Intelligent scheduling method and system for computing power network

Abstract

The invention relates to the technical field of computing power networks, in particular to an intelligent scheduling method and system for a computing power network. The method comprises the following steps: collecting computing power node data, network link data and task demand data of each node in a computing power network, and generating a standardized data sample; inputting the standardized data sample into a demand analysis model to complete a task demand portrait; combining the task demand level with the real-time resource state of the computing power network, constructing a multi-target optimization scheduling model, and solving the model to obtain an optimal scheduling scheme adaptive to the task and the network state; according to the optimal scheduling scheme, computing power resource allocation and task issuing are completed; the system comprises a data acquisition module, a task demand analysis module, an intelligent scheduling decision module and an execution module. Through the above mode, systematic consideration of the computing power resource state, the task demand characteristics and the dynamic adaptation adjustment is realized, and accurate dynamic matching of the computing power resource and the task demand can be realized.

Description

Technical Field

[0001] This invention relates to the field of computing power network technology, and in particular to a method and system for intelligent scheduling of computing power networks. Background Technology

[0002] With the booming development of the digital economy, computing power has become a core production factor supporting the implementation of emerging technologies such as artificial intelligence, big data, and cloud computing. As a key infrastructure for integrating distributed computing power resources and realizing on-demand allocation of computing power, the scheduling efficiency of computing power networks directly determines the utilization value of computing power resources and the quality of business services.

[0003] Currently, computing networks encompass massive heterogeneous computing nodes, complex network links, and diverse task requirements, and scheduling technology is developing towards intelligence and precision. Existing computing network scheduling methods typically involve basic steps such as data collection, demand analysis, scheme decision-making, and execution monitoring. They attempt to integrate computing resources with task demand information to achieve rational resource allocation, meet service needs in different business scenarios, and promote the large-scale application of computing networks.

[0004] However, the existing computing power network scheduling process lacks a systematic consideration of the status of computing power resources, task requirements, and dynamic adaptation and adjustment, making it difficult to achieve accurate dynamic matching between computing power resources and task requirements. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for intelligent scheduling of computing power networks, which aims to solve the technical problem that the existing computing power network scheduling process lacks a systematic consideration of the status of computing power resources, task requirements and dynamic adaptation, making it difficult to achieve accurate dynamic matching of computing power resources and task requirements.

[0006] To achieve the above objectives, the present invention employs an intelligent scheduling method for computing power networks, comprising the following steps: Collect computing node data, network link data, and task requirement data from each node in the computing power network, and preprocess the collected data to generate standardized data samples. Input standardized data samples into the demand analysis model to complete the task demand profile and output its corresponding three demand levels: computing power, latency, and reliability. By combining the task requirement level with the real-time resource status of the computing network, a multi-objective optimization scheduling model is constructed and solved to obtain the optimal scheduling scheme that adapts to the task and network status. The optimal scheduling scheme is used to allocate computing resources and distribute tasks, and the network operation status and task execution progress are monitored in real time to ensure the effectiveness of scheduling.

[0007] Among the steps, the following steps are involved: collecting computing node data, network link data, and task requirement data from each node in the computing power network, and preprocessing the collected data to generate standardized data samples: Deploy distributed data collection nodes to cover the entire computing power network, and collect computing power node data, network link data, and task requirement data for each computing power node; The three types of data collected were initially screened to remove obviously invalid and redundant data; For missing values ​​in the filtered data, interpolation is used to complete the data and make it complete. All the completed data are normalized to eliminate the dimensional differences between different types of data and generate standardized data samples.

[0008] Among them, the steps of deploying distributed collection nodes to cover the entire computing power network and collecting computing power node data, network link data, and task requirement data for each computing power node are as follows: Computing node data includes CPU utilization, memory usage, remaining storage capacity, and computing power output; network link data includes link bandwidth, transmission latency, packet loss rate, and link load; task requirement data includes task data volume, computational complexity, task priority, latency threshold, and reliability requirements.

[0009] Among them, in the steps of inputting standardized data samples into the demand analysis model, completing the task demand profile, and outputting its corresponding three demand levels of computing power, latency, and reliability: Build a requirements analysis model and train the model to learn the correlation between different types of task data and requirements levels. Standardized data samples are input and adapted according to the format required by the model and then passed into the requirements analysis model; The requirements analysis model is used to perform feature matching and pattern recognition on the input data to construct a requirements profile that includes the core requirements of the task. Based on the demand profile, the computing power demand level, latency demand level, and reliability demand level corresponding to the task are determined and output respectively.

[0010] In the step of determining and outputting the computing power requirement level, latency requirement level, and reliability requirement level corresponding to the task based on the requirement profile: The computing power requirement is divided into three levels: high, medium, and low; the latency requirement is divided into three levels: stringent, normal, and lenient; and the reliability requirement is divided into three levels: special, level one, and level two.

[0011] Among the steps, the following steps are involved: constructing and solving a multi-objective optimization scheduling model by combining the task requirement level with the real-time resource status of the computing power network to obtain the optimal scheduling scheme that adapts to the task and network status: The system summarizes the task requirement levels and synchronously collects the current real-time resource distribution status of the computing power network, including information on the remaining computing power of each node and the current available bandwidth of each link. With the goals of maximizing resource utilization, minimizing task execution latency, and minimizing scheduling energy consumption, a multi-objective optimization scheduling model is constructed by combining the constraints corresponding to the task requirement levels. An intelligent optimization algorithm is used to solve the multi-objective optimization scheduling model, searching for a scheduling solution that satisfies all constraints and can achieve multi-objective balance. The feasibility of the obtained scheduling solution is verified, and the optimal solution that adapts to the current task requirements and network resource status is selected and transformed into an optimal scheduling scheme that includes the target computing power node, transmission link path, and resource allocation ratio.

[0012] Among these steps, the allocation of computing resources and task distribution are completed according to the optimal scheduling scheme, and the network operation status and task execution progress are monitored in real time to ensure the effectiveness of scheduling: The optimal scheduling scheme is analyzed, resource allocation instructions and task issuance instructions are generated, and then issued to the corresponding computing power nodes and network devices through the computing power network management and control platform. After the target computing node receives the instruction, it reserves a corresponding proportion of computing power resources and memory resources. The network device adjusts the link configuration according to the instruction, and after completing the resource allocation, it receives and starts the task execution. Deploy monitoring nodes to collect real-time data on the resource usage status of computing nodes, the transmission status of network links, and the execution progress of tasks; Real-time analysis of monitoring data is performed to determine whether there are any abnormal situations such as resource overload, link congestion, or task execution delays.

[0013] This invention also provides an intelligent scheduling system for computing power networks, comprising a data acquisition module, a task requirement analysis module, an intelligent scheduling decision module, and an execution module; wherein: The data acquisition module is used to collect computing node data, network link data and task requirement data of each node in the computing power network, and to preprocess the collected data to generate standardized data samples. The task requirement analysis module is used to input standardized data samples into the requirement analysis model, complete the task requirement profile, and output its corresponding three requirement levels: computing power, latency, and reliability. The intelligent scheduling decision module is used to combine the task requirement level and the real-time resource status of the computing network to construct and solve a multi-objective optimization scheduling model, and obtain the optimal scheduling scheme that adapts to the task and network status. The execution module is used to allocate computing resources and issue tasks according to the optimal scheduling scheme, and to monitor the network operation status and task execution progress in real time to ensure the effectiveness of scheduling.

[0014] This invention discloses an intelligent scheduling method and system for computing power networks. The method comprises a data acquisition module, a task requirement analysis module, an intelligent scheduling decision module, and an execution module, performing the following steps: collecting computing power node data, network link data, and task requirement data from each node in the computing power network; preprocessing the collected data to generate standardized data samples; inputting the standardized data samples into a requirement analysis model to complete a task requirement profile and output its corresponding three requirement levels: computing power, latency, and reliability; combining the task requirement levels with the real-time resource status of the computing power network, constructing and solving a multi-objective optimization scheduling model to derive the optimal scheduling scheme that adapts to the task and network status; allocating computing power resources and issuing tasks according to the optimal scheduling scheme; and conducting real-time monitoring of network operation status and task execution progress to ensure scheduling effectiveness. Through this approach, a systematic consideration of computing power resource status, task requirement characteristics, and dynamic adaptation and adjustment is achieved, enabling precise dynamic matching of computing power resources and task requirements. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the intelligent scheduling method for computing power networks of the present invention.

[0017] Figure 2 This is a flowchart of steps S100 of the present invention.

[0018] Figure 3 This is a flowchart of steps S200 of the present invention.

[0019] Figure 4 This is a flowchart of steps S300 of the present invention.

[0020] Figure 5 This is a flowchart of steps S400 of the present invention.

[0021] Figure 6 This is a schematic diagram of the principle of the intelligent scheduling system for computing power networks of the present invention.

[0022] 501 - Data Acquisition Module, 502 - Task Requirements Analysis Module, 503 - Intelligent Scheduling Decision Module, 504 - Execution Module. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0024] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0025] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0026] Please see Figures 1-5 This invention provides an intelligent scheduling method for computing power networks, comprising the following steps: S100: Collects computing node data, network link data, and task requirement data from each node in the computing network, and preprocesses the collected data to generate standardized data samples.

[0027] In this embodiment, computing node data, network link data, and task requirement data of each node in the computing power network are collected, and the collected data is preprocessed to generate standardized data samples. The specific process is as follows: S101: Deploy distributed collection nodes to cover the entire computing power network, and collect computing power node data, network link data and task requirement data for each computing power node; S102: Perform preliminary screening on the three types of collected data to remove obviously invalid and redundant data; S103: For missing values ​​in the filtered data, use interpolation to complete the data and make it complete. S104: Normalize all the completed data to eliminate the dimensional differences between different types of data and generate standardized data samples.

[0028] In the above process, firstly, distributed acquisition nodes are deployed to cover the entire computing power network. Data is collected from each computing power node, network link data, and task requirement data. The computing power node data includes CPU utilization, memory usage, remaining storage capacity, and computing power output power; network link data includes link bandwidth, transmission latency, packet loss rate, and link load; and task requirement data includes task data volume, computational complexity, task priority, latency threshold, and reliability requirements. Then, the collected data undergoes preliminary screening to remove obviously invalid and redundant data. For missing values ​​in the screened data, interpolation is used to complete the data. All completed data is then normalized to eliminate dimensional differences between different data types, generating standardized data samples. The distributed acquisition nodes adopt an edge + center collaborative architecture. Edge nodes are responsible for real-time local data acquisition, while the center node is responsible for comprehensive data aggregation and verification. Simultaneously, the acquisition frequency is dynamically adjusted according to the task type: latency-sensitive tasks have increased acquisition frequency to ensure data timeliness, while non-latency-sensitive tasks have appropriately reduced frequency to conserve resources.

[0029] S200: Input standardized data samples into the requirements analysis model to complete the task requirements profile and output the corresponding three requirements levels: computing power, latency, and reliability.

[0030] In this implementation, standardized data samples are input into the demand analysis model to complete the task demand profile and output its corresponding three demand levels: computing power, latency, and reliability. The specific process is as follows: S201: Construct a requirements analysis model and train the model, learning the correlation between different types of task data and requirements levels. S202: Input standardized data samples into the requirements analysis model according to the required format; S203: Use the requirements analysis model to perform feature matching and pattern recognition on the input data to construct a requirements profile that includes the core requirements of the task; S204: Based on the demand profile, determine and output the corresponding computing power demand level, latency demand level, and reliability demand level for each task.

[0031] In the above process, firstly, a demand analysis model is constructed and trained to learn the correlation between different types of task data and demand levels. Then, standardized data samples are input and adapted according to the model's required format and fed into the demand analysis model. Next, the demand analysis model is used to perform feature matching and pattern recognition on the input data to construct a demand profile containing the core requirements of the task. Based on the demand profile, the corresponding computing power demand level, latency demand level, and reliability demand level are determined and output respectively. The computing power demand level is divided into three levels: high, medium, and low; the latency demand level is divided into three levels: stringent, normal, and lenient; and the reliability demand level is divided into three levels: special, level one, and level two. The demand analysis model will perform a credibility verification on the output demand level. If the verification fails, it returns to S100 to reacquire and process the relevant data of the corresponding task to ensure the accuracy of the demand level determination.

[0032] S300: Combining task requirement levels with the real-time resource status of the computing network, a multi-objective optimization scheduling model is constructed and solved to obtain the optimal scheduling scheme that adapts to the task and network status.

[0033] In this embodiment, a multi-objective optimization scheduling model is constructed and solved by combining the task requirement level and the real-time resource status of the computing network to obtain the optimal scheduling scheme that adapts to the task and network status. The specific process is as follows: S301: Summarizes the output of each task requirement level and synchronously collects the current real-time resource distribution status of the computing power network, including information on the remaining computing power of each node and the current available bandwidth of each link. S302: With the goals of maximizing resource utilization, minimizing task execution latency, and minimizing scheduling energy consumption, a multi-objective optimization scheduling model is constructed by combining the constraints corresponding to the task requirement level. S303: Use intelligent optimization algorithms to solve the multi-objective optimization scheduling model, and search for a scheduling solution that satisfies all constraints and can achieve multi-objective balance; S304: Verify the feasibility of the obtained scheduling solution, select the optimal solution that fits the current task requirements and network resource status, and transform it into an optimal scheduling scheme that includes the target computing power node, transmission link path, and resource allocation ratio.

[0034] In the above process, firstly, the output task requirement levels are summarized, and the real-time resource distribution status of the computing network is collected synchronously, including the remaining computing power of each node and the current available bandwidth of each link. With the objectives of maximizing resource utilization, minimizing task execution latency, and minimizing scheduling energy consumption, a multi-objective optimization scheduling model is constructed, combined with the constraints corresponding to the task requirement levels. Then, an intelligent optimization algorithm is used to solve the multi-objective optimization scheduling model, searching for a scheduling solution that satisfies all constraints and achieves multi-objective balance. Next, the feasibility of the obtained scheduling solution is verified, and the optimal solution that adapts to the current task requirements and network resource status is selected, transforming it into an optimal scheduling scheme that includes target computing power nodes, transmission link paths, and resource allocation ratios. During model construction, the weights of each optimization objective are adjusted according to task priority, with high-priority tasks prioritizing the objectives corresponding to their core requirements. During the solution process, changes in network resource status are synchronized in real time to ensure the feasibility of the scheduling scheme.

[0035] S400: Allocates computing resources and distributes tasks according to the optimal scheduling scheme, and conducts real-time monitoring of network operation status and task execution progress to ensure scheduling effectiveness.

[0036] In this implementation, computing resources are allocated and tasks are distributed according to the optimal scheduling scheme, and the network operation status and task execution progress are monitored in real time to ensure the effectiveness of scheduling. The specific process is as follows: S401: Analyzes the optimal scheduling scheme, generates resource allocation instructions and task issuance instructions, and issues them to the corresponding computing power nodes and network devices through the computing power network management and control platform; S402: After the target computing node receives the instruction, it reserves a corresponding proportion of computing power resources and memory resources. The network device adjusts the link configuration according to the instruction, and after completing the resource allocation, it receives and starts the task execution. S403: Deploy monitoring nodes to collect real-time data on the resource usage status of computing nodes, the transmission status of network links, and the execution progress of tasks; S404: Perform real-time analysis of monitoring data to determine whether there are any abnormal situations such as resource overload, link congestion, or task execution delay.

[0037] In the above process, the optimal scheduling scheme is first analyzed to generate resource allocation instructions and task issuance instructions, which are then issued to the corresponding computing power nodes and network devices through the computing power network management platform. After the target computing power node receives the instructions, it reserves the corresponding proportion of computing power resources and memory resources. The network device adjusts the link configuration according to the instructions, completes the resource allocation, and then receives and starts task execution. Then, monitoring nodes are deployed to collect the resource occupancy status of computing power nodes, the transmission status of network links, and the task execution progress data in real time. The monitoring data is then analyzed in real time to determine whether there are any abnormal situations such as resource overload, link congestion, or task execution delay. If an abnormal situation is detected, the corresponding adjustment mechanism will be triggered according to the severity of the abnormality. For minor abnormalities, the resource allocation ratio will be fine-tuned. For severe abnormalities, the process will return to S200 to re-analyze the requirements and generate a new scheduling scheme to ensure the effectiveness of scheduling execution and the stable operation of the computing power network.

[0038] Please see Figure 6 This invention provides an intelligent scheduling system for computing power networks, comprising a data acquisition module 501, a task requirement analysis module 502, an intelligent scheduling decision module 503, and an execution module 504; wherein: The data acquisition module 501 is used to collect computing node data, network link data and task requirement data of each node in the computing power network, and to preprocess the collected data to generate standardized data samples. The task requirement analysis module 502 is used to input standardized data samples into the requirement analysis model, complete the task requirement profile, and output its corresponding three requirement levels: computing power, latency, and reliability. The intelligent scheduling decision module 503 is used to combine the task requirement level and the real-time resource status of the computing network to construct a multi-objective optimization scheduling model and solve it to obtain the optimal scheduling scheme that adapts to the task and network status. The execution module 504 is used to complete the allocation of computing resources and task distribution according to the optimal scheduling scheme, and to carry out real-time monitoring of network operation status and task execution progress to ensure the effectiveness of scheduling.

[0039] In this embodiment, the data acquisition module 501 collects computing node data, network link data, and task requirement data from each node in the computing network, and preprocesses the collected data to generate standardized data samples. The task requirement analysis module 502 inputs the standardized data samples into the requirement analysis model to complete the task requirement profile and output its corresponding computing power, latency, and reliability requirement levels. The intelligent scheduling decision module 503 combines the task requirement level with the real-time resource status of the computing network to construct and solve a multi-objective optimization scheduling model, deriving the optimal scheduling scheme that adapts to the task and network status. The execution module 504 completes the allocation of computing resources and task distribution according to the optimal scheduling scheme, and conducts real-time monitoring of the network operation status and task execution progress to ensure the effectiveness of scheduling.

[0040] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0041] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for intelligent scheduling of computing power networks, characterized in that, Includes the following steps: Collect computing node data, network link data, and task requirement data from each node in the computing power network, and preprocess the collected data to generate standardized data samples. Input standardized data samples into the demand analysis model to complete the task demand profile and output its corresponding three demand levels: computing power, latency, and reliability. By combining the task requirement level with the real-time resource status of the computing network, a multi-objective optimization scheduling model is constructed and solved to obtain the optimal scheduling scheme that adapts to the task and network status. The optimal scheduling scheme is used to allocate computing resources and distribute tasks, and the network operation status and task execution progress are monitored in real time to ensure the effectiveness of scheduling.

2. The intelligent scheduling method for computing power networks as described in claim 1, characterized in that, In the steps of collecting computing node data, network link data, and task requirement data from each node in the computing power network, and preprocessing the collected data to generate standardized data samples: Deploy distributed data collection nodes to cover the entire computing power network, and collect computing power node data, network link data, and task requirement data for each computing power node; The three types of data collected were initially screened to remove obviously invalid and redundant data; For missing values ​​in the filtered data, interpolation is used to complete the data and make it complete. All the completed data are normalized to eliminate the dimensional differences between different types of data and generate standardized data samples.

3. The intelligent scheduling method for computing power networks as described in claim 2, characterized in that, In the steps of deploying distributed data collection nodes to cover the entire computing power network, and collecting computing power node data, network link data, and task requirement data from each computing power node: Computing node data includes CPU utilization, memory usage, remaining storage capacity, and computing power output; network link data includes link bandwidth, transmission latency, packet loss rate, and link load; task requirement data includes task data volume, computational complexity, task priority, latency threshold, and reliability requirements.

4. The intelligent scheduling method for computing power networks as described in claim 1, characterized in that, In the steps of inputting standardized data samples into the demand analysis model, completing the task demand profile, and outputting its corresponding three demand levels of computing power, latency, and reliability: Build a requirements analysis model and train the model to learn the correlation between different types of task data and requirements levels. Standardized data samples are input and adapted according to the format required by the model and then passed into the requirements analysis model; The requirements analysis model is used to perform feature matching and pattern recognition on the input data to construct a requirements profile that includes the core requirements of the task. Based on the demand profile, the computing power demand level, latency demand level, and reliability demand level corresponding to the task are determined and output respectively.

5. The intelligent scheduling method for computing power networks as described in claim 4, characterized in that, In the steps of determining and outputting the computing power requirement level, latency requirement level, and reliability requirement level corresponding to the task based on the requirement profile: The computing power requirement is divided into three levels: high, medium, and low; the latency requirement is divided into three levels: stringent, normal, and lenient; and the reliability requirement is divided into three levels: special, level one, and level two.

6. The intelligent scheduling method for computing power networks as described in claim 1, characterized in that, In the steps of combining task requirement levels and real-time computing network resource status, constructing and solving a multi-objective optimization scheduling model to obtain the optimal scheduling scheme that adapts to the task and network status: The system summarizes the task requirement levels and synchronously collects the current real-time resource distribution status of the computing power network, including information on the remaining computing power of each node and the current available bandwidth of each link. With the goals of maximizing resource utilization, minimizing task execution latency, and minimizing scheduling energy consumption, a multi-objective optimization scheduling model is constructed by combining the constraints corresponding to the task requirement levels. An intelligent optimization algorithm is used to solve the multi-objective optimization scheduling model, searching for a scheduling solution that satisfies all constraints and can achieve multi-objective balance. The feasibility of the obtained scheduling solution is verified, and the optimal solution that adapts to the current task requirements and network resource status is selected and transformed into an optimal scheduling scheme that includes the target computing power node, transmission link path, and resource allocation ratio.

7. The intelligent scheduling method for computing power networks as described in claim 1, characterized in that, In the steps of allocating computing resources and issuing tasks according to the optimal scheduling scheme, and conducting real-time monitoring of network operation status and task execution progress to ensure the effectiveness of scheduling: The optimal scheduling scheme is analyzed, resource allocation instructions and task issuance instructions are generated, and then issued to the corresponding computing power nodes and network devices through the computing power network management and control platform. After the target computing node receives the instruction, it reserves a corresponding proportion of computing power resources and memory resources. The network device adjusts the link configuration according to the instruction, and after completing the resource allocation, it receives and starts the task execution. Deploy monitoring nodes to collect real-time data on the resource usage status of computing nodes, the transmission status of network links, and the execution progress of tasks; Real-time analysis of monitoring data is performed to determine whether there are any abnormal situations such as resource overload, link congestion, or task execution delays.

8. A computing power network intelligent scheduling system, applied to the computing power network intelligent scheduling method as described in claim 1, characterized in that, It includes a data acquisition module, a task requirements analysis module, an intelligent scheduling decision-making module, and an execution module; among which: The data acquisition module is used to collect computing node data, network link data and task requirement data of each node in the computing power network, and to preprocess the collected data to generate standardized data samples. The task requirement analysis module is used to input standardized data samples into the requirement analysis model, complete the task requirement profile, and output its corresponding three requirement levels: computing power, latency, and reliability. The intelligent scheduling decision module is used to combine the task requirement level and the real-time resource status of the computing network to construct and solve a multi-objective optimization scheduling model, and obtain the optimal scheduling scheme that adapts to the task and network status. The execution module is used to allocate computing resources and issue tasks according to the optimal scheduling scheme, and to monitor the network operation status and task execution progress in real time to ensure the effectiveness of scheduling.

Images