Heterogeneous computing power resource management method and device for intelligent computing center cloud platform

Abstract

The invention provides a heterogeneous computing power resource management method and device for an intelligent computing center cloud platform, and the method comprises the steps: S1, registering a plurality of first computing power resources corresponding to a plurality of different operators in a management platform, and obtaining a plurality of registered second computing power resources; s2, receiving a user use request input by a user on the management platform; and S3, performing matching in the plurality of second computing power resources based on the user use request to obtain at least one third computing power resource, and scheduling the at least one third computing power resource to process the computing power operation task. According to the invention, the complexity of management, operation and maintenance is greatly reduced, and the unified management and scheduling of various heterogeneous computing power resources managed by the intelligent computing center cloud platform are realized.

Description

Technical Field

[0001] This invention relates to the fields of intelligent computing centers, smart computing centers, computing infrastructure, and smart cloud technologies, specifically to a method and apparatus for managing heterogeneous computing resources in an intelligent computing center cloud platform. Background Technology

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

[0003] An "intelligent computing center" refers to a facility that provides the necessary computing power, data, and algorithms for artificial intelligence applications (such as the development, training, and inference of deep learning models) by utilizing large-scale heterogeneous computing resources, including general-purpose and intelligent computing power. Intelligent computing centers encompass facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enablement.

[0004] "Intelligent computing center" includes, but is not limited to, "intelligent computing center".

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

[0006] "Computing power" is the core of "intelligent computing center" and "smart computing center". It is the ability of computer equipment or computing / data center to process information. It is the ability of computer hardware and software to work together to perform a certain computing requirement. It is the computing power to achieve the target output by processing information data. It is a new type of productivity that integrates information computing power, network carrying capacity and data storage capacity. It mainly provides services to society through computing power infrastructure.

[0007] Since the emergence of intelligent computing centers, they have undertaken ultra-large-scale computing tasks such as model training and inference. Due to the diverse types of heterogeneous computing resources, and the fact that the same computing resource may be provided by multiple operators, management and operation complexity is extremely high. Therefore, how to uniformly manage and schedule various heterogeneous computing resources within intelligent computing centers is an urgent problem to be solved. Summary of the Invention

[0008] This invention provides a method and apparatus for managing heterogeneous computing resources in an intelligent computing center cloud platform, which solves the problem in the prior art that intelligent computing centers cannot uniformly manage and schedule multiple computing resources.

[0009] To solve the above problems, the present invention is implemented as follows:

[0010] In a first aspect, the present invention provides a method for managing heterogeneous computing resources in an intelligent computing center cloud platform, the method comprising:

[0011] Step S1: Register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources.

[0012] Step S2: Receive a user usage request input by the user on the management platform. The user usage request is used to request the management platform to call computing resources to process computing power running tasks.

[0013] Step S3: Based on the user's request, match the multiple second computing resources to obtain at least one third computing resource, and schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

[0014] Optionally, the management platform includes a first controller and a second controller, wherein the first controller is used to control the plurality of second computing resources, and the second controller is used to schedule the plurality of second computing resources.

[0015] Optionally, step S1 includes:

[0016] Step S11: Obtain the multiple first computing resources corresponding to the multiple different operators. The multiple first computing resources include at least one of the following: AI computing cluster, host, storage, network, virtual machine and containerized computing resources.

[0017] Step S12: Register the plurality of first computing resources in the first controller to obtain the plurality of registered second computing resources and a plurality of resource tags. The plurality of second computing resources and the plurality of resource tags correspond one-to-one, and the resource tags are used to indicate the resource information of the corresponding second computing resources.

[0018] Optionally, after registering the plurality of first computing resources in the first controller in step S12, and obtaining the plurality of registered second computing resources and the plurality of resource tags, the method further includes:

[0019] Step S13: Store the resources in the second controller based on the multiple resource tags;

[0020] Step S14: Based on the multiple resource tags, schedule the multiple second computing resources.

[0021] Optionally, the management platform further includes an application programming interface (API), which is used to perform unified management of the first controller and the second controller based on the API.

[0022] Optionally, step S3 includes:

[0023] Step S31: Based on the user's request, determine the configuration environment information and computing resource information corresponding to the computing power running task;

[0024] Step S32: Based on the configuration environment information and the computing resource information, match the plurality of second computing resources to obtain the at least one third computing resource;

[0025] Step S33: In the management platform, schedule at least one third computing power resource to process the computing power operation task.

[0026] Secondly, embodiments of the present invention also provide a heterogeneous computing power resource management device for an intelligent computing center cloud platform, the device comprising:

[0027] The registration module is used to register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources.

[0028] The receiving module is used to receive user usage requests input by users on the management platform, wherein the user usage requests are used to request the management platform to call computing resources to process computing power running tasks;

[0029] The matching module is used to match the multiple second computing resources based on the user's request to obtain at least one third computing resource, and to schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

[0030] Thirdly, the present invention also provides an electronic device including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, performs the steps of the method described in the first aspect above.

[0031] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect above.

[0032] Fifthly, the present invention also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the method described in the first aspect above.

[0033] This invention provides a method and apparatus for managing heterogeneous computing resources in an intelligent computing center cloud platform. The method includes: Step S1, registering multiple first computing resources corresponding to multiple different operators in a management platform to obtain multiple registered second computing resources. The management platform is used to control and schedule the multiple second computing resources, and the multiple first computing resources are different types of computing resources; Step S2, receiving a user request input by a user in the management platform, the user request being used to request the management platform to call computing resources to process computing tasks; Step S3, matching the multiple second computing resources based on the user request to obtain at least one third computing resource, and scheduling the at least one third computing resource to process the computing tasks, the third computing resource being any one of the multiple second computing resources. This invention registers multiple first computing resources belonging to different heterogeneous computing resources in a management platform, controls and schedules multiple registered second computing resources, and upon receiving a user request, matches the multiple second computing resources according to the user request to determine at least one third computing resource to process the computing task. This greatly reduces the complexity of management and operation and maintenance, and realizes unified management and scheduling of multiple heterogeneous computing resources under the management of the intelligent computing center cloud platform. Attached Figure Description

[0034] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of a method for managing heterogeneous computing resources in an intelligent computing center cloud platform provided by the present invention;

[0036] Figure 2 This is a structural diagram of a heterogeneous computing power resource management device for an intelligent computing center cloud platform provided by the present invention;

[0037] Figure 3 This is a structural diagram of an electronic device provided by the present invention. Detailed Implementation

[0038] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0039] The “computing power” mentioned in this invention refers to: the ability of computer equipment or computing / data center to process information; the ability of computer hardware and software to work together to perform a certain computing requirement; the computing power to achieve the target result output by processing information data; and a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity, mainly providing services to society through computing power infrastructure.

[0040] The "computational power" (CP) described in this invention refers to the ability of a data center server to process data and output results. It is a comprehensive indicator of a data center's computing power, encompassing general computing power, supercomputing power, and intelligent computing power. The commonly used unit of measurement is floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), with higher values ​​indicating stronger overall computing power. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A supercomputers, 500,000 mainstream server CPUs, or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 +CP 智能 +CP 超级 .

[0041] The "Network Power" (NP) mentioned in this invention refers to the performance of data transmission capability of computing facilities, which includes comprehensive capabilities such as network architecture, network bandwidth, transmission latency, intelligent management and scheduling, and involves network transmission within and between data centers. It is a comprehensive indicator for measuring network transmission scheduling capability.

[0042] The "Storage Power" (SP) described in this invention refers to the comprehensive capabilities of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon operation. It is a comprehensive indicator for measuring the data storage capacity of a data center, including external storage devices such as storage arrays and internal storage devices in servers. The commonly used unit of measurement for storage capacity is exabytes (EB, 1EB = 2^60 bytes), the commonly used unit of measurement for performance is the number of read / write operations per second per unit capacity (IOPS / TB), and the disaster recovery ratio is an important indicator of security and reliability.

[0043] The "computing infrastructure" mentioned in this invention refers to a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage capacity, enabling centralized computing, storage, transmission, and application of information.

[0044] The "new information infrastructure" mentioned in this invention refers to network infrastructure such as 5G networks, fiber optic broadband networks, backbone networks, international communication networks, and satellite internet; computing infrastructure such as data centers, general computing centers, intelligent computing centers, and supercomputing centers; and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0045] The “computing power” mentioned in this invention includes: general computing power, intelligent computing power, and supercomputing power.

[0046] The "general computing power" mentioned in this 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.

[0047] The "intelligent computing power" mentioned in this invention refers to: a computing platform deployed on a large scale based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit) for various artificial intelligence innovative applications, such as natural language processing and machine vision.

[0048] The “supercomputing power” mentioned in this invention 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 uses a dedicated operating system to handle extremely complex or data-intensive problems. It is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, and gene analysis.

[0049] The "intelligent computing center cloud platform" mentioned in this invention, abbreviated as "intelligent computing cloud", refers to a cloud computing platform that integrates hardware and software resources based on an intelligent computing center.

[0050] The "intelligent computing center" mentioned in this invention includes, but is not limited to, "smart computing center".

[0051] The "intelligent computing center" mentioned in this invention, also known as an artificial intelligence computing center, is a type of computing 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.

[0052] The "computing center" mentioned in this invention refers to a facility that is mainly composed of infrastructure such as wind, thermal, hydro, and electricity, and IT hardware and software equipment, and has computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0053] The "supercomputing center" mentioned in this invention refers to a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters. It can provide large-scale computing, storage and network services and is widely used in aerospace, defense, oil exploration, climate modeling and genome sequencing and other application scenarios.

[0054] The “computing resources” mentioned in this invention refer to the technologies and facilities required for the development of the digital society that have the ability to compute, transmit, store and apply information, including but not limited to computing resources such as CPUs and GPUs, computing resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and supporting and guaranteeing resources such as wind, fire, water and electricity.

[0055] The "model" mentioned in this invention includes, but is not limited to, "large language model" and "multimodal large model".

[0056] The "large language model" mentioned in this invention refers to a large-scale language model (LLM), which is a language model with a large number of parameters. It is designed to understand and generate human language, and is trained with a large amount of text data. It can perform a wide range of tasks, including text summarization, translation, and sentiment analysis.

[0057] The “Multimodal Large Models” mentioned in this invention refer to models that combine multimodal information such as text, images, videos, and audio for training, including but not limited to multimodal large language models.

[0058] The “computing power running task” mentioned in this invention refers to a specific workload or job executed on computing power resources that requires a certain amount of computing power support. It usually involves complex data processing, numerical calculation, model training or simulation scenarios.

[0059] The "heterogeneous computing resources" mentioned in this invention refer to a comprehensive system composed of different types of computing units such as CPU, GPU, and NPU. It not only includes the computing capabilities of various hardware, but also integrates supporting resources such as storage, network, and scheduling management to support its efficient operation. By coordinating and adapting to different task requirements, it achieves optimal utilization of overall computing power.

[0060] The "integration and management" mentioned in this invention refers to the unified management and scheduling of at least two types of computing resources to achieve centralized management of computing resources and improve the utilization rate of computing resources.

[0061] Please see Figure 1 , Figure 1 This is a flowchart of a method for managing heterogeneous computing resources in an intelligent computing center cloud platform provided by the present invention, such as... Figure 1 As shown, it includes the following steps:

[0062] Step S1: Register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources.

[0063] In this invention, multiple first computing resources can be different types of computing resources provided by different operators, such as AI computing clusters, virtual machines, etc. These computing resources belong to heterogeneous computing resources. In this embodiment, the management platform registers multiple first computing resources to obtain registered second computing resources. These registered second computing resources can be used within the management platform; that is, the management platform controls and schedules the multiple second computing resources. It should be noted that controlling the resources means performing operations such as merging, adding, or deleting the registered second computing resources. Scheduling the resources means that after receiving a user's computing task, the platform can process the user's computing task by scheduling the second computing resources.

[0064] In a feasible embodiment, the management platform may optionally include a first controller and a second controller, wherein the first controller is used to control the plurality of second computing resources, and the second controller is used to schedule the plurality of second computing resources.

[0065] The management platform comprises a first controller, Kamaji, and a second controller, Sveltos. Kamaji is a distributed computing platform designed to provide users with efficient and low-cost computing resources. As a Kubernetes controller, Kamaji allows for the unified management of different types of computing resources (including physical hosts, virtual machines, containers, etc.) within the Kubernetes cluster. These resources can span different cloud platforms or on-premises data centers. By implementing resource scheduling at a layer above Kubernetes, Kamaji enables users to manage different types of computing resources through a unified API, without needing to concern themselves with the differences in underlying resources (such as different cloud vendors or hardware environments). Kubernetes (often abbreviated as "K8s") is an open-source container orchestration platform designed to automate the deployment, scaling, and management of containerized applications.

[0066] Sveltos is an open-source distributed computing platform designed to simplify and optimize the management and execution of large-scale computing tasks, especially in cloud environments and edge computing. Sveltos allows users to dynamically schedule and manage computing resources across multiple cloud platforms, providing cross-cloud resource scheduling capabilities. It supports unified management of computing resources from different cloud vendors, integrating different types of computing resource pools (AI clusters, virtual machines, containers, etc.) into a single platform. Sveltos manages computing resources through automated policies, such as dynamically scaling resources up or down based on demand, or replicating an already deployed AlayaNeW cluster to a newly managed cluster via Sveltos, achieving automated deployment. Different labels can be used to determine different operations on different clusters.

[0067] In this invention, Kamaji and Sveltos utilize Kubernetes' resource scheduling and management mechanism, combined with their respective functions, to achieve unified control and scheduling of different types of computing resources and computing power resources from different cloud vendors. Kamaji is primarily used to control multiple secondary computing power resources, while Sveltos is primarily used to schedule multiple secondary computing power resources.

[0068] Step S2: Receive a user request input by the user on the management platform. The user request is used to request the management platform to call computing resources to process computing tasks.

[0069] In this embodiment, the user request is input by the user on the management platform. The user request includes a computing power task that the user requests to be processed, and this task requires computing power resources from the management platform. After receiving the user request, the management platform analyzes the computing power task to determine the computing power resources that match it.

[0070] Step S3: Based on the user's request, match the multiple second computing resources to obtain at least one third computing resource, and schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

[0071] In this invention, based on the user's request, at least one third computing resource is determined by matching among multiple registered second computing resources. The third computing resource can be any one of the multiple second computing resources. That is, the third computing resource can be at least one second computing resource. Therefore, at least one third computing resource is scheduled to process the computing task.

[0072] This invention registers multiple first computing resources belonging to different heterogeneous computing resources in a management platform, controls and schedules multiple registered second computing resources, and upon receiving a user request, matches the multiple second computing resources according to the user request to determine at least one third computing resource to process the computing task. This greatly reduces the complexity of management and operation and maintenance, and realizes unified management and scheduling of multiple heterogeneous computing resources under the management of the intelligent computing center cloud platform.

[0073] In some feasible implementations, step S1 may optionally include:

[0074] Step S11: Obtain the multiple first computing resources corresponding to the multiple different operators, wherein the multiple first computing resources include at least one of: AI computing cluster, host, virtual machine and containerized computing resources;

[0075] Step S12: Register the plurality of first computing resources in the first controller to obtain the plurality of registered second computing resources and a plurality of resource tags. The plurality of second computing resources and the plurality of resource tags correspond one-to-one, and the resource tags are used to indicate the resource information of the corresponding second computing resources.

[0076] In this embodiment, multiple first computing resources are provided by multiple different operators, and different operators can provide one or more computing resources. Specifically, the first computing resources include at least one of the following: AI computing clusters, hosts, storage, networks, virtual machines, and containerized computing resources, thereby realizing unified control of computing resources from multiple different operators.

[0077] Specifically, multiple first computing resources are first registered in the first controller to obtain multiple registered second computing resources and multiple resource tags. Each resource tag corresponds to a second computing resource, that is, the resource tag is used to indicate the resource information of the corresponding second computing resource. The resource information may include configuration information, computing resource information, container information, or container information, etc.

[0078] The first controller, Kamaji, registers different resource pools with the cluster through the Kubernetes controller and uses different tags to determine which resource pool to run containers or virtual machines on. It provides load balancing, fault tolerance, and resource optimization functions across resource pools. The second controller, Sveltos, is responsible for resource management across cloud environments. It can uniformly schedule resource pools from different cloud vendors, support resource management across multiple cloud platforms and computing resource types, and ensure resource utilization and high availability.

[0079] Optionally, after registering the plurality of first computing resources in the first controller in step S12, and obtaining the plurality of registered second computing resources and the plurality of resource tags, the method further includes:

[0080] Step S13: Store the resources in the second controller based on the multiple resource tags;

[0081] Step S14: Based on the multiple resource tags, schedule the multiple second computing resources.

[0082] In this invention, after generating multiple resource tags, these tags need to be stored in a second controller. This facilitates the second controller's scheduling of second computing resources, enabling resource scheduling across different operators. Specifically, the first and second controllers can schedule computing resources from different cloud vendors to a Kubernetes cluster, and the uniformly managed resource pool automatically schedules tasks according to policies. This allows users to seamlessly scale applications and deploy and manage computing resources across clouds.

[0083] During the scheduling of the second computing power resources, multiple resource tags are matched with the task information corresponding to the computing power running tasks to determine the resource tags corresponding to the computing power running tasks, and multiple second computing power resources are scheduled through the resource tags.

[0084] Optionally, the management platform further includes an application programming interface (API), which is used to perform unified management of the first controller and the second controller based on the API.

[0085] In this embodiment, the management platform is also equipped with an application programming interface (API), which allows users to manage these computing resources through the unified API provided by Kubernetes, without needing to concern themselves with the specific implementation details of the underlying resources. Kamaji and Sveltos will handle cross-platform and cross-resource pool scheduling and management, thereby achieving unified management of Kamaji and Sveltos.

[0086] Optionally, step S3 includes:

[0087] Step S31: Based on the user's request, determine the configuration environment information and computing resource information corresponding to the computing power running task;

[0088] Step S32: Based on the configuration environment information and the computing resource information, match the plurality of second computing resources to obtain the at least one third computing resource;

[0089] Step S33: In the management platform, schedule at least one third computing power resource to process the computing power operation task.

[0090] In this invention, user requests are analyzed to determine the configuration environment information and computing resource information corresponding to the computing power running task. The configuration environment information may include hardware environment, operating system, software environment, virtual environment, etc. The computing resource information may include CPU information, GPU information, memory information, etc.

[0091] Based on configuration environment information and computing resource information, the system matches the system with multiple second computing resources to determine at least one third computing resource. The computing tasks are then processed through this at least one third computing resource, thereby ensuring the processing efficiency and accuracy of the computing tasks.

[0092] This invention registers multiple first computing resources belonging to different heterogeneous computing resources in a management platform, controls and schedules multiple registered second computing resources, and upon receiving a user request, matches the multiple second computing resources according to the user request to determine at least one third computing resource to process the computing task. This greatly reduces the complexity of management and operation and maintenance, and realizes unified management and scheduling of multiple heterogeneous computing resources under the management of the intelligent computing center cloud platform.

[0093] Please see Figure 2 , Figure 2 This is a structural diagram of a heterogeneous computing power resource management device for an intelligent computing center cloud platform provided by the present invention, as shown below. Figure 2 The heterogeneous computing power resource management device 200 of the intelligent computing center cloud platform shown includes:

[0094] The registration module 210 is used to register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources.

[0095] The receiving module 220 is used to receive a user's usage request input by the user on the management platform, wherein the user usage request is used to request the management platform to call computing resources to process computing power running tasks;

[0096] The matching module 230 is used to match the multiple second computing resources based on the user's request to obtain at least one third computing resource, and to schedule the at least one third computing resource to process the computing power running task, wherein the third computing resource is any one of the multiple second computing resources.

[0097] Optionally, the management platform includes a first controller and a second controller, wherein the first controller is used to control the plurality of second computing resources, and the second controller is used to schedule the plurality of second computing resources.

[0098] Optionally, the registration module 210 includes:

[0099] The acquisition submodule is used to acquire the multiple first computing resources corresponding to the multiple different operators. The multiple first computing resources include at least one of the following: AI computing cluster, host, storage, network, virtual machine and containerized computing resources.

[0100] The registration submodule is used to register the plurality of first computing resources in the first controller to obtain the registered plurality of second computing resources and a plurality of resource tags. The plurality of second computing resources and the plurality of resource tags correspond one-to-one, and the resource tags are used to indicate the resource information of the corresponding second computing resources.

[0101] Optional, also includes:

[0102] A storage submodule is used to store the multiple resource tags in the second controller;

[0103] The scheduling submodule is used to schedule the multiple second computing resources based on the multiple resource tags.

[0104] Optionally, the management platform further includes an application programming interface (API), which is used to perform unified management of the first controller and the second controller based on the API.

[0105] Optionally, the matching module 230 includes:

[0106] The determination submodule is used to determine the configuration environment information and computing resource information corresponding to the computing power running task based on the user's usage request;

[0107] A matching submodule is used to match the plurality of second computing resources based on the configuration environment information and the computing resource information to obtain the at least one third computing resource.

[0108] The processing submodule is used to schedule the at least one third computing power resource in the management platform to process the computing power operation task.

[0109] This invention registers multiple first computing resources belonging to different heterogeneous computing resources in a management platform, controls and schedules multiple registered second computing resources, and upon receiving a user request, matches the multiple second computing resources according to the user request to determine at least one third computing resource to process the computing task. This greatly reduces the complexity of management and operation and maintenance, and realizes unified management and scheduling of multiple heterogeneous computing resources under the management of the intelligent computing center cloud platform.

[0110] This invention also provides an electronic device. Please refer to [link to relevant documentation]. Figure 3 The electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and capable of running on the processor 301.

[0111] When program 3021 is executed by processor 301, it can achieve the following: Figure 1 Any step in the corresponding method embodiment:

[0112] Step S1: Register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources.

[0113] Step S2: Receive a user usage request input by the user on the management platform. The user usage request is used to request the management platform to call computing resources to process computing power running tasks.

[0114] Step S3: Based on the user's request, match the multiple second computing resources to obtain at least one third computing resource, and schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

[0115] Optionally, the management platform includes a first controller and a second controller, wherein the first controller is used to control the plurality of second computing resources, and the second controller is used to schedule the plurality of second computing resources.

[0116] Optionally, step S1 includes:

[0117] Step S11: Obtain the multiple first computing resources corresponding to the multiple different operators. The multiple first computing resources include at least one of the following: AI computing cluster, host, storage, network, virtual machine and containerized computing resources.

[0118] Step S12: Register the plurality of first computing resources in the first controller to obtain the plurality of registered second computing resources and a plurality of resource tags. The plurality of second computing resources and the plurality of resource tags correspond one-to-one, and the resource tags are used to indicate the resource information of the corresponding second computing resources.

[0119] Optionally, after registering the plurality of first computing resources in the first controller in step S12, and obtaining the plurality of registered second computing resources and the plurality of resource tags, the method further includes:

[0120] Step S13: Store the resources in the second controller based on the multiple resource tags;

[0121] Step S14: Based on the multiple resource tags, schedule the multiple second computing resources.

[0122] Optionally, the management platform further includes an application programming interface (API), which is used to perform unified management of the first controller and the second controller based on the API.

[0123] Optionally, step S3 includes:

[0124] Step S31: Based on the user's request, determine the configuration environment information and computing resource information corresponding to the computing power running task;

[0125] Step S32: Based on the configuration environment information and the computing resource information, match the plurality of second computing resources to obtain the at least one third computing resource;

[0126] Step S33: In the management platform, schedule at least one third computing power resource to process the computing power operation task.

[0127] This invention registers multiple first computing resources belonging to different heterogeneous computing resources in a management platform, controls and schedules multiple registered second computing resources, and upon receiving a user request, matches the multiple second computing resources according to the user request to determine at least one third computing resource to process the computing task. This greatly reduces the complexity of management and operation and maintenance, and realizes unified management and scheduling of multiple heterogeneous computing resources under the management of the intelligent computing center cloud platform.

[0128] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described embodiment of the heterogeneous computing resource management method for the intelligent computing center cloud platform, achieving the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0129] This invention also provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described method for managing heterogeneous computing resources in a smart computing center cloud platform, and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0130] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0131] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, 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 to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0132] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for managing heterogeneous computing resources in an intelligent computing center cloud platform, characterized in that, The method includes: Step S1: Register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources. Step S2: Receive a user usage request input by the user on the management platform. The user usage request is used to request the management platform to call computing resources to process computing power running tasks. Step S3: Based on the user's request, match the multiple second computing resources to obtain at least one third computing resource, and schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

2. The method according to claim 1, characterized in that, The management platform includes a first controller and a second controller. The first controller is used to control the plurality of second computing resources, and the second controller is used to schedule the plurality of second computing resources.

3. The method according to claim 2, characterized in that, Step S1 includes: Step S11: Obtain the multiple first computing resources corresponding to the multiple different operators. The multiple first computing resources include at least one of the following: AI computing cluster, host, storage, network, virtual machine and containerized computing resources. Step S12: Register the plurality of first computing resources in the first controller to obtain the plurality of registered second computing resources and a plurality of resource tags. The plurality of second computing resources and the plurality of resource tags correspond one-to-one, and the resource tags are used to indicate the resource information of the corresponding second computing resources.

4. The method according to claim 3, characterized in that, Step S12, registering the plurality of first computing resources in the first controller, and after obtaining the registered plurality of second computing resources and the plurality of resource tags, the method further includes: Step S13: Store the resources in the second controller based on the multiple resource tags; Step S14: Based on the multiple resource tags, schedule the multiple second computing resources.

5. The method according to any one of claims 1-4, characterized in that, The management platform also includes an application programming interface (API), which is used to manage the first controller and the second controller in a unified manner based on the API.

6. The method according to claim 1, characterized in that, Step S3 includes: Step S31: Based on the user's request, determine the configuration environment information and computing resource information corresponding to the computing power running task; Step S32: Based on the configuration environment information and the computing resource information, match the plurality of second computing resources to obtain the at least one third computing resource; Step S33: In the management platform, schedule at least one third computing power resource to process the computing power operation task.

7. A heterogeneous computing power resource management device for an intelligent computing center cloud platform, characterized in that, The device includes: The registration module is used to register multiple first computing resources corresponding to multiple different operators in the management platform to obtain multiple second computing resources after registration. The management platform is used to control and schedule the multiple second computing resources. The multiple first computing resources are different types of computing resources. The receiving module is used to receive user usage requests input by users on the management platform, wherein the user usage requests are used to request the management platform to call computing resources to process computing power running tasks; The matching module is used to match the multiple second computing resources based on the user's request to obtain at least one third computing resource, and to schedule the at least one third computing resource to process the computing power running task. The third computing resource is any one of the multiple second computing resources.

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

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 6.

Images