Multi-agent computing power scheduling method and device for intelligent computing center cloud platform

Abstract

The invention provides a multi-agent computing power scheduling method and device for an intelligent computing center cloud platform, and relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructures, and the method comprises the steps: S1, operating task management agents through the computing power of the intelligent computing center cloud platform, obtaining computing power resource demand information of a first computing power operation task in the computing power operation task scheduling queue; s2, acquiring computing power resource use information of a plurality of acceleration cards through a computing power resource monitoring agent of the intelligent computing center cloud platform; and S3, through a computing power scheduling agent of the intelligent computing center cloud platform, based on the computing power resource demand information and the computing power resource use information, a first accelerator card is allocated to process a first computing power operation task, and the computing power resource use information of the first accelerator card is matched with the computing power resource demand information of the first computing power operation task. According to the method, the computing power resource utilization rate can be greatly improved, so that the computing power resources are universally applied.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent computing centers, intelligent computing centers and computing power infrastructure, and particularly relates to a method and device for multi-agent scheduling of computing power in a cloud platform of an intelligent computing center. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have emerged as the times require.

[0003] An "intelligent computing center" refers to a facility that provides the required computing power, data and algorithms mainly for artificial intelligence applications (such as scenarios of artificial intelligence deep learning model development, model training and model inference, etc.) by using large-scale heterogeneous computing power resources, including general computing power and intelligent computing power. The intelligent computing center covers facilities, hardware and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0004] The "intelligent computing center" includes but is not limited to the "intelligent computing center".

[0005] An "intelligent computing center", that is, an artificial intelligence computing center, is a type of computing power infrastructure that provides computing power services, data services and algorithm services required for artificial intelligence applications based on artificial intelligence theory and adopting an artificial intelligence computing architecture.

[0006] "Computing power" is the core of "intelligent computing centers", "cloud platforms of intelligent computing centers" and "intelligent computing centers". It is the ability of computer devices or computing / data centers to process parameters, the ability of computer hardware and software to cooperate to execute a certain computing requirement together, the computing ability to achieve the output of target results by processing parameter data, and a new type of productive force integrating parameter computing power, network carrying capacity and data storage capacity. It mainly provides services to the society through computing power infrastructure.

[0007] In an intelligent computing center, computing power resources are usually provided by acceleration cards to execute computing power operation tasks. In the prior art, a pre-written scheduling program is used to obtain the computing power operation tasks to be scheduled and the status of current acceleration cards, and then the scheduling program is used to allocate acceleration cards for the computing power operation tasks. However, in the prior art, the pre-written scheduling program can only simply allocate acceleration cards according to preset rules, and cannot flexibly adapt to different situations. There is a situation where different computing power operation tasks and the status of different acceleration cards cannot be taken into account, resulting in a situation where the computing power resources of the acceleration cards are surplus after allocation, and the utilization rate of the computing power resources of the intelligent computing center is very low. At the same time, due to the surplus of the computing power resources provided by the intelligent computing center, users also need to bear this part of the computing power resources when renting computing power services, resulting in a high rental cost and making it difficult to achieve the wide application of inclusive computing power.

[0008] It can be seen that in the prior art, there are problems of very low utilization rate of computing power resources in the intelligent computing center and very high computing power leasing cost. Summary of the Invention

[0009] Embodiments of the present invention provide a method and device for multi-agent scheduling of computing power in a cloud platform of an intelligent computing center to solve the problems of very low utilization rate of computing power resources in the intelligent computing center and very high computing power leasing cost in the prior art.

[0010] To solve the above problems, the present invention is implemented as follows: In a first aspect, embodiments of the present invention provide a method for multi-agent scheduling of computing power in a cloud platform of an intelligent computing center, including: Step S1: Obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue through the computing power operation task management agent of the intelligent computing center cloud platform; Step S2: Obtain the computing power resource usage information of multiple acceleration cards through the computing power resource monitoring agent of the intelligent computing center cloud platform; Step S3: Through the computing power scheduling agent of the intelligent computing center cloud platform, allocate a first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information, where the first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power operation task.

[0011] In one embodiment, the computing power operation task management agent, the computing power resource monitoring agent, and the computing power scheduling agent are agents in a first virtual space, and the first virtual space is used to schedule the multiple acceleration cards to process the computing power operation tasks included in the computing power operation task scheduling queue; The step S3 includes: Step S31: Broadcast the computing power resource requirement information in the first virtual space through the computing power operation task management agent; Step S32: Broadcast the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power resource monitoring agent; Step S33: Receive the computing power resource requirement information and the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power scheduling agent; Step S34: Allocate a first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent.

[0012] In one embodiment, the step S3 includes: Step S31': Run the task management agent for the computing power, and broadcast the first information, where the first information includes a first identifier and the computing power resource requirement information, and the first identifier is used to represent the computing power operation task scheduled for the multiple acceleration cards to process; Step S32': Through the computing power resource monitoring agent, broadcast the second information, where the second information includes the first identifier and the computing power resource usage information of the multiple acceleration cards; Step S33': Through the computing power scheduling agent, receive the first information and the second information based on the first identifier; Step S34': Through the computing power scheduling agent, allocate the first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information.

[0013] In one embodiment, the step S1 includes: Step S11: Through the task management agent for the computing power, obtain the task priority of each computing power operation task in at least one computing power operation task included in the computing power operation task scheduling queue; Step S12: Through the task management agent for the computing power, determine the first computing power operation task from the at least one computing power operation task based on the task priority; Step S13: Through the task management agent for the computing power, obtain the computing power resource requirement information of the first computing power operation task.

[0014] In one embodiment, the step S11 includes: Step S111: Through the task management agent for the computing power, obtain the task priority of each computing power operation task in at least one computing power operation task included in the computing power operation task scheduling queue based on the first model context protocol (MCP) interface; The step S13 includes: Step S131: Through the task management agent for the computing power, obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue based on the first MCP interface.

[0015] In one embodiment, the step S2 includes: Step S21: Through the computing power resource monitoring agent, obtain the computing power resource usage information of multiple acceleration cards from the computing power resource collector based on the second MCP interface, where the computing power resource collector is a component deployed in the intelligent computing center cloud platform for collecting the computing power resource usage information of the multiple acceleration cards; The step S3 includes: Step S31'', determine the first acceleration card based on the computing power resource demand information and the computing power resource usage information through the computing power scheduling agent; Step S32'', allocate the first acceleration card through the computing power scheduling agent to process the first computing power operation task based on the third MCP interface.

[0016] In a second aspect, an embodiment of the present invention further provides a multi-agent scheduling computing power device for an intelligent computing center cloud platform, including: A first acquisition module, configured to obtain the computing power resource demand information of the first computing power operation task in the computing power operation task scheduling queue through the computing power operation task management agent of the intelligent computing center cloud platform; A second acquisition module, configured to obtain the computing power resource usage information of multiple acceleration cards through the computing power resource monitoring agent of the intelligent computing center cloud platform; A scheduling module, configured to allocate the first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information through the computing power scheduling agent of the intelligent computing center cloud platform, where the first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource demand information of the first computing power operation task.

[0017] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps in the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0019] In a fifth aspect, the present invention further provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the steps in the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0020] In the present invention, in step S1, the computing power operation task management agent is run through the computing power of the intelligent computing center cloud platform to obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue; in step S2, the computing power resource monitoring agent of the intelligent computing center cloud platform is used to obtain the computing power resource usage information of multiple acceleration cards; in step S3, the computing power scheduling agent of the intelligent computing center cloud platform allocates the first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information, where the first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power operation task. In this way, by using the computing power operation task management agent to obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue, the management of the computing power operation tasks in the computing power operation task scheduling queue can be realized; by using the computing power resource monitoring agent to obtain the computing power resource usage information of multiple acceleration cards, the monitoring of the computing power resource usage information of different acceleration cards can be realized; by using the computing power scheduling agent to allocate the first acceleration card to execute the first computing power task, the management of the computing power operation tasks, the monitoring of the computing power resource usage information of the acceleration cards, and the allocation of the acceleration cards are taken into account by different agents. Compared with the pre-written scheduling program, it can more flexibly adapt to different situations, take into account the states of different computing power operation tasks and different acceleration cards, and greatly improve the utilization rate of the computing power resources of the intelligent computing center. At the same time, when users lease computing power services, they no longer need to bear the excess computing power resources, greatly reducing the leasing cost and realizing the wide application of inclusive computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0022] Figure 1 is a flowchart of a method for multi-agent scheduling of computing power in an intelligent computing center cloud platform provided by an embodiment of the present invention; Figure 2 is an interaction diagram between agents provided by an embodiment of the present invention; Figure 3 is a structural diagram of a device for multi-agent scheduling of computing power in an intelligent computing center cloud platform provided by an embodiment of the present invention; Figure 4 is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] The "computing power" described in the present invention refers to: the ability of a computer device or a computing / data center to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to achieve the output of a target result by processing information data, a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly providing services to society through computing power infrastructure.

[0025] The "computational power" (Computational Power, CP) described in the present invention refers to: the ability of a data center server to process data and achieve result output, a comprehensive index for measuring the computing ability of a data center, including general computing ability, supercomputing ability, and intelligent computing ability. The commonly used measurement unit is the number of floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), and the larger the value, the stronger the comprehensive computing ability. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 + CP 智能 + CP 超级 .

[0026] The "carrying capacity" (Network Power, NP) described in the present invention refers to: the manifestation of the data transmission ability of computing power facilities, a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., involving network transmission within and between data centers, and a comprehensive index for measuring network transmission scheduling ability.

[0027] The "Storage Power" (SP) described in the present invention refers to the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon. It is a comprehensive indicator for measuring the data storage capacity of a data center, including external storage devices such as storage arrays and built-in storage devices in servers. The commonly used measurement unit for storage capacity is exabyte (EB, 1EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read and write operations per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0028] The "computing power infrastructure" described in the present invention refers to a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage power, and can realize the centralized computing, storage, transmission, and application of information.

[0029] The "new type of information infrastructure" described in the present invention mainly includes network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, and satellite Internet, computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, and supercomputing centers, and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0030] The "computing power" described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0031] The "general computing power" described in the present invention refers to the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0032] The "intelligent computing power" described in the present invention refers to a computing platform that is scaled up for various artificial intelligence innovation applications based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit), such as natural language processing and machine vision.

[0033] The "super computing power" described in the present invention mainly refers to the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system. It is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, and gene analysis.

[0034] The "Intelligent Computing Center" described in the present invention refers to a facility that uses large-scale heterogeneous computing power resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.), and mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios like artificial intelligence deep learning model development, model training, and model inference). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0035] The "Intelligent Computing Center Cloud Platform" described in the present invention refers to a cloud computing platform that comprehensively serves based on the hardware resources and software resources of the intelligent computing center.

[0036] The "Intelligent Computing Center" described in the present invention includes, but is not limited to, the "Intelligent Computing Center".

[0037] The "Intelligent Computing Center" described in the present invention, namely the artificial intelligence computing center, is a type of computing power infrastructure based on artificial intelligence theory, adopting an artificial intelligence computing architecture, and providing computing power services, data services, and algorithm services required for artificial intelligence applications.

[0038] The "Computing Power Center" described in the present invention refers to a facility mainly composed of infrastructure such as wind, fire, water, and electricity and IT software and hardware devices, and having computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0039] The "Supercomputing Center" described in the present invention, namely the supercomputing data center, is a data center based on supercomputers or large-scale computing clusters, capable of providing functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0040] The "Computing Power Resources" described in the present invention refers to technologies and facilities with information computing, transmission, storage, and application capabilities required for the development of the digital society, including, but not limited to, computing resources such as CPU and GPU, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, as well as support and guarantee resources such as wind, fire, water, and electricity.

[0041] The "Model" described in the present invention includes, but is not limited to, the "Large Language Model" and the "Multimodal Large Model".

[0042] The "Large Language Model" described in the present invention refers to a large-scale language model (LLM), which is a language model with a relatively large number of parameters, aiming to understand and generate human language, trained with a large amount of text data, and can perform a wide range of tasks including text summarization, translation, sentiment analysis, etc.

[0043] The "Multimodal Large Models" described in the present invention refer to models that jointly train multimodal information such as text, images, videos, and audio, including but not limited to multimodal large language models.

[0044] The "Agent" described in the present invention refers to an agent that can perceive the environment and take actions to achieve specific goals. It can be software, hardware, or a system, and has autonomy, adaptability, and interaction capabilities. The agent perceives changes in the environment (such as through sensors or data input), makes judgments and decisions based on the knowledge and algorithms it has learned, and then executes actions to affect the environment or achieve a predetermined goal. Agents are widely used in the field of artificial intelligence, commonly found in automated systems, robots, virtual assistants, and game characters, etc. The core lies in the ability to learn autonomously and evolve continuously to better complete tasks and adapt to complex environments.

[0045] The "computing power operation task" described in the present invention refers to a specific workload or job that is executed on computing power resources and requires a certain amount of computing power support, usually involving scenarios such as complex data processing, numerical calculations, model training, or simulation.

[0046] The "inclusive computing power" described in the present invention refers to providing appropriate and effective computing power services to all social strata and groups with computing power service needs at an affordable cost based on the requirements of equal opportunity and the principle of commercial sustainability.

[0047] Please refer to Figure 1 , Figure 1 which is a flowchart of a method for multi-agent scheduling of computing power in an intelligent computing center cloud platform provided by an embodiment of the present invention. As Figure 1 shown, it includes the following steps: Step S1: Through the computing power operation task management agent of the intelligent computing center cloud platform, obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue.

[0048] The above-mentioned computing power operation task management agent is an agent created based on the computing power resources of the intelligent computing center cloud platform and is used to obtain relevant information of the computing power operation task scheduling queue. Among them, the computing power operation task management agent is used to act as an expert proficient in task management, and through the computing power operation task management agent, it is possible to better manage the computing power operation tasks in the task management queue.

[0049] It should be noted that there are usually multiple computing power operation tasks in the computing power operation task scheduling queue. Through the computing power operation task management agent, it is possible to manage multiple computing power operation tasks based on multiple aspects such as the importance, priority, and computing power resource requirement information of different computing power operation tasks.

[0050] The above computing power resource requirement information is used to characterize the computing power resource situation required for the computing power operation task. By obtaining the computing power resource requirement information of the first computing power operation task, it is convenient for the subsequent computing power scheduling agent to allocate the first acceleration card for the first computing power operation task.

[0051] The above first acceleration card is an acceleration card of the intelligent computing center cloud platform. The acceleration cards other than the first acceleration card in the intelligent computing center cloud platform can be referred to as "other acceleration cards" or "second acceleration cards".

[0052] Step S2: Through the computing power resource monitoring agent of the intelligent computing center cloud platform, obtain the computing power resource usage information of multiple acceleration cards.

[0053] The above computing power resource monitoring agent is an agent created based on the computing power resources of the intelligent computing center cloud platform, and is used to monitor the resources of different acceleration cards in the intelligent computing center cloud platform. Among them, the computing power resource monitoring agent is used to act as an expert proficient in task monitoring, and through the computing power resource monitoring agent, it is possible to better monitor the computing power resource usage information of different acceleration cards.

[0054] The above computing power resource usage information is used to characterize the computing power resource usage situation of the acceleration card. It should be noted that in the intelligent computing center cloud platform, usually different acceleration cards (such as GPU acceleration cards) provide computing power resources. At the same time, the intelligent computing center cloud platform provides computing power resources for different computing power operation tasks to execute the computing power operation tasks. In this case, the computing power resources of the acceleration card are in a changing state (such as not using computing power resources, using some of the computing power resources of the acceleration card, and using all of the computing power resources of the acceleration card). By monitoring the computing power resource usage information of different acceleration cards through the computing power resource monitoring agent, it is convenient for the subsequent computing power scheduling agent to allocate the first acceleration card for the first computing power operation task.

[0055] Step S3: Through the computing power scheduling agent of the intelligent computing center cloud platform, allocate the first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information. The first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power operation task.

[0056] The above computing power scheduling agent is an agent created for the computing power resources of the intelligent computing center cloud platform, and is used to schedule the computing power resources of the intelligent computing center cloud platform to allocate different acceleration cards for different computing power operation tasks. Among them, the computing power scheduling agent is used to act as an expert proficient in task scheduling, and can allocate better acceleration cards for the computing power operation tasks, so that the computing power resources of each acceleration card can be fully utilized to improve the utilization rate of the computing power resources.

[0057] In the present invention, in step S1, the computing power operation task management agent of the intelligent computing center cloud platform is used to obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue; in step S2, the computing power resource monitoring agent of the intelligent computing center cloud platform is used to obtain the computing power resource usage information of multiple acceleration cards; in step S3, the computing power scheduling agent of the intelligent computing center cloud platform is used to allocate a first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information, where the first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power operation task. In this way, by using the computing power operation task management agent to obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue, the management of the computing power operation tasks in the computing power operation task scheduling queue can be realized; by using the computing power resource monitoring agent to obtain the computing power resource usage information of multiple acceleration cards, the monitoring of the computing power resource usage information of different acceleration cards can be realized; by using the computing power scheduling agent to allocate the first acceleration card to execute the first computing task, thus, through different agents, the management of the computing power operation tasks, the monitoring of the computing power resource usage information of the acceleration cards, and the allocation of the acceleration cards are taken into account, which can be more flexible in adapting to different situations compared with the pre-written scheduling program, taking into account the states of different computing power operation tasks and different acceleration cards, and greatly improving the utilization rate of the computing power resources of the intelligent computing center. At the same time, users no longer need to bear the excess computing power resources when renting computing power services, greatly reducing the rental cost and realizing the wide application of inclusive computing power.

[0058] Further, as Figure 2 shown, the information interaction between the computing power operation task management agent, the computing power resource monitoring agent, and the computing power scheduling agent is realized in the form of broadcasting, so that after the computing power operation task management agent obtains the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue, and the computing power resource monitoring agent obtains the computing power resource usage information of multiple acceleration cards, the computing power scheduling agent can obtain the computing power resource requirement information of the first computing power operation task and the computing power resource usage information of multiple acceleration cards through broadcasting, and then allocate the first acceleration card to process the first computing power operation task.

[0059] It should be noted that information interaction through broadcasting will cause other irrelevant intelligent agents to also receive the computing power resource demand information and the computing power resource usage information of multiple acceleration cards. In order to enable the computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent to accurately perform information interaction, two methods are provided in the present invention. One is to set the computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent in a virtual space. The other is to determine the information interaction between the computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent through identification.

[0060] Specifically, in one embodiment, the computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent are intelligent agents in a first virtual space, and the first virtual space is used to schedule the multiple acceleration cards to process the computing power operation tasks included in the computing power operation task scheduling queue; The step S3 includes: Step S31, through the computing power operation task management intelligent agent, broadcast the computing power resource demand information in the first virtual space; Step S32, through the computing power resource monitoring intelligent agent, broadcast the computing power resource usage information of the multiple acceleration cards in the first virtual space; Step S33, through the computing power scheduling intelligent agent, receive the computing power resource demand information and the computing power resource usage information of the multiple acceleration cards in the first virtual space; Step S34, through the computing power scheduling intelligent agent, allocate the first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information.

[0061] The above-mentioned first virtual space can be as Figure 2 shown. The computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent are in the same first virtual space. The data broadcast by the intelligent agents in the first virtual space can only be received by other intelligent agents in the first virtual space, thus realizing stable information interaction between the intelligent agents. In this way, the computing power operation task management intelligent agent, the computing power resource monitoring intelligent agent, and the computing power scheduling intelligent agent are configured in the first virtual space to obtain the computing power resource demand information and the computing power resource usage information of multiple acceleration cards, and to realize the allocation of the first acceleration card by the computing power scheduling intelligent agent to process the first computing power operation task.

[0062] Specifically, the computing power operation task management agent is used to broadcast the computing power resource demand information in the first virtual space; the computing power resource monitoring agent is used to broadcast the computing power resource usage information of the multiple acceleration cards in the first virtual space; the computing power scheduling agent is used to receive the computing power resource demand information and the computing power resource usage information of the multiple acceleration cards in the first virtual space. In this way, information interaction is realized through the first virtual space.

[0063] In addition, in one embodiment, step S3 includes: Step S31': The computing power operation task management agent broadcasts a first piece of information, where the first piece of information includes a first identifier and the computing power resource demand information, and the first identifier is used to represent the computing power operation task scheduled to be processed by the multiple acceleration cards; Step S32': The computing power resource monitoring agent broadcasts a second piece of information, where the second piece of information includes the first identifier and the computing power resource usage information of the multiple acceleration cards; Step S33': The computing power scheduling agent receives the first piece of information and the second piece of information based on the first identifier; Step S34': The computing power scheduling agent allocates a first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information.

[0064] Different from using the first virtual space, in the present invention, information interaction between the computing power operation task management agent, the computing power resource monitoring agent, and the computing power scheduling agent is realized through the first identifier. Specifically, the computing power operation task management agent broadcasts a first piece of information, which includes a first identifier and the computing power resource demand information, and the first identifier is used to represent the computing power operation task scheduled to be processed by the multiple acceleration cards; the computing power resource monitoring agent broadcasts a second piece of information, which includes the first identifier and the computing power resource usage information of the multiple acceleration cards; the computing power scheduling agent receives the first piece of information and the second piece of information based on the first identifier. In this way, different computing power scheduling agents determine whether the received first piece of information and second piece of information are the required information through the first identifier, and then determine the computing power resource demand information and the computing power resource usage information based on the first piece of information and the second piece of information including the first identifier, so as to realize the allocation of the first acceleration card by the computing power scheduling agent to process the first computing power operation task.

[0065] In one embodiment, step S1 includes: Step S11: The computing power operation task management agent obtains the task priority of each computing power operation task included in the computing power operation task scheduling queue; Step S12: Run the task management agent through the computing power, and determine the first computing power operation task from the at least one computing power operation task based on the task priority; Step S13: Through the computing power operation task management agent, obtain the computing power resource requirement information of the first computing power operation task.

[0066] It should be noted that when managing the computing power operation tasks in the computing power operation task scheduling queue, the importance or priority of the computing power operation tasks is different, and it is necessary to determine the computing power tasks to be executed by the computing power resources to be scheduled in sequence according to the importance or priority of each computing power operation task.

[0067] Specifically, in the present invention, through the computing power operation task management agent, obtain the task priority of each computing power operation task in the at least one computing power operation task included in the computing power operation task scheduling queue; through the computing power operation task management agent, determine the first computing power operation task from the at least one computing power operation task based on the task priority; through the computing power operation task management agent, obtain the computing power resource requirement information of the first computing power operation task. In this way, obtain the task priority and computing power resource requirement information of the computing power operation task through the computing power operation task scheduling queue, and determine the order of scheduling the computing power resources to process the computing power operation task according to the task priority, so as to determine the first computing power operation task.

[0068] In some embodiments, the above step S1 can also obtain the computing power resource requirement information of the first computing power operation task through the following steps. Specifically, step S1 includes: Step S11': Through the computing power operation task management agent, obtain the task priority and computing power resource requirement information of each computing power operation task in the at least one computing power operation task included in the computing power operation task scheduling queue.

[0069] Step S12': Through the computing power operation task management agent, determine the first computing power operation task from the at least one computing power operation task based on the task priority and the computing power resource requirement information.

[0070] In this way, determine the computing power operation task with the highest priority for scheduling in the queue by combining the task priority and the computing power resource requirement information.

[0071] In some embodiments, step S1 may further include: Step S11': Through the computing power operation task management agent, obtain the task priority, computing power resource requirement information, and time information of each computing power operation task in the at least one computing power operation task included in the computing power operation task scheduling queue, where the time information is used to represent the time when each computing power operation task is added to the task scheduling.

[0072] Step S12': Run the task management agent through the computing power, and determine the first computing power running task from the at least one computing power running task based on the task priority, the computing power resource requirement information, and the time learning.

[0073] In this way, by combining the task priority, the computing power resource requirement information, and the time information, the computing power running task with the highest priority for scheduling in the queue is determined.

[0074] In one embodiment, the step S11 includes: Step S111: Through the computing power running task management agent, obtain the task priority of each computing power running task in the at least one computing power running task included in the computing power running task scheduling queue based on the first Model Context Protocol (MCP) interface. The step S13 includes: Step S131: Through the computing power running task management agent, obtain the computing power resource requirement information of the first computing power running task in the computing power running task scheduling queue based on the first MCP interface.

[0075] It should be noted that the protocols for the computing power running task scheduling queue to interact with the computing power running task management agent may be different, and smooth data interaction cannot be achieved. Therefore, in the present invention, the interaction between the computing power running task scheduling queue and the computing power running task management agent is realized through the first MCP interface.

[0076] Among them, through the MCP service, the encapsulation of the application programming interface (API) of the computing power running task scheduling queue can be realized. The API of the computing power running task scheduling queue can access the data of the computing power running task scheduling queue. By encapsulating the API through the MCP service, the first MCP interface is obtained, which realizes the standardization of the interface, enabling the computing power running task management agent to obtain the computing power resource requirement information of the first computing power running task in the computing power running task scheduling queue through the first MCP interface.

[0077] In one embodiment, the step S2 includes: Step S21: Through the computing power resource monitoring agent, obtain the computing power resource usage information of multiple acceleration cards from the computing power resource collector based on the second Model Context Protocol MCP interface. The computing power resource collector is a component deployed on the intelligent computing center cloud platform for collecting the computing power resource usage information of the multiple acceleration cards.

[0078] The step S3 includes: Step S31'', determine the first acceleration card by the computing power scheduling agent based on the computing power resource demand information and the computing power resource usage information; Step S32'', allocate the first acceleration card by the computing power scheduling agent to process the first computing power operation task based on the third model context protocol MCP interface.

[0079] It should be noted that, similar to the interaction between the computing power operation task scheduling queue and the computing power operation task management agent, there is also a situation where the interaction between the computing power resource monitoring agent and the computing power resource collector cannot have smooth data interaction due to different protocols. Therefore, in the present invention, the interaction between the computing power resource monitoring agent and the computing power resource collector is realized through the second MCP interface.

[0080] Specifically, by encapsulating the API of the computing power resource collector through MCP, the API of the computing power resource collector can access the data of the computing power resource collector. By encapsulating the API through the MCP service, the second MCP interface is obtained, realizing the standardization of the interface, so that the computing power resource monitoring agent can obtain the computing power resource usage information of multiple acceleration cards of the computing power resource collector through the second MCP interface.

[0081] Similarly, the computing power scheduling agent realizes the allocation of the first acceleration card to process the first computing power operation task through the third MCP interface. Among them, the computing power scheduling agent realizes the allocation of the first acceleration card to process the first computing power operation task through the third MCP interface, specifically through the computing power resource scheduler. The computing power resource scheduler is a component deployed in the intelligent computing center cloud platform for allocating acceleration cards for computing power operation tasks. The computing power scheduling agent realizes data interaction with the computing power resource scheduler through the third MCP interface.

[0082] Specifically, in the case where it is necessary to allocate the first acceleration card to process the first computing power operation task, the computing power scheduling agent sends a scheduling instruction to the computing power resource scheduler through the third MCP interface. After receiving the scheduling instruction, the computing power resource scheduler allocates the first acceleration card to process the first computing power operation task. Among them, the scheduling instruction includes the identifier of the first acceleration card and the identifier of the first computing power operation task.

[0083] Please refer to Figure 3 , Figure 3 is the structural diagram of a device for multi-agent scheduling of computing power in an intelligent computing center cloud platform provided by an embodiment of the present invention. As Figure 3 shown, the device 300 for multi-agent scheduling of computing power in the intelligent computing center cloud platform includes: The first acquisition module 301 is used to obtain the computing power resource demand information of the first computing power operation task in the computing power operation task scheduling queue through the computing power operation task management agent of the intelligent computing center cloud platform; A second acquisition module 302, configured to obtain the computing power resource usage information of multiple acceleration cards through the computing power resource monitoring agent of the intelligent computing center cloud platform; A scheduling module 303, configured to allocate a first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information through the computing power scheduling agent of the intelligent computing center cloud platform, where the first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource demand information of the first computing power operation task.

[0084] In one embodiment, the computing power operation task management agent, the computing power resource monitoring agent, and the computing power scheduling agent are agents in a first virtual space, and the first virtual space is used to schedule the multiple acceleration cards to process the computing power operation tasks included in the computing power operation task scheduling queue; The step scheduling module 303 includes: A first broadcast unit, configured to broadcast the computing power resource demand information in the first virtual space through the computing power operation task management agent; A second broadcast unit, configured to broadcast the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power resource monitoring agent; A first receiving unit, configured to receive the computing power resource demand information and the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power scheduling agent; A first scheduling unit, configured to allocate a first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information through the computing power scheduling agent.

[0085] In one embodiment, the scheduling module 303 includes: A third broadcast unit, configured to broadcast a first message through the computing power operation task management agent, where the first message includes a first identifier and the computing power resource demand information, and the first identifier is used to represent the computing power operation task scheduled to be processed by the multiple acceleration cards; A fourth broadcast unit, configured to broadcast a second message through the computing power resource monitoring agent, where the second message includes the first identifier and the computing power resource usage information of the multiple acceleration cards; A second receiving unit, configured to receive the first message and the second message based on the first identifier through the computing power scheduling agent; A second scheduling unit, configured to allocate a first acceleration card to process the first computing power operation task based on the computing power resource demand information and the computing power resource usage information through the computing power scheduling agent.

[0086] In one embodiment, the first acquisition module 301 of the steps includes: A first acquisition unit, configured to run a task management agent through the computing power, and acquire the task priority of each computing power operation task included in the computing power operation task scheduling queue; A first determination unit, configured to determine the first computing power operation task from the at least one computing power operation task based on the task priority through the computing power operation task management agent; A second acquisition unit, configured to run a task management agent through the computing power, and acquire the computing power resource requirement information of the first computing power operation task.

[0087] In one embodiment, the first acquisition unit includes: A first acquisition subunit, configured to run a task management agent through the computing power, and acquire the task priority of each computing power operation task included in the computing power operation task scheduling queue based on a first model context protocol (MCP) interface; The second acquisition unit includes: A second acquisition subunit, configured to run a task management agent through the computing power, and acquire the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue based on the first model context protocol (MCP) interface.

[0088] In one embodiment, the second acquisition module 302 includes: A third acquisition unit, configured to run a computing power resource monitoring agent, and acquire the computing power resource usage information of multiple acceleration cards from a computing power resource collector based on a second model context protocol (MCP) interface, where the computing power resource collector is a component deployed on the intelligent computing center cloud platform for collecting the computing power resource usage information of the multiple acceleration cards.

[0089] The scheduling module 303 includes: A second determination unit, configured to determine the first acceleration card based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent; A third scheduling unit, configured to allocate the first acceleration card to process the first computing power operation task based on a third MCP interface through the computing power scheduling agent.

[0090] The device for multi-agent scheduling of computing power in the intelligent computing center cloud platform provided by the embodiments of the present invention can implement each process of the above-mentioned method for multi-agent scheduling of computing power in the intelligent computing center cloud platform. The technical features correspond one by one and can achieve the same technical effects. To avoid repetition, details are not described here again.

[0091] It should be noted that the multi-agent scheduling computing power device of the intelligent computing center cloud platform in the embodiments of the present invention can be a device, or a component, integrated circuit, or chip in an electronic device.

[0092] The present invention also provides an electronic device. Refer to Figure 4 , Figure 4 which is a schematic structural diagram of an electronic device provided in an embodiment of the present invention. The electronic device includes a memory 401, a processor 402, and a program or instruction running on the memory 401. When the program or instruction is executed by the processor 402, it can implement Figure 1 any step in the method embodiment of the multi-agent scheduling computing power of the corresponding intelligent computing center cloud platform and achieve the same beneficial effects, which will not be elaborated here.

[0093] Among them, the processor 402 can be a CPU, ASIC, FPGA, or GPU.

[0094] Those of ordinary skill in the art can understand that all or part of the steps of implementing the method embodiment of the multi-agent scheduling computing power of the above intelligent computing center cloud platform can be completed by hardware related to program instructions, and the program can be stored in a readable medium.

[0095] The present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement any step in the method embodiment of the multi-agent scheduling computing power of the above Figure 1 corresponding intelligent computing center cloud platform, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. The storage medium can be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0096] The present invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement each process in the method embodiment of the multi-agent scheduling computing power of the above Figure 1 corresponding intelligent computing center cloud platform, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0097] The terms "first", "second", etc. in the present invention are used to distinguish similar objects and do not necessarily describe a specific order or sequence. In addition, the terms "comprising", "having" and any of their variations are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. In addition, the use of "and / or" in this application means at least one of the connected objects. For example, A and / or B and / or C means including the 7 cases of A alone, B alone, C alone, A and B both present, B and C both present, A and C both present, and A, B and C all present.

[0098] It should be noted that in this document, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising such element.

[0099] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of this application, 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 is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which may be a mobile phone, computer, server, air conditioner, or a second terminal device, etc.) to execute the methods of the various embodiments of this application.

[0100] The embodiments of this application are described above in conjunction with the accompanying drawings, but this application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of this application, those of ordinary skill in the art can also make many forms without departing from the purpose of this application and the scope protected by the claims, and all of them fall within the protection scope of this application.

Claims

1. A method for multi-agent scheduling computing power in an intelligent computing center cloud platform, characterized in that, Including: Step S1: Run a task management agent through the computing power of the intelligent computing center cloud platform to obtain the computing power resource requirement information of the first computing power running task in the computing power running task scheduling queue; Step S2: Obtain the computing power resource usage information of multiple acceleration cards through the computing power resource monitoring agent of the intelligent computing center cloud platform; Step S3: Through the computing power scheduling agent of the intelligent computing center cloud platform, allocate a first acceleration card to process the first computing power running task based on the computing power resource requirement information and the computing power resource usage information. The first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power running task.

2. The method according to claim 1, wherein The task management agent for computing power operation, the computing power resource monitoring agent, and the computing power scheduling agent are agents in a first virtual space, and the first virtual space is used to schedule the multiple acceleration cards to process the computing power running tasks included in the computing power running task scheduling queue; The step S3 includes: Step S31: Broadcast the computing power resource requirement information in the first virtual space through the task management agent for computing power operation; Step S32: Broadcast the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power resource monitoring agent; Step S33: Receive the computing power resource requirement information and the computing power resource usage information of the multiple acceleration cards in the first virtual space through the computing power scheduling agent; Step S34: Allocate a first acceleration card to process the first computing power running task based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent.

3. The method according to claim 1, wherein The step S3 includes: Step S31': Broadcast a first message through the task management agent for computing power operation. The first message includes a first identifier and the computing power resource requirement information, and the first identifier is used to represent the computing power running task scheduled to be processed by the multiple acceleration cards; Step S32': Broadcast a second message through the computing power resource monitoring agent. The second message includes the first identifier and the computing power resource usage information of the multiple acceleration cards; Step S33': Receive the first message and the second message based on the first identifier through the computing power scheduling agent; Step S34': Allocate a first acceleration card to process the first computing power running task based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent.

4. The method according to any one of claims 1 to 3, characterized in that, The step S1 includes: Step S11: Obtain the task priority of each computing power running task in at least one computing power running task included in the computing power running task scheduling queue through the task management agent for computing power operation; Step S12: Determine the first computing power running task from the at least one computing power running task based on the computing power running task priority through the task management agent for computing power operation; Step S13: Obtain the computing power resource requirement information of the first computing power running task through the task management agent for computing power operation.

5. The method according to claim 4, wherein The step S11 includes: Step S111: Run the task management agent through the computing power, and obtain the task priority of each computing power operation task included in the computing power operation task scheduling queue based on the first model context protocol MCP interface; The step S13 includes: Step S131: Run the task management agent through the computing power, and obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue based on the first model context protocol MCP interface.

6. The method according to claim 1, characterized in that, The step S2 includes: Step S21: Through the computing power resource monitoring agent, obtain the computing power resource usage information of multiple acceleration cards from the computing power resource collector based on the second model context protocol MCP interface, where the computing power resource collector is a component deployed in the intelligent computing center cloud platform for collecting the computing power resource usage information of the multiple acceleration cards; The step S3 includes: Step S31'': Determine the first acceleration card based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent; Step S32'': Allocate the first acceleration card to process the first computing power operation task based on the third model context protocol MCP interface through the computing power scheduling agent.

7. An apparatus for multi-agent scheduling computing power of an intelligent computing center cloud platform, characterized in that, It includes: The first acquisition module is used to obtain the computing power resource requirement information of the first computing power operation task in the computing power operation task scheduling queue through the task management agent of the intelligent computing center cloud platform; The second acquisition module is used to obtain the computing power resource usage information of multiple acceleration cards through the computing power resource monitoring agent of the intelligent computing center cloud platform; The scheduling module is used to allocate the first acceleration card to process the first computing power operation task based on the computing power resource requirement information and the computing power resource usage information through the computing power scheduling agent of the intelligent computing center cloud platform. The first acceleration card is one of the multiple acceleration cards, and the computing power resource usage information of the first acceleration card matches the computing power resource requirement information of the first computing power operation task.

8. An electronic device, characterized in that, It includes: A processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, the steps of the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the steps of the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that, It includes computer instructions. When the computer instructions are executed by the processor, the steps of the method for multi-agent scheduling of computing power of the intelligent computing center cloud platform as described in any one of claims 1 to 6 are implemented.

Images