedule tasks. The specific steps are as follows:

[0075] Data preprocessing: Collect historical task execution data and resource usage data, and perform data cleaning and preprocessing.

[0076] Model training: Use the preprocessed data to train the deep learning model. The model input is task characteristics and resource status, and the output is the scheduling strategy.

[0077] Real-time prediction: During the task execution process, the current task characteristics and resource status are input in real time, and the trained model is used to predict the scheduling strategy.

[0078] Task allocation: Assign tasks to appropriate computing nodes based on the predicted scheduling strategy.

[0079] Example 2

[0080] Task scheduling based on reinforcement learning. In this embodiment, the scheduling engine module uses reinforcement learning algorithm to schedule tasks. The specific steps are as follows:

[0081] Environment modeling: Model the task scheduling problem as a reinforcement learning environment and define the state, action, and reward functions.

[0082] Strategy training: Use reinforcement learning algorithms (such as Q-learning and DQN) to train scheduling strategies. The model optimizes the scheduling strategy through continuous trial and feedback.

[0083] Strategy application: During task execution, the current status is input in real time, and the trained strategy model is used to make scheduling decisions.

[0084] Task allocation: Assign tasks to appropriate computing nodes based on the decisions of the reinforcement learning model.

[0085] It should also be noted that the present invention aims to solve the problems of insufficient flexibility, high real-time and computational complexity, centralized architecture bottlenecks and risks, lack of intelligence and adaptive capabilities, etc. in existing task scheduling technologies. Through intelligent and adaptive task scheduling, the present invention can significantly improve the efficiency and speed of task processing. By introducing artificial intelligence technology, the intelligent and adaptive task scheduling is realized, solving the deficiencies of existing technologies in terms of flexibility and intelligence. At the same time, the system can adjust the scheduling strategy in real time in a dynamically changing computing environment, optimize resource allocation, and improve overall performance and reliability.

[0086] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0087] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. An efficient task scheduling system based on AI, characterized by: The system mainly consists of the following task management module, resource monitoring module, scheduling engine module, execution module and data analysis module; The task management module is responsible for the creation, allocation and management of tasks; The resource monitoring module is used to monitor the resource usage of each computing node in real time; The scheduling engine module performs task scheduling and resource allocation based on AI algorithm; The execution module is responsible for the execution and feedback of specific tasks; The data analysis module is used to analyze historical data and real-time data and optimize scheduling strategies.

2. An efficient task scheduling method based on AI, which is used in an efficient task scheduling system based on AI, and is characterized by: The AI-based efficient task scheduling method includes the following steps: Step 1: Task creation: Users create new tasks through the task management module and specify the basic information and requirements of the task; Step 2: Resource status monitoring. The resource monitoring module monitors the resource usage of each computing node in real time, including CPU, memory, and network. Step 3: Scheduling strategy generation: The scheduling engine module generates the optimal scheduling strategy using AI algorithms based on real-time resource status and historical data. Step 4: Task allocation: assign tasks to appropriate computing nodes according to the generated scheduling strategy; Step 5: Task execution: The execution module executes the task on the specified computing node and monitors the execution status of the task in real time; Step 6: Feedback of execution results: After the task is completed, the execution results are fed back to the task management module; Step 7: Data analysis: The data analysis module analyzes the execution results and historical data to further optimize the scheduling strategy and improve the overall performance of the system.

3. The AI-based efficient task scheduling method according to claim 2, characterized in that: The principle of scheduling strategy generation includes the following steps: S1. Environment modeling: Model the task scheduling problem as a reinforcement learning environment and define the state, action, and reward functions; S2. Strategy training: The scheduling strategy is trained using reinforcement learning algorithms. The model optimizes the scheduling strategy through continuous trial and error and feedback. S3, Strategy testing and verification: Test and verify the trained strategy model to ensure the effectiveness of the strategy; S4, real-time input of current status: during the task execution, the current status is input in real time; S5, Scheduling decision: Use the trained strategy model to make scheduling decisions; S6, Task allocation: Assign tasks to appropriate computing nodes based on the decision of the reinforcement learning model; S7, Task execution and feedback: Execute tasks and collect execution results and feedback data; S8, reward feedback: According to the task execution results, the reward value is calculated and fed back to the strategy model; S9, Strategy update: Update the strategy model in real time or periodically to adapt to new tasks and resource status.

4. The AI-based efficient task scheduling system and scheduling method according to claim 3, characterized in that: Reinforcement learning algorithms include Q-learning or DQN algorithms.

5. The AI-based efficient task scheduling method according to claim 3, characterized in that: Real-time input of current status for task characteristics and resource status.

6. The AI-based efficient task scheduling method according to claim 2, characterized in that: AI algorithms include deep learning or reinforcement learning algorithms.

Images