Integrated computing and networking resource scheduling system and method, electronic device, and storage medium

Abstract

The application relates to the technical field of computing power network, and provides a computing power network integrated resource scheduling system and method, an electronic device and a storage medium.The computing power network integrated resource scheduling system comprises a computing power network arrangement module, a computing power network scheduling module, a computing power network sensing module and an intelligent decision module; the computing power network sensing module is used for connecting the interface of the computing power network resource and collecting computing power network resource data; the computing power network resource comprises one or more of general computing power, supercomputing and intelligent computing; the intelligent decision module is used for providing computing power network resource arrangement strategies in various business scenarios based on preset artificial intelligence algorithms; the computing power network arrangement module is used for obtaining business demands, determining a target computing power network resource arrangement strategy from the computing power network resource arrangement strategies according to the business demands, performing computing power network resource scheduling arrangement on the computing power network resource data according to the target computing power network resource arrangement strategy in combination with the business demands, and obtaining a scheduling scheme; and the computing power network scheduling module is used for executing the scheduling scheme.The application can improve the scheduling efficiency of the computing power network resource.

Description

Technical Field

[0001] This application relates to the field of computing network technology, specifically to an integrated computing network resource scheduling system, method, electronic device, and storage medium. Background Technology

[0002] As a crucial infrastructure connecting computing power and networks, computing power networks have developed rapidly under the dual impetus of policy and social demand. With the development of computing power networks, underlying computing power resources are increasingly becoming the foundation for the digital transformation of countries and enterprises. As computing power resources increase, how to effectively match supply and demand and balance utilization has become a significant challenge.

[0003] In resource scheduling scenarios of computing networks, the following problems typically exist:

[0004] 1. Regarding resource scheduling, on the one hand, due to the heterogeneity problem, it is currently difficult to achieve unified scheduling of heterogeneous computing power such as general computing, supercomputing, and intelligent computing.

[0005] 2. At the business level, it is difficult to combine the deployment status, nodes, and locations of upper-layer applications with the underlying resources. At the same time, user scheduling requests are usually a collection of multiple factors, and it is impossible to quickly parse and respond when a large number of users are scheduling at the same time.

[0006] 3. Regarding the intelligent aspect, current industry scheduling of cloud, network, and other resources is mostly rule-based scheduling that is user-specified or rule-matched, resulting in insufficient accuracy and timeliness.

[0007] This results in low efficiency when scheduling computing network resources. Summary of the Invention

[0008] This application provides an integrated computing and network resource scheduling system, method, electronic device, and storage medium to solve the problem of low efficiency in current computing and network resource scheduling.

[0009] In a first aspect, embodiments of this application provide an integrated computing network resource scheduling system, including a computing network orchestration module, a computing network scheduling module, a computing network perception module, and an intelligent decision-making module;

[0010] The computing network sensing module is used to connect to the interface of the managed computing network resources and collect computing network resource data; the computing network resources include one or more of general computing power, supercomputing and intelligent computing;

[0011] The intelligent decision-making module is used to provide network resource orchestration strategies for various business scenarios based on pre-set artificial intelligence algorithms;

[0012] The computing network orchestration module is used to acquire business requirements, determine a target computing network resource orchestration strategy from various computing network resource orchestration strategies based on the business requirements, and perform computing network resource scheduling and orchestration on the computing network resource data based on the target computing network resource orchestration strategy and the business requirements to obtain a scheduling scheme.

[0013] The computing network scheduling module is used to execute the scheduling scheme.

[0014] In one embodiment, the network orchestration module includes an intent analysis unit, which is used for:

[0015] Obtain a scheduling request, identify the intent of the scheduling request, and obtain the business requirements corresponding to the scheduling request.

[0016] In one embodiment, the computing network scheduling module includes a task splitting unit and a task scheduling unit;

[0017] The task splitting unit is used to split the scheduling scheme to obtain a corresponding number of subtasks;

[0018] The task scheduling unit is used to schedule each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource.

[0019] In one embodiment, the computing network sensing module includes a sensing processing unit;

[0020] The perception processing unit is used to perform capacity expansion processing on the corresponding computing network resources if it is determined that the health of the corresponding computing network resources is lower than a preset health threshold during the process of the task scheduling unit scheduling each subtask to the corresponding computing network resources for execution through the interface of the managed computing network resources.

[0021] Secondly, embodiments of this application provide a method for integrated computing and network resource scheduling, including:

[0022] Obtain business requirements and collect computing network resource data from the managed computing network resources; the computing network resources include one or more of general computing power, supercomputing, and intelligent computing.

[0023] Obtain computing network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms;

[0024] Based on the aforementioned business requirements, determine the target computing network resource orchestration strategy from the various computing network resource orchestration strategies;

[0025] Based on the target computing network resource orchestration strategy and the business requirements, the computing network resource data is scheduled and orchestrated to obtain a scheduling scheme.

[0026] Execute the aforementioned scheduling scheme.

[0027] In one embodiment, obtaining business requirements includes:

[0028] Obtain a scheduling request, identify the intent of the scheduling request, and obtain the business requirements corresponding to the scheduling request.

[0029] In one embodiment, executing the scheduling scheme includes

[0030] The scheduling scheme is broken down into a corresponding number of subtasks;

[0031] Each subtask is scheduled to be executed on the corresponding computing network resource through the interface of the managed computing network resource.

[0032] In one embodiment, when scheduling each subtask to be executed on the corresponding computing network resource through the interface of the managed computing network resource, the method further includes:

[0033] If it is determined that the health of the corresponding computing network resource is lower than a preset health threshold, then the corresponding computing network resource will be expanded.

[0034] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the integrated computing and network resource scheduling method described in the second aspect.

[0035] Fourthly, embodiments of this application provide a storage medium, which is a computer-readable storage medium including a computer program. When the computer program is executed by a processor, it implements the integrated computer network resource scheduling method described in the second aspect.

[0036] The integrated computing and network resource scheduling system, method, electronic device, and storage medium provided in this application embodiment, through an integrated computing and network resource scheduling system including a computing and network orchestration module, a computing and network scheduling module, a computing and network perception module, and an intelligent decision-making module, can interface with one or more computing and network resources among the managed general computing power, supercomputing, and intelligent computing resources and collect computing and network resource data; provide computing and network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms; thereby, it can obtain business requirements, determine the target computing and network resource orchestration strategy from various computing and network resource orchestration strategies according to the business requirements, and perform computing and network resource scheduling and orchestration on the computing and network resource data according to the target computing and network resource orchestration strategy and business requirements to obtain a scheduling scheme; thereby, the scheduling scheme can be executed to realize the resource scheduling of the computing power network. By providing pre-defined network resource orchestration strategies for various business scenarios through artificial intelligence algorithms, network resource orchestration strategies can be quickly determined based on business requirements. Then, based on the determined network resource orchestration strategies and business requirements, network resource orchestration can be performed on heterogeneous network resource data across domains and vendors, such as general computing power, supercomputing, and intelligent computing. This enables rapid business response and intelligent and precise scheduling of heterogeneous network resources across domains and vendors, thereby improving the scheduling efficiency of network resources. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of the architecture of the integrated computing and network resource scheduling system provided in the embodiments of this application;

[0039] Figure 2 This is a schematic diagram of a scenario for the integrated computing and network resource scheduling system provided in this application embodiment;

[0040] Figure 3 This is a flowchart illustrating the integrated computing and network resource scheduling method provided in this application embodiment;

[0041] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. In the description of this specification, the reference to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., means that the specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the embodiments of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and combine the different embodiments or examples and the features of different embodiments or examples described in this specification. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0043] The following describes in detail the integrated computing and network resource scheduling system, method, electronic device, and storage medium provided in this application with reference to embodiments.

[0044] It should be noted that there are still many problems in the current integrated resource scheduling of computing networks: 1. From the perspective of scheduling form, most of them are currently resource-based scheduling services. How to combine resource-based services with task-based services needs to be solved urgently; 2. From the perspective of scheduling strategy, how to achieve intelligent scheduling from rule-based scheduling needs to be solved urgently.

[0045] Figure 1 This is a schematic diagram of the architecture of the integrated computing and network resource scheduling system provided in an embodiment of this application. (Refer to...) Figure 1 This application provides an integrated computing network resource scheduling system, which may include a computing network orchestration module, a computing network scheduling module, a computing network perception module, and an intelligent decision-making module.

[0046] The computing network perception module is used to connect to the interface of the managed computing network resources and collect computing network resource data; computing network resources include one or more of general computing power, supercomputing and intelligent computing.

[0047] The intelligent decision-making module is used to provide network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms;

[0048] The computing network orchestration module is used to obtain business requirements, determine the target computing network resource orchestration strategy from various computing network resource orchestration strategies based on the business requirements, and perform computing network resource scheduling and orchestration on computing network resource data based on the target computing network resource orchestration strategy and business requirements to obtain a scheduling scheme.

[0049] The network scheduling module is used to execute scheduling schemes.

[0050] In this application, users can input information through the integrated computing and network resource scheduling system and initiate scheduling requests based on the input information.

[0051] The computing network orchestration module in the integrated computing and network resource scheduling system can parse the received scheduling requests to form business requirements for computing power and network.

[0052] Furthermore, the computing network orchestration module can identify the business scenario in which the business requirement is located, and obtain the corresponding computing network resource orchestration strategy from the various computing network resource orchestration strategies provided by the intelligent decision-making module as the target computing network resource orchestration strategy based on the business scenario.

[0053] Furthermore, the computing network orchestration module can design and orchestrate a scheduling scheme for the computing network resource data collected by the computing network perception module according to business needs and the target computing network resource orchestration strategy, thereby forming a scheduling scheme for the business needs, and then sending the scheduling scheme to the computing network scheduling module.

[0054] After receiving the scheduling plan, the computing network scheduling module can execute the scheduling plan to complete the scheduling of computing network resources according to business needs.

[0055] It should be noted that, Figure 2 This is a schematic diagram illustrating a scenario of the integrated computing and network resource scheduling system provided in an embodiment of this application. (Refer to...) Figure 2 The network orchestration module of this application may include an intent analysis unit, which can obtain the scheduling request initiated by the user and perform intent recognition on the input information in the scheduling request; specifically, it can perform resource intent analysis and business intent analysis on the input information, and obtain the business requirements of the scheduling request after completing the analysis.

[0056] Resource intent analysis can include a comprehensive assessment of computing power, network, storage, etc., while business intent analysis analyzes business needs based on the type of application and task, as well as the industry and business tags of the customer.

[0057] The network orchestration module of this application may further include a capability policy management unit. This unit can provide unified management of orchestration capabilities and maintain the orchestration capability service catalog. It can also design and manage orchestration strategies, which are used for branching, looping, retries, and capability filtering during orchestration.

[0058] The computational network orchestration module of this application may also include a design center unit. This design center unit uses a visual approach to design and orchestrate execution templates using schemes and strategies, providing functions such as querying a scheme library and strategies, drag-and-drop functionality, and attribute configuration, allowing users to customize orchestration strategies. Simultaneously, it can also adjust the computational network resource orchestration strategies obtained from the intelligent decision-making module, supporting adjustments to the execution order of capabilities via a graphical interface using drag-and-drop methods.

[0059] The computing network scheduling module in this application can analyze the scheduling scheme to generate various sub-tasks, schedule each sub-task to the corresponding computing resources by calling the open interface of the managed computing network resources, and monitor and manage the execution process of the scheduling.

[0060] Specifically, the network scheduling module may include a scheduling strategy unit, which can provide design and management of scheduling strategies. The scheduling strategy is used to handle the initiation, monitoring, repetition, exception and error handling during scheduling execution, and mainly includes rule strategies, trigger strategies, task initiation strategies, repetition mechanisms, exception and error handling, etc.

[0061] The network scheduling module may also include a scheduling execution unit, which can initiate scheduling tasks according to the scheduling scheme, covering the entire lifecycle management of scheduling, including task management, task splitting, task scheduling, and task scheduling control.

[0062] Furthermore, the scheduling execution unit may include a task splitting unit and a task scheduling unit.

[0063] The task splitting unit can split the scheduling scheme into a corresponding number of subtasks;

[0064] The task scheduling unit can further schedule each subtask to the corresponding computing network resources for execution through the interface of the managed computing network resources.

[0065] The network scheduling module may also include a scheduling monitoring unit, which provides the ability to monitor ongoing scheduling tasks in real time. It can view the execution process, execution progress, node execution status, node execution start time, execution end time, duration, input data, output data, conversion rules, execution results, etc.

[0066] The computing network perception module in this application can connect to the unified open interface of the managed computing network resources, collect various perception data, including events and performance of objects such as computing power, network, and applications, and analyze the perception data to discover targets for further optimization of scheduling.

[0067] The computing network perception module can include a perception access unit. This unit can perceive data from computing network resources such as intelligent computing, general computing power, and supercomputing through the managed interface. It provides unified access to various data sources (such as resources, performance, alarms, events, logs, business data, applications, and services). It interfaces with managed applications and task statuses, allowing for task management and monitoring to track task execution. It can also retry or repeat abnormal and erroneous tasks according to scheduling rules. Furthermore, it provides unified processing of various data, such as data cleaning and data storage.

[0068] The network perception module may also include a perception analysis unit, which can provide the ability to analyze data on access network resources. Specifically, it can analyze various types of managed network resources from a macro perspective, including health, resource capacity, storage capacity, load, utilization, etc. Through perception analysis, the overall operating status of managed devices' resources, services, and applications can be obtained, which can determine the selection of the integrated management machine when applications and tasks are deployed.

[0069] In this application, the network perception module may further include a perception processing unit. The perception processing unit can process the relevant perception analysis results and schedule them according to the scheduling scheme. Based on the analysis results, during the scheduling process according to the scheduling scheme, it optimizes computing resources, network resources, storage resources, the utilization rate, load, and current operating status of each scheduling node, as well as the smoothness of applications and services, etc., issues policies and commands, migrates or expands applications and services whose perception health is lower than a preset health threshold, and optimizes resource data.

[0070] For example, during the process of the task scheduling unit scheduling each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource, if it is determined that the health of the corresponding computing network resource is lower than the preset health threshold, then the corresponding computing network resource will be expanded.

[0071] The intelligent decision-making module in this application can model and analyze the perception data of heterogeneous computing network resources such as general computing power, supercomputing, and intelligent computing. It can train algorithms to complete the training of multi-factor, intelligent decision-making scenarios and optimize the decision-making model. At the same time, it provides management capabilities for decision-making scenarios and algorithms.

[0072] Specifically, the intelligent decision-making module may include a scenario management unit, which can uniformly manage decision-making scenarios, classify various business scenarios, and provide a business scenario classification table, such as application deployment, application migration, and task scheduling. Based on the operational characteristics of different business scenarios and combined with the pre-built artificial intelligence algorithms in the algorithm management unit, corresponding computing network resource orchestration strategies are formed for the computing network orchestration module to call.

[0073] The intelligent decision-making module may also include a model training unit, which can provide the submission and training of model training tasks, and supports the creation and submission of training tasks. It also supports unified management of training tasks, including querying, statistics, recent trends, and analysis of training duration.

[0074] The intelligent decision-making module may also include an algorithm management unit, which can uniformly manage the artificial intelligence algorithm library. AI (artificial intelligence) capabilities and various types of algorithms are categorized for different business scenarios and can be operated according to those scenarios. All AI capabilities and algorithms are extracted to form an AI capability registry for the scenario management unit to call; a general-purpose algorithm library is formed, enabling efficient reuse for common scenarios.

[0075] This application embodiment utilizes an integrated computing network resource scheduling system comprising a computing network orchestration module, a computing network scheduling module, a computing network perception module, and an intelligent decision-making module. It can interface with one or more computing network resources from the managed general computing power, supercomputing, and intelligent computing networks, and collect computing network resource data. Based on pre-built artificial intelligence algorithms, it provides computing network resource orchestration strategies for various business scenarios. Furthermore, it can acquire business requirements, determine a target computing network resource orchestration strategy from various computing network resource orchestration strategies based on these requirements, and perform computing network resource scheduling and orchestration based on the target computing network resource orchestration strategy and business requirements to obtain a scheduling scheme. This scheduling scheme can then be executed to achieve resource scheduling of the computing network. By providing pre-defined network resource orchestration strategies for various business scenarios through artificial intelligence algorithms, network resource orchestration strategies can be quickly determined based on business requirements. Then, based on the determined network resource orchestration strategies and business requirements, network resource orchestration can be performed on heterogeneous network resource data across domains and vendors, such as general computing power, supercomputing, and intelligent computing. This enables rapid business response and intelligent and precise scheduling of heterogeneous network resources across domains and vendors, thereby improving the scheduling efficiency of network resources.

[0076] The integrated computing and network resource scheduling system in this application connects the underlying computing resources and the operational status of upper-level applications, promoting the interconnection and scheduling of various platforms. It also incorporates artificial intelligence strategies to realize AI-driven computing resources, meeting the needs of cross-regional and cross-industry data resource fusion computing. Furthermore, it focuses on continuously deepening the multi-level and diversified resource and capability layout, strengthening the deep integration of various elements centered on the computing network, and continuously optimizing and providing flexible and universal scheduling services, creating an integrated computing resource scheduling service system covering the entire process, all stages, and all fields.

[0077] This application enables real-time cross-domain and cross-vendor scheduling of managed general-purpose computing power, intelligent computing, and supercomputing; it integrates multiple underlying scheduling elements and previous business elements to achieve comprehensive and all-round scheduling; and it incorporates technologies such as reinforcement learning and big data to empower intelligent scheduling. Specifically, it can: 1) effectively improve the utilization rate of various computing resources and achieve balanced use of various resources; 2) improve the final scheduling accuracy through multi-element fusion and other technologies; and 3) greatly improve the overall scheduling efficiency by assisting artificial intelligence technology.

[0078] Figure 3 This is a flowchart illustrating the integrated computing and network resource scheduling method provided in an embodiment of this application. (Refer to...) Figure 3 This application provides a method for integrated computing and network resource scheduling, which may include:

[0079] Step 100: Obtain business requirements and collect computing network resource data from the managed computing network resources; computing network resources include one or more of general computing power, supercomputing, and intelligent computing.

[0080] Step 200: Obtain the computing network resource orchestration strategy for each business scenario based on the pre-built artificial intelligence algorithm;

[0081] Step 300: Determine the target computing network resource orchestration strategy from the various computing network resource orchestration strategies based on business requirements;

[0082] Step 400: Based on the target computing network resource orchestration strategy and business requirements, perform computing network resource scheduling and orchestration on the computing network resource data to obtain a scheduling scheme;

[0083] Step 500: Execute the scheduling plan.

[0084] In one embodiment, obtaining business requirements includes:

[0085] Step 101: Obtain the scheduling request, perform intent recognition on the scheduling request, and obtain the business requirements corresponding to the scheduling request.

[0086] In one embodiment, the scheduling scheme is executed, including

[0087] Step 501: Decompose the scheduling scheme to obtain a corresponding number of subtasks;

[0088] Step 502: Schedule each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource.

[0089] In one embodiment, when scheduling each subtask to be executed on the corresponding computing network resource through the interface of the managed computing network resource, the method further includes:

[0090] Step 5021: If it is determined that the health of the corresponding computing network resource is lower than the preset health threshold, then the corresponding computing network resource is expanded.

[0091] It should be noted that the execution entity of the integrated computing and network resource scheduling method provided in this application embodiment can be the aforementioned integrated computing and network resource scheduling system.

[0092] In this application, users can input information through the integrated computing and network resource scheduling system and initiate scheduling requests based on the input information.

[0093] The computing network orchestration module in the integrated computing and network resource scheduling system can parse the received scheduling requests to form business requirements for computing power and network.

[0094] Furthermore, the computing network orchestration module can identify the business scenario in which the business requirement is located, and obtain the corresponding computing network resource orchestration strategy from the various computing network resource orchestration strategies provided by the intelligent decision-making module as the target computing network resource orchestration strategy based on the business scenario.

[0095] Furthermore, the computing network orchestration module can design and orchestrate a scheduling scheme for the computing network resource data collected by the computing network perception module according to business needs and the target computing network resource orchestration strategy, thereby forming a scheduling scheme for the business needs, and then sending the scheduling scheme to the computing network scheduling module.

[0096] After receiving the scheduling plan, the computing network scheduling module can execute the scheduling plan to complete the scheduling of computing network resources according to business needs.

[0097] It should be noted that, Figure 2 This is a schematic diagram illustrating a scenario of the integrated computing and network resource scheduling system provided in an embodiment of this application. (Refer to...) Figure 2 The network orchestration module of this application may include an intent analysis unit, which can obtain the scheduling request initiated by the user and perform intent recognition on the input information in the scheduling request; specifically, it can perform resource intent analysis and business intent analysis on the input information, and obtain the business requirements of the scheduling request after completing the analysis.

[0098] Resource intent analysis can include a comprehensive assessment of computing power, network, storage, etc., while business intent analysis analyzes business needs based on the type of application and task, as well as the industry and business tags of the customer.

[0099] The network orchestration module of this application may further include a capability policy management unit. This unit can provide unified management of orchestration capabilities and maintain the orchestration capability service catalog. It can also design and manage orchestration strategies, which are used for branching, looping, retries, and capability filtering during orchestration.

[0100] The computational network orchestration module of this application may also include a design center unit. This design center unit uses a visual approach to design and orchestrate execution templates using schemes and strategies, providing functions such as querying a scheme library and strategies, drag-and-drop functionality, and attribute configuration, allowing users to customize orchestration strategies. Simultaneously, it can also adjust the computational network resource orchestration strategies obtained from the intelligent decision-making module, supporting adjustments to the execution order of capabilities via a graphical interface using drag-and-drop methods.

[0101] The computing network scheduling module in this application can analyze the scheduling scheme to generate various sub-tasks, schedule each sub-task to the corresponding computing resources by calling the open interface of the managed computing network resources, and monitor and manage the execution process of the scheduling.

[0102] Specifically, the network scheduling module may include a scheduling strategy unit, which can provide design and management of scheduling strategies. The scheduling strategy is used to handle the initiation, monitoring, repetition, exception and error handling during scheduling execution, and mainly includes rule strategies, trigger strategies, task initiation strategies, repetition mechanisms, exception and error handling, etc.

[0103] The network scheduling module may also include a scheduling execution unit, which can initiate scheduling tasks according to the scheduling scheme, covering the entire lifecycle management of scheduling, including task management, task splitting, task scheduling, and task scheduling control.

[0104] Furthermore, the scheduling execution unit may include a task splitting unit and a task scheduling unit.

[0105] The task splitting unit can split the scheduling scheme into a corresponding number of subtasks;

[0106] The task scheduling unit can further schedule each subtask to the corresponding computing network resources for execution through the interface of the managed computing network resources.

[0107] The network scheduling module may also include a scheduling monitoring unit, which provides the ability to monitor ongoing scheduling tasks in real time. It can view the execution process, execution progress, node execution status, node execution start time, execution end time, duration, input data, output data, conversion rules, execution results, etc.

[0108] The computing network perception module in this application can connect to the unified open interface of the managed computing network resources, collect various perception data, including events and performance of objects such as computing power, network, and applications, and analyze the perception data to discover targets for further optimization of scheduling.

[0109] The computing network perception module can include a perception access unit. This unit can access and manage interfaces, collect computing network resources, and provide unified access to various data sources (such as resources, performance data, alarms, events, logs, business data, applications, and services). It interfaces with managed applications and task statuses, allowing for task management and monitoring to track task execution, retry or repeat execution of abnormal and erroneous tasks according to scheduling rules, and providing unified processing of various data, such as data cleaning and data storage.

[0110] The network perception module may also include a perception analysis unit, which can provide the ability to analyze data on access network resources. Specifically, it can analyze various types of managed network resources from a macro perspective, including health, resource capacity, storage capacity, load, utilization, etc. Through perception analysis, the overall operating status of managed devices' resources, services, and applications can be obtained, which can determine the selection of the integrated management machine when applications and tasks are deployed.

[0111] In this application, the network perception module may further include a perception processing unit. The perception processing unit can process the relevant perception analysis results and schedule them according to the scheduling scheme. Based on the analysis results, during the scheduling process according to the scheduling scheme, it optimizes computing resources, network resources, storage resources, the utilization rate, load, and current operating status of each scheduling node, as well as the smoothness of applications and services, etc., issues policies and commands, migrates or expands applications and services whose perception health is lower than a preset health threshold, and optimizes resource data.

[0112] For example, during the process of the task scheduling unit scheduling each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource, if it is determined that the health of the corresponding computing network resource is lower than the preset health threshold, then the corresponding computing network resource will be expanded.

[0113] The intelligent decision-making module in this application can model and analyze the perception data of heterogeneous computing network resources such as general computing power, supercomputing, and intelligent computing. It can train algorithms to complete the training of multi-factor, intelligent decision-making scenarios and optimize the decision-making model. At the same time, it provides management capabilities for decision-making scenarios and algorithms.

[0114] Specifically, the intelligent decision-making module may include a scenario management unit, which can uniformly manage decision-making scenarios, classify various business scenarios, and provide a business scenario classification table, such as application deployment, application migration, and task scheduling. Based on the operational characteristics of different business scenarios and combined with the pre-built artificial intelligence algorithms in the algorithm management unit, corresponding computing network resource orchestration strategies are formed for the computing network orchestration module to call.

[0115] The intelligent decision-making module may also include a model training unit, which can provide the submission and training of model training tasks, and supports the creation and submission of training tasks. It also supports unified management of training tasks, including querying, statistics, recent trends, and analysis of training duration.

[0116] The intelligent decision-making module may also include an algorithm management unit, which can uniformly manage the artificial intelligence algorithm library. AI (artificial intelligence) capabilities and various types of algorithms are categorized for different business scenarios and can be operated according to those scenarios. All AI capabilities and algorithms are extracted to form an AI capability registry for the scenario management unit to call; a general-purpose algorithm library is formed, enabling efficient reuse for common scenarios.

[0117] The integrated computing network resource scheduling method provided in this application embodiment, through an integrated computing network resource scheduling system including a computing network orchestration module, a computing network scheduling module, a computing network perception module, and an intelligent decision-making module, can connect to interfaces of one or more computing network resources among the managed general computing power, supercomputing, and intelligent computing, and collect computing network resource data; based on pre-built artificial intelligence algorithms, it provides computing network resource orchestration strategies for various business scenarios; then it can obtain business requirements, determine the target computing network resource orchestration strategy from various computing network resource orchestration strategies according to the business requirements, and perform computing network resource scheduling and orchestration on the computing network resource data according to the target computing network resource orchestration strategy and business requirements to obtain a scheduling scheme; thereby, the scheduling scheme can be executed to realize the resource scheduling of the computing power network. By providing pre-defined network resource orchestration strategies for various business scenarios through artificial intelligence algorithms, network resource orchestration strategies can be quickly determined based on business requirements. Then, based on the determined network resource orchestration strategies and business requirements, network resource orchestration can be performed on heterogeneous network resource data across domains and vendors, such as general computing power, supercomputing, and intelligent computing. This enables rapid business response and intelligent and precise scheduling of heterogeneous network resources across domains and vendors, thereby improving the scheduling efficiency of network resources.

[0118] The integrated computing and network resource scheduling system in this application connects the underlying computing resources and the operational status of upper-level applications, promoting the interconnection and scheduling of various platforms. It also incorporates artificial intelligence strategies to realize AI-driven computing resources, meeting the needs of cross-regional and cross-industry data resource fusion computing. Furthermore, it focuses on continuously deepening the multi-level and diversified resource and capability layout, strengthening the deep integration of various elements centered on the computing network, and continuously optimizing and providing flexible and universal scheduling services, creating an integrated computing resource scheduling service system covering the entire process, all stages, and all fields.

[0119] This application enables real-time cross-domain and cross-vendor scheduling of managed general-purpose computing power, intelligent computing, and supercomputing; it integrates multiple underlying scheduling elements and previous business elements to achieve comprehensive and all-round scheduling; and it incorporates technologies such as reinforcement learning and big data to empower intelligent scheduling. Specifically, it can: 1) effectively improve the utilization rate of various computing resources and achieve balanced use of various resources; 2) improve the final scheduling accuracy through multi-element fusion and other technologies; and 3) greatly improve the overall scheduling efficiency by assisting artificial intelligence technology.

[0120] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call a computer program in the memory 430 to execute the steps of the integrated computer network resource scheduling method, such as including:

[0121] Obtain business requirements and collect computing network resource data from the managed computing network resources; the computing network resources include one or more of general computing power, supercomputing, and intelligent computing.

[0122] Obtain computing network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms;

[0123] Based on the aforementioned business requirements, determine the target computing network resource orchestration strategy from the various computing network resource orchestration strategies;

[0124] Based on the target computing network resource orchestration strategy and the business requirements, the computing network resource data is scheduled and orchestrated to obtain a scheduling scheme.

[0125] Execute the aforementioned scheduling scheme.

[0126] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0127] On the other hand, embodiments of this application also provide a storage medium, which is a computer-readable storage medium storing a computer program. The computer program is used to cause a processor to execute the steps of the methods provided in the above embodiments, including, for example:

[0128] Obtain business requirements and collect computing network resource data from the managed computing network resources; the computing network resources include one or more of general computing power, supercomputing, and intelligent computing.

[0129] Obtain computing network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms;

[0130] Based on the aforementioned business requirements, determine the target computing network resource orchestration strategy from the various computing network resource orchestration strategies;

[0131] Based on the target computing network resource orchestration strategy and the business requirements, the computing network resource data is scheduled and orchestrated to obtain a scheduling scheme.

[0132] Execute the aforementioned scheduling scheme.

[0133] The computer-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical storage (e.g., CD, DVD, BD, HVD), and semiconductor storage (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0134] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0135] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A computing-network integrated resource scheduling system, characterized in that, It includes a computing network orchestration module, a computing network scheduling module, a computing network perception module, and an intelligent decision-making module; The computing network sensing module is used to connect to the interface of the managed computing network resources and collect computing network resource data; the computing network resources include one or more of general computing power, supercomputing and intelligent computing; The intelligent decision-making module is used to provide computing network resource orchestration strategies for various business scenarios based on pre-set artificial intelligence algorithms for the computing network orchestration module to call. The computing network orchestration module is used to acquire business requirements, determine a target computing network resource orchestration strategy from various computing network resource orchestration strategies based on the business requirements, and perform computing network resource scheduling and orchestration on the computing network resource data based on the target computing network resource orchestration strategy and the business requirements to obtain a scheduling scheme. The computing network scheduling module is used to execute the scheduling scheme; The computing network scheduling module includes a task splitting unit and a task scheduling unit; The task splitting unit is used to split the scheduling scheme to obtain a corresponding number of subtasks; The task scheduling unit is used to schedule each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource; The computing network sensing module includes a sensing processing unit; The perception processing unit is used to perform capacity expansion processing on the corresponding computing network resources if it is determined that the health of the corresponding computing network resources is lower than a preset health threshold during the process of the task scheduling unit scheduling each subtask to the corresponding computing network resources for execution through the interface of the managed computing network resources.

2. The integrated computing and network resource scheduling system according to claim 1, characterized in that, The network orchestration module includes an intent analysis unit, which is used for: Obtain a scheduling request, identify the intent of the scheduling request, and obtain the business requirements corresponding to the scheduling request.

3. A method for integrated computing and network resource scheduling, characterized in that, include: Obtain business requirements and collect computing network resource data from the managed computing network resources; The computing network resources include one or more of general computing power, supercomputing, and intelligent computing. Obtain computing network resource orchestration strategies for various business scenarios based on pre-built artificial intelligence algorithms; Based on the aforementioned business requirements, determine the target computing network resource orchestration strategy from the various computing network resource orchestration strategies; Based on the target computing network resource orchestration strategy and the business requirements, the computing network resource data is scheduled and orchestrated to obtain a scheduling scheme. Execute the aforementioned scheduling scheme; The scheduling scheme is broken down into a corresponding number of subtasks; Each subtask is scheduled to be executed on the corresponding computing network resource through the interface of the managed computing network resource; When scheduling each subtask to the corresponding computing network resource for execution through the interface of the managed computing network resource, it also includes: If it is determined that the health of the corresponding computing network resource is lower than a preset health threshold, then the corresponding computing network resource will be expanded.

4. The integrated computing and network resource scheduling method according to claim 3, characterized in that, The acquisition of business requirements includes: Obtain a scheduling request, identify the intent of the scheduling request, and obtain the business requirements corresponding to the scheduling request.

5. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the integrated computer network resource scheduling method according to any one of claims 3 to 4.

6. A storage medium, said storage medium being a computer-readable storage medium, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the integrated computer network resource scheduling method according to any one of claims 3 to 4.

Images