Method and device for intelligent computing center to provide computing power resources through computing power package

Abstract

The invention provides a method and device for an intelligent computing center to provide computing power resources through computing power packages, and relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructure, and the method comprises the steps: S1, determining a computing power package corresponding to a to-be-processed computing power task under the condition that the computing power resources need to be called, the computing power task is a task needing to call computing power resource processing, and the computing power package is used for metering the computing power of the intelligent computing center according to the calculated amount; s2, computing power resources are called based on an application program interface corresponding to the computing power package, the application program interface comprises a code program corresponding to operation for executing the computing power task, and the computing power resources called by the application program interface are matched with the operation environment of the computing power task. According to the method, the scheduling efficiency of the computing power resources can be improved, the computing power is metered according to the calculated amount through the computing power package, and wide application of the general computing power is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent computing centers, smart computing centers and computing power infrastructure, and specifically to a method and device for an intelligent computing center to provide computing power resources through a computing power package. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have come into being.

[0003] "Intelligent computing center" refers to a facility that provides the required computing power, data and algorithms for artificial intelligence applications (such as artificial intelligence deep learning model development, model training and model reasoning) by using large-scale heterogeneous computing 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 bottom-level computing power to top-level application enablement.

[0004] “Intelligent Computing Center” includes but is not limited to “Smart Computing Center”.

[0005] "Intelligent Computing Center" refers to an artificial intelligence computing center. It is a type of computing infrastructure that is based on artificial intelligence theory, adopts artificial intelligence computing architecture, and provides computing power services, data services, and algorithm services required for artificial intelligence applications.

[0006] "Computing power" is the core of "intelligent computing center" and "intelligent 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 execute certain computing needs. It is the computing power to process information data and achieve target result output. 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] At present, cloud servers can directly provide computing resources that meet user needs, and users use computing resources to develop models and other tasks. Before executing computing tasks for model development through cloud servers, testers need to build an operating framework based on the specific parameters of the computing tasks and set the corresponding operating environment parameters. Only then can the cloud server execute computing tasks based on the operating framework and operating environment parameters. The process of building an operating framework and setting the corresponding operating environment parameters is relatively complicated and time-consuming, resulting in low efficiency in scheduling computing resources to process computing tasks by the intelligent computing center. At the same time, since the GPU resources of the intelligent computing center need to be occupied for a long time during model development, business personnel need to rent the computing services of the intelligent computing center for a long time, resulting in high costs for model development, making it difficult to achieve widespread application of inclusive computing power.

[0008] It can be seen that the existing technology has problems such as low efficiency of intelligent computing centers in scheduling computing resources to process computing tasks and high model development costs. Summary of the invention

[0009] The present invention provides a method and device for an intelligent computing center to provide computing power resources through a computing power package, so as to solve the problems in the prior art of low efficiency in scheduling computing power tasks and using computing power, as well as high model development cost.

[0010] To solve the above problems, the present invention is achieved as follows: In a first aspect, the present invention provides a method for an intelligent computing center to provide computing resources through a computing power package, comprising: Step S1: when computing power resources need to be called, determine the computing power package corresponding to the computing power task to be processed, where the computing power task is a task that needs to be processed by calling computing power resources, and the computing power package is used to measure the computing power of the intelligent computing center according to the computing amount; Step S2: Call computing resources based on the application program interface corresponding to the computing package, where the application program interface includes a code program corresponding to the operation of executing the computing task, and the computing resources called by the application program interface match the operating environment of the computing task.

[0011] In one embodiment, the computing power package corresponds to a plurality of application program interfaces, and step S2 includes: Step S21, determining the target task type corresponding to the computing task; Step S22: querying a target application program interface corresponding to the target task type based on a preset mapping table, wherein the preset mapping table includes a plurality of task types and the plurality of application program interfaces, and a mapping relationship between different task types and different application program interfaces, and the target application program interface is one of the plurality of application program interfaces; Step S23: calling computing resources based on the target application program interface.

[0012] In one embodiment, the multiple task types include at least one of a model training task type, an inference task type, and a data fine-tuning task type, and each task type corresponds to at least one application programming interface.

[0013] In one embodiment, after step S2, the method further includes: Step S3: Processing the computing task based on the computing resources called by the target application program interface; The computing task is a computing task processed by the target container, and step S3 includes at least one of the following: Step S31: when the load data is greater than or equal to a first preset threshold, expand the target container based on the target application program interface, and the load data is the load data of the target container; Step S32: When the load data is less than or equal to a second preset threshold, shrink the target container based on the target application program interface.

[0014] In one embodiment, step S31 includes: Step S311: When the load data is greater than or equal to the first preset threshold, obtain first computing resource information of a first server, where the first server is the server to which the target container belongs; Step S312: When the first computing power resource information satisfies the load data, create a first container based on the computing power resources of the first server through the target application program interface; Step S313: deploy the relevant data of the computing task to the first container through the target application program interface; Step S314: Execute the computing task based on the first container.

[0015] In one embodiment, the step S31 further includes: Step S315: When the first computing power resource information does not satisfy the load data, obtain network information and second computing power resource information of a second server, where the second server and the first server are in the same area or in different areas; Step S316: When the second computing power resource information satisfies the load data and the network information satisfies the preset network condition, create a second container based on the computing power resources of the second server; Step S317: deploy the relevant data of the computing task to the second container; Step S318: Execute the computing task based on the second container.

[0016] In one embodiment, after step S3, the method further includes: Step S4: Measure the computing resource consumption of the computing package based on the target container, the first container or the second container, where the target container, the first container and the second container are containers for executing the computing task.

[0017] In a second aspect, the present invention further provides a device for an intelligent computing center to provide computing resources through a computing power package, comprising: A determination module, used to determine a computing power package corresponding to a computing power task to be processed when computing power resources need to be called, wherein the computing power task is a task that needs to be processed by calling computing power resources, and the computing power package is used to measure the computing power of the intelligent computing center according to the computing amount; A calling module is used to call computing power resources based on the application program interface corresponding to the computing power package. The application program interface includes a code program corresponding to the operation of executing the computing power task. The computing power resources called by the application program interface match the operating environment of the computing power task.

[0018] In a third aspect, the present invention further provides an electronic device comprising 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, implements the steps in the method of providing computing power resources through a computing power package by an intelligent computing center as described in the first aspect above.

[0019] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the intelligent computing center providing computing power resources through a computing power package as described in the first aspect above are implemented.

[0020] In a fifth aspect, the present invention further provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps in the method of providing computing power resources through a computing power package by an intelligent computing center as described in the first aspect above.

[0021] In the present invention, when it is necessary to call computing power resources, determine the computing power package corresponding to the computing power task to be processed, the computing power task is a task that needs to call computing power resources for processing, and the computing power package is used to measure the computing power of the intelligent computing center according to the amount of calculation; based on the application program interface corresponding to the computing power package, the computing power resources are called, and the application program interface includes a code program corresponding to the operation of executing the computing power task, and the computing power resources called by the application program interface match the operating environment of the computing power task. In this way, when it is necessary to execute a computing power task, the computing power resources of the intelligent computing center can be quickly scheduled to process the computing power tasks through the application program interface of the intelligent computing center, and the user does not need to configure the operating framework and set the operating environment parameters for each computing power task, which effectively improves the scheduling efficiency of computing power resources, and then can effectively improve the execution efficiency of computing power tasks. At the same time, by measuring computing power according to the amount of calculation through the computing power package, the problem of high costs caused by business personnel leasing the computing power services of the intelligent computing center for a long time can be avoided, and the widespread application of inclusive computing power can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use 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 ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0023] Figure 1 It is a flow chart of a method provided by the present invention for an intelligent computing center to provide computing power resources through a computing power package; Figure 2 It is a structural diagram of the intelligent computing center provided by the present invention; Figure 3 It is a structural diagram of a device for providing computing power resources through a computing power package in an intelligent computing center provided by the present invention; Figure 4 It is a structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The "computing power" mentioned in the present 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 execute certain computing requirements, and the computing power to achieve target result output by processing information data. It is a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity, and it mainly provides services to the society through computing power infrastructure.

[0026] The "computing power" (Computational Power, CP) mentioned in the present invention refers to: it is the ability of a data center server to process data and output results. It is a comprehensive indicator to measure the computing power of a data center, including general computing power, super computing power and intelligent computing power. The commonly used unit of measurement is the number of floating-point operations performed per second (FLOPS, 1EFLOPS=10^18FLOPS). The larger the value, the stronger the comprehensive computing power. According to calculations, 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream notebooks. The calculation formula is: CP=CP 通用 +CP 智能 +CP 超级 .

[0027] The "carrying capacity" (Network Power, NP) mentioned in the present invention refers to: it is the performance of the data transmission capacity of the computing power facilities, including the comprehensive capabilities of network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc. It involves network transmission within and between data centers, and is a comprehensive indicator for measuring network transmission scheduling capabilities.

[0028] The "Storage Power" (SP) mentioned in the present invention refers to: the comprehensive capabilities of a data center in terms of data storage capacity, performance, safety 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 server built-in storage devices. The commonly used unit of measurement for storage capacity is exabyte (EB, 1EB=2^60bytes), and the commonly used unit of measurement for performance is the number of reads and writes per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB). The disaster recovery ratio is an important manifestation of safety and reliability.

[0029] The "computing power infrastructure" mentioned in the present invention refers to: a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage capacity, which can realize centralized computing, storage, transmission and application of information, and presents characteristics such as diversity and ubiquity, intelligence and agility, safety and reliability, green and low-carbon. It is of great significance to promote industrial transformation and upgrading, enable my country's scientific and technological innovation, meet people's better life and realize high-efficiency social governance.

[0030] The "new information infrastructure" mentioned in the present invention refers to: mainly including network infrastructure such as 5G network, fiber broadband network, backbone network, international communication network, satellite Internet, computing power infrastructure such as data center, general computing power center, intelligent computing center, supercomputing center, and new technology facilities such as artificial intelligence, blockchain, and quantum computing. With the emergence and promotion and application of new general technologies, the new information infrastructure will become more diverse.

[0031] The “computing power” mentioned in the present invention includes: general computing power, intelligent computing power and super computing power.

[0032] The “general computing power” mentioned 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.

[0033] The "intelligent computing power" mentioned in the present invention refers to: a computing platform based on GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit) and other dedicated chips for various innovative artificial intelligence applications, such as natural language processing, machine vision, etc.

[0034] The "super computing power" mentioned in the present invention refers to: the computing power mainly provided by high-performance computing clusters such as supercomputers. It uses 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 calculations in cutting-edge scientific fields, such as planetary simulation, drug molecule design, gene analysis, etc.

[0035] The "intelligent computing center" mentioned in the present invention refers to a facility that provides the required computing power, data and algorithms for artificial intelligence applications (such as artificial intelligence deep learning model development, model training and model reasoning) by using large-scale heterogeneous computing resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from bottom-level computing power to top-level application enablement.

[0036] The “intelligent computing center” mentioned in the present invention includes but is not limited to the “intelligent computing center”.

[0037] The "intelligent computing center" mentioned in the present invention is an artificial intelligence computing center, which is a type of computing power infrastructure that is based on artificial intelligence theory, adopts artificial intelligence computing architecture, and provides computing power services, data services, and algorithm services required for artificial intelligence applications.

[0038] The "computing power center" mentioned in the present invention refers to: a facility that is mainly composed of infrastructure such as wind, fire, water, electricity, and IT hardware and software equipment, and has computing power, transportation capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0039] The "supercomputing center" mentioned in the present invention refers to: a supercomputing data center, which is a data center based on a supercomputer or a large-scale computing cluster, which can provide 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 resources" mentioned in the present invention refer to: technologies and facilities with information computing, transmission, storage and application capabilities required for the development of a digital society, including but not limited to computing resources such as CPUs and GPUs, 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, and supporting and guarantee resources such as wind, fire, water and electricity.

[0041] The "inclusive computing power" mentioned in the present invention refers to providing appropriate and effective computing power services at an affordable cost to all social classes and groups that have computing power service needs based on the requirements of equal opportunity and the principle of commercial sustainability.

[0042] The “model” mentioned in the present invention includes but is not limited to a “large language model” and a “multimodal large model”.

[0043] The “large language model” mentioned in the present invention refers to a large language model (LLM), which is a language model with a large parameter scale, designed 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.

[0044] The "Multimodal Large Models" mentioned in the present invention refer to models that combine multimodal information such as text, images, videos, audio, etc. for training, including but not limited to multimodal large language models.

[0045] 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 with 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 performs actions to affect the environment or achieve predetermined goals. Agents are widely used in the field of artificial intelligence and are commonly seen in automation systems, robots, virtual assistants, and game characters. The core of agents is the ability to learn autonomously and continuously evolve to better complete tasks and adapt to complex environments.

[0046] The "computing power package" mentioned in the present invention refers to the following three points: 1. After the computing power resources are packaged with "degree" as the basic unit of measurement, the computing power service is provided to the user. The user can purchase or lease the computing power of the intelligent computing center through the "computing power service". Among them, the measurement of computing power resources with "degree" as the basic unit of measurement can refer to the patent document CN 118760575B of the applicant: "1 degree computing power" is 1 unit of computing power, which can be specifically X TFLOPS·h, that is, X trillion floating-point operations per second × hours, where "X TFLOPS·h" can be but not limited to: "1TFLOPS·h", "312 TFLOPS·h", "624TFLOPS·h", "1024 TFLOPS·h", "10000 TFLOPS·h" and other arbitrary numbers. The numbers are only illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms of extensions without departing from the scope of protection of the present invention and the claims, all of which belong to the protection of the present invention.

[0047] 2. It is used to measure the computing capacity of the intelligent computing center according to the amount of computing, realize billing according to the amount of computing, and only charge when the computing task is running. It can be purchased on demand and used immediately after purchase.

[0048] 3. Flexible scheduling of GPU resources, elastic expansion or contraction of computing resources on demand, greatly reducing the waste of computing resources; achieving out-of-the-box use of more economical and high-performance "computing power + algorithm" integrated artificial intelligence services, providing a complete large model + agent development tool chain; and ultimately achieving widespread application of "inclusive computing power".

[0049] Specifically, the "computing power package" described in the present invention includes the following three functional service components: 1. Application Programming Interface. The application programming interface includes the code program for executing computing task operations. That is, the application programming interface defines a set of fully functional, clear, flexible and easy-to-use application programming interfaces, through which the computing resources corresponding to the computing tasks can be quickly called. Among them, the application programming interface can automatically match the computing resources that meet the needs based on the computing resource requirements, machine resource occupancy, GPU type, network interconnection, storage usage and other conditions, and efficiently complete the user's computing needs.

[0050] 2. Capacity expansion and contraction service component: The capacity expansion and contraction service component is used to expand and contract the containers that perform computing tasks, so that the computing resources occupied by the containers remain at a reasonable level. While ensuring the smooth execution of computing tasks, less computing resources of the intelligent computing center are occupied, allowing the intelligent computing center to provide computing services to more users and realize the widespread application of inclusive computing power.

[0051] 3. Metering component: The metering component is used to measure the computing resources consumed in the process of executing computing tasks. In the process of measuring computing resource consumption, only the computing resources consumed when the container executes computing tasks are measured, and the computing resources consumed during the expansion and contraction of the container are not measured, which improves the accuracy of measurement, reduces the computing resources that users need to rent, and reduces the user's usage costs, thereby realizing the widespread application of inclusive computing power.

[0052] The "computing power package" described in the present invention may also be referred to as a "computing power set", "computing power library", "computing power group", etc. The names are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field may make many forms of extensions without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

[0053] In the prior art, users need to use a large amount of computing resources during the process of software or model development. In order to reduce the cost of using computing resources, users usually choose to rent computing services from cloud servers. When software or model development is needed, the cloud servers are used to provide computing power to run the software or models, thereby reducing the cost of directly purchasing hardware equipment.

[0054] See also Figure 1 , Figure 1 It is a flow chart of a method provided by the present invention for an intelligent computing center to provide computing resources through a computing package, such as Figure 1 As shown, the following steps are included: Step S1: When computing power resources need to be called, determine the computing power package corresponding to the computing power task to be processed, where the computing power task is a task that needs to call computing power resources for processing, and the computing power package is used to measure the computing power of the intelligent computing center according to the computing amount.

[0055] The above computing tasks are tasks that users need the intelligent computing center to perform. The computing tasks can be tasks such as training models, reasoning and fine-tuning data using the intelligent agents of the intelligent computing center, etc. It should be noted that the computing tasks are computing tasks created by users based on the computing packages of the intelligent computing center, that is, each computing task corresponds to a computing package, and the computing tasks created are implemented through the algorithm framework, data set and / or preset model provided by the computing package. The computing package of the intelligent computing center can call computing resources to process computing tasks.

[0056] The above-mentioned intelligent computing center is used to provide computing resources to perform computing tasks. Among them, the computing resources are packaged in the form of computing packages, and the computing power consumed is measured according to the specific computing amount consumed during the execution of computing tasks.

[0057] It should be noted that in the prior art, computing power is mainly provided based on the length of the lease, and the length of time for leasing computing power services is determined by the length of time that users need to develop software. Computing power resources are consumed when the software or model is running, but they do not need to be consumed when the software or model is adjusted, resulting in underutilization of computing power resources and high costs for users to use computing power services.

[0058] In the present invention, computing power servers are provided based on computing volume through computing power packages. Compared with the method of providing computing power based on rental time, the computing power utilization rate can be more fully improved, while reducing the user's usage cost, thereby realizing the widespread application of inclusive computing power.

[0059] Step S2: Call computing resources based on the application program interface corresponding to the computing package, where the application program interface includes a code program corresponding to the operation of executing the computing task, and the computing resources called by the application program interface match the operating environment of the computing task.

[0060] The above-mentioned application program interface encapsulates the computing operations of different computing tasks. Through the application program interface, it is possible to quickly schedule computing resources that match the computing tasks and execute computing tasks based on the computing resources. Users do not need to configure the operating framework and set the operating environment parameters for each computing task, which can effectively improve the efficiency of computing resource scheduling and thus improve the execution efficiency of computing tasks.

[0061] It should be noted that the application programming interface is the application programming interface corresponding to the computing power package. When the user chooses to purchase or lease computing power resources in the form of a computing power package (that is, computing power resources are measured by computing volume), the application programming interface needs to be called to process computing power tasks. When the user does not choose to purchase or lease computing power resources in the form of a computing power package (that is, computing power resources are not measured by computing volume), there is no need to call the application programming interface to process computing power tasks.

[0062] Specifically, the application program interface includes the code program corresponding to the operation of the computing task, that is, the application program interface defines a set of fully functional, clear, flexible and easy-to-use application program interfaces, through which the computing resources corresponding to the computing task can be quickly called. Among them, the application program interface can call computing resource requirements, machine resource occupancy, GPU type, network interconnection, storage usage and other conditions to automatically match the computing resources that meet the needs, and efficiently complete the user's computing needs.

[0063] like Figure 2As shown, the intelligent computing center includes the user layer, the application program interface layer of the computing power package, infrastructure software and infrastructure hardware, among which the user layer is used to provide users with a large model tool chain, and users can directly use the various tools included in the large model tool chain. Among them, the large model tool chain includes a large model framework, a data set, and various preset models. For example, the large model framework can be PyTorch, vLLM, llama.cpp, etc.; the application program interface layer of the computing power package encapsulates different functions to adapt to different computing tasks, such as model training, reasoning, fine-tuning and other functions; the infrastructure software includes various software platforms to provide basic software services, such as the Alaya NeW platform, DingoStack, DingoFS, etc.; the infrastructure hardware includes the servers, GPU accelerator cards, storage systems and high-speed network hardware of the intelligent computing center.

[0064] It can be understood that by configuring the application program interface for the intelligent computing center in the above manner, the computing resources of the intelligent computing center can be quickly scheduled to process computing tasks through the application program interface. There is no need for users to configure the operating framework and set the operating environment parameters for each computing task, which can effectively improve the efficiency of computing resource scheduling and thus improve the execution efficiency of computing tasks.

[0065] In the present invention, when it is necessary to call computing power resources, determine the computing power package corresponding to the computing power task to be processed, the computing power task is a task that needs to call computing power resources for processing, and the computing power package is used to measure the computing power of the intelligent computing center according to the amount of calculation; based on the application program interface corresponding to the computing power package, the computing power resources are called, and the application program interface includes a code program corresponding to the operation of executing the computing power task, and the computing power resources called by the application program interface match the operating environment of the computing power task. In this way, when it is necessary to execute a computing power task, the computing power resources of the intelligent computing center can be quickly scheduled to process the computing power tasks through the application program interface of the intelligent computing center, and the user does not need to configure the operating framework and set the operating environment parameters for each computing power task, which effectively improves the scheduling efficiency of computing power resources, and then can effectively improve the execution efficiency of computing power tasks. At the same time, by measuring computing power according to the amount of calculation through the computing power package, the problem of high costs caused by business personnel leasing the computing power services of the intelligent computing center for a long time can be avoided, and the widespread application of inclusive computing power can be achieved.

[0066] In one embodiment, the computing power package corresponds to a plurality of application program interfaces, and step S2 includes: Step S21, determining the target task type corresponding to the computing task; Step S22: querying a target application program interface corresponding to the target task type based on a preset mapping table, wherein the preset mapping table includes a plurality of task types and the plurality of application program interfaces, and a mapping relationship between different task types and different application program interfaces, and the target application program interface is one of the plurality of application program interfaces; Step S23: calling computing resources based on the target application program interface.

[0067] The above-mentioned computing power package corresponds to multiple application programming interfaces. When the computing power package of the intelligent computing center is required to provide computing power resources for different computing power tasks, computing power resources can be scheduled for each computing power task through different application programming interfaces, thereby realizing the processing of different computing power tasks.

[0068] It should be noted that different computing tasks have different specific contents. For example, the operations performed when the computing task is a model training task are different from the operations performed when the computing task is a data fine-tuning task. If the same application program interface is used to adapt to different types of computing tasks, the application program interface needs to set more codes to improve compatibility, occupying more resources of the intelligent computing center, making it difficult to set more application program interfaces, and limiting the number of users of computing services that the intelligent computing center can provide.

[0069] Therefore, in this embodiment, different application programming interfaces are set for specific types of computing tasks, so that each application programming interface can provide rapid scheduling of computing resource services of the intelligent computing center while occupying fewer resources, thereby increasing the number of users of computing services that the intelligent computing center can provide.

[0070] Specifically, firstly, the computing tasks are divided into different task types according to the specific content of the computing tasks; different application program interfaces are created, each of which is created for computing tasks of a task type. Through the application program interfaces, the computing resources of the intelligent computing center can be scheduled to execute computing tasks; a mapping relationship is established between each application program interface and different task types, and a preset mapping table is created based on the mapping relationship. After the preset mapping table is created, the target application program interface can be determined according to the task type of the computing task.

[0071] The above-mentioned determination of the target task type corresponding to the computing task (step S21) may specifically be that the user sets the task type corresponding to the computing task when uploading the computing task; or, the specific content of the computing task is identified to determine the task type corresponding to the computing task.

[0072] In the present invention, the target task type corresponding to the computing task is determined; the target application program interface corresponding to the target task type is queried based on a preset mapping table, the preset mapping table includes multiple task types and the multiple application program interfaces, and the mapping relationship between different task types and different application program interfaces, the target application program interface is one of the multiple application program interfaces; the computing resource is called based on the target application program interface. In this way, the target application program interface is determined through the target task type of the computing task, and then the scheduling efficiency of the computing resource is effectively improved through the target application program interface, and then the execution efficiency of the computing task is effectively improved.

[0073] In one embodiment, the multiple task types include at least one of a model training task type, an inference task type, and a data fine-tuning task type, and each task type corresponds to at least one application programming interface.

[0074] In the present invention, multiple task types include at least one of a model training task type, an inference task type, and a data fine-tuning task type. Each task type corresponds to at least one application program interface. In this way, at least one application program interface is set for different task types, so that the intelligent computing center can simultaneously provide computing power resources for computing power tasks of different task types, and provide computing power resources for multiple computing power tasks of the same task type.

[0075] It should be noted that in the field of intelligent computing centers, intelligent computing centers can provide a large amount of computing resources to perform computing tasks. In the case of insufficient computing power in the intelligent computing center, computing resource services will no longer be provided to new users. For users who have already provided computing services, there are changes in computing resource consumption. For example, more computing resources are consumed during the running of the program, and less computing resources are consumed when the program is adjusted, resulting in the situation where the intelligent computing center has redundant computing resources but cannot provide computing services to more users.

[0076] In order to provide computing power services of the intelligent computing center to more users, the computing power provided by the intelligent computing center is elastically scaled in the present invention, so that the intelligent computing center can provide computing power resources to more users.

[0077] Specifically, in one embodiment, after step S2, the method further includes: Step S3: Processing the computing task based on the computing resources called by the target application program interface; The computing task is a computing task processed by the target container, and step S3 includes at least one of the following: Step S31: when the load data is greater than or equal to a first preset threshold, expand the target container based on the target application program interface, and the load data is the load data of the target container; Step S32: When the load data is less than or equal to a second preset threshold, shrink the target container based on the target application program interface.

[0078] The above-mentioned first preset threshold is used to measure whether the load data of the target container is too large. When the load data is greater than or equal to the first preset threshold, the load of the target container is large. In order to meet the execution of the computing power task in the target container, it is necessary to further expand the capacity of the target container to ensure that the computing power task is executed through sufficient computing power resources.

[0079] The above-mentioned second preset threshold is used to measure whether the load data of the target container is too small. When the load data is less than or equal to the second preset threshold, the load of the target container is small and there is an excess of computing resources. In order to improve the utilization rate of computing resources of the intelligent computing center and reduce the waste of computing resources, it is necessary to reduce the capacity of the target container and release more spare computing power so that the intelligent computing center can provide computing services to more users, thereby further realizing the widespread application of inclusive computing power.

[0080] Among them, the first preset threshold is greater than the second preset threshold. When the load data of the target container is between the first preset threshold and the second preset threshold, the computing power resources occupied by the target container are at a relatively stable level, and there is no need to expand or shrink the target container. This avoids the situation where the target container needs to be frequently expanded or reduced due to setting the first preset threshold and the second preset threshold to the same, so that the computing power resources of the intelligent computing center are in a relatively stable state.

[0081] Furthermore, when the target container needs to be expanded or reduced in capacity, it can be directly implemented through the target application program interface corresponding to the computing power package, realizing the automatic scheduling of computing power resources without the need for user settings or adjustments.

[0082] In the present invention, when the load data is greater than or equal to the first preset threshold, the target container is expanded based on the target application interface, and the load data is the load data of the target container; when the load data is less than or equal to the second preset threshold, the target container is reduced based on the target application interface. In this way, the target container is expanded or reduced based on the load data of the target container, so that the target container can occupy less computing resources of the intelligent computing center while ensuring sufficient computing resources to execute computing tasks, so that the intelligent computing center can provide computing services to more users, and further realize the widespread application of inclusive computing power.

[0083] In some embodiments, the computing resources of the intelligent computing center include local computing resources and remote computing resources. The way in which the intelligent computing center schedules computing resources in different areas for expansion and reduction is also different. Parameters such as the network and the remaining computing power of different servers need to be considered so that the intelligent computing center can provide lower latency and sufficient computing resources.

[0084] Among them, first determine whether the local computing resources are sufficient, and then dispatch computing resources in other places when local resources are insufficient.

[0085] Specifically, in one embodiment, step S31 includes: Step S311: When the load data is greater than or equal to the first preset threshold, obtain first computing resource information of a first server, where the first server is the server to which the target container belongs; Step S312: When the first computing power resource information satisfies the load data, create a first container based on the computing power resources of the first server through the target application program interface; Step S313: deploy the relevant data of the computing task to the first container through the target application program interface; Step S314: Execute the computing task based on the first container.

[0086] The above-mentioned first server is the server to which the target container belongs, that is, when the target container needs to be expanded, priority is given to scheduling local computing resources. When the local computing resources meet the scheduling requirements, the local computing resources are directly scheduled; when the local computing resources do not meet the scheduling requirements, remote computing resources are then scheduled.

[0087] Specifically, when capacity expansion is required, the first computing resource information of the first server is obtained, and the first computing resource information is used to characterize the remaining computing resources of the first server. When the first computing resource information satisfies the load data, it is considered that the first server meets the capacity expansion requirements of the target container, and the computing resources of the first server are directly scheduled for capacity expansion.

[0088] Among them, the first computing power resource information satisfies the load data, which may be that the first computing power resource information includes the remaining computing power of the first server, and the load data includes the required computing power. When the remaining computing power is greater than the required computing power, the first computing power resource information satisfies the load data.

[0089] The relevant data of the computing task specifically includes at least one of the environmental parameters of the computing task, the data to be processed, the model, the operating parameters, and the application program. The relevant data of the computing task is deployed to the first container, specifically, at least one of the environmental parameters of the computing task, the data to be processed, the model, the operating parameters, and the application program is deployed to the first container, so that the first container can process the computing task according to the deployed relevant data.

[0090] In the present invention, when the load data is greater than or equal to the first preset threshold, the first computing resource information of the first server is obtained, and the first server is the server to which the target container belongs; when the first computing resource information satisfies the load data, the first container is created based on the computing resources of the first server through the target application interface; the relevant data of the computing task is deployed to the first container through the target application interface; and the computing task is executed based on the first container. In this way, the target container is expanded through the computing resources of the first server, the corresponding computing service is quickly launched, and the computing process is diverted to the new machine, thereby realizing efficient execution of computing tasks.

[0091] When local computing resources are insufficient, the intelligent computing center needs to dispatch computing resources from other places to process computing tasks. Specifically, in one embodiment, step S31 further includes: Step S315: When the first computing power resource information does not satisfy the load data, obtain network information and second computing power resource information of a second server, where the second server and the first server are in the same area or in different areas; Step S316: When the second computing power resource information satisfies the load data and the network information satisfies the preset network condition, create a second container based on the computing power resources of the second server; Step S317: deploy the relevant data of the computing task to the second container; Step S318: Execute the computing task based on the second container.

[0092] The above-mentioned second server is a server other than the first server. When the computing resources of the first server are insufficient, the intelligent computing center needs to dispatch computing resources through the second server in a different location to process computing tasks.

[0093] In some embodiments, the second server may be a server located in the same region as the first server. It should be noted that the second server and the first server are servers in the same region, and the hardware equipment of the second server and the hardware equipment of the first server may be located in the same or adjacent computer room, that is, the physical distance between the second server and the first server is relatively close. In this case, the latency between the second server and the first server is relatively low, and the computing resources of the second server can be preferentially called to expand the target container.

[0094] In some embodiments, the second server may be a server located in a different region from the first server. It should be noted that the second server and the first server are servers in different regions, and the hardware equipment of the second server and the hardware equipment of the first server may be located in computer rooms that are far away from each other. For example, the hardware equipment of the first server is located in City A, and the hardware equipment of the second server is located in City B. In this case, the latency between the second server and the first server is relatively high, and it can be used as a backup method for calling computing resources remotely.

[0095] Specifically, the step S315 is as follows: When the first computing power resource information does not satisfy the load data, obtaining computing power resource information and network information of the second server in the same area; When the computing power resource information of the second server in the same area does not satisfy the load data, or when the network information of the second server in the same area satisfies the preset network condition, acquiring the computing power resource information and network information of the second server in a different area; The step S316 is specifically as follows: When the computing power resource information of the second server in the same area meets the load data and the network information of the second server in the same area meets the preset network condition, creating the second container based on the computing power resources of the second server in the same area; When the computing power resource information of the second servers in different areas meets the load data and the network information of the second servers in different areas meets the preset network conditions, the second container is created based on the computing power resources of the second servers in different areas.

[0096] The preset network condition is used to measure whether the network information of the second server meets the requirements for scheduling computing resources with low latency. The preset network condition may include at least one of a latency threshold, a bandwidth threshold, and a connectivity parameter threshold, and the network status of the second server is determined by at least one of latency, bandwidth, and connectivity parameters.

[0097] In some implementations, when the second server and the first server are in the same region, obtaining network information of all servers in the first region, the first region being the region where the first server is located; When the network information of the second server is optimal among the network information of all servers, the second server meets the preset network condition.

[0098] In some implementations, when the second server and the first server are in different regions, obtaining at least one of a latency, a bandwidth, and a connectivity parameter of the second server; When the delay of the second server meets the delay threshold, and / or the bandwidth of the second server meets the bandwidth threshold, and / or the connectivity of the second server meets the connectivity threshold, the second server meets the preset network condition.

[0099] The above-mentioned deployment of the relevant data of the computing task to the second container (step S313) can synchronously deploy the relevant data of the computing task to the second container when the latency is low (the second server and the first server are in the same area), and needs to asynchronously deploy the relevant data of the computing task to the second container when the latency is high (the second server and the first server are in different areas).

[0100] Specifically, when the second server and the first server are in the same area, at least one of the environmental parameters, data to be processed, model, operating parameters, and application programs of the computing task is synchronously copied to the storage hardware of the second server, and then the computing device is selected to create a second container, and at least one of the environmental parameters, data to be processed, model, operating parameters, and application programs of the computing task is deployed to the second container, so that the second container can deploy related data processing computing tasks.

[0101] When the second server and the first server are in different areas, at least one of the environmental parameters, data to be processed, model, operating parameters, and application programs of the computing task is asynchronously copied to the storage hardware of the second server, and then the computing device is selected to create a second container, and at least one of the environmental parameters, data to be processed, model, operating parameters, and application programs of the computing task is deployed to the second container, so that the second container can deploy related data processing computing tasks.

[0102] In the present invention, when the first computing power resource information does not meet the load data, the network information and the second computing power resource information of the second server are obtained, and the second server and the first server are in the same area or different areas; when the second computing power resource information meets the load data and the network information meets the preset network conditions, the second container is created based on the computing power resources of the second server through the target application interface; the relevant data of the computing power task is deployed to the second container through the target application interface; and the computing power task is executed based on the second container. In this way, when the local computing power resources are insufficient, the intelligent computing center can dispatch the computing power resources of the second server to process the computing power task.

[0103] Furthermore, when the load data of the target container is low, the target container needs to be reduced in size to release the computing resources of the intelligent computing center so that the intelligent computing center can provide more computing resources to users.

[0104] It should be noted that, unlike expanding the capacity of the target container, when the target container needs to be reduced in capacity, it is not necessary to consider other servers, and only the computing resources of the first server where the target container is located need to be released.

[0105] In some implementations, when the load data is less than or equal to a second preset threshold, scaling down the target container based on the target application program interface may be to release computing resources of the target container at a preset ratio.

[0106] In some implementations, when the load data is less than or equal to a second preset threshold, scaling down the target container based on the target application program interface may be to release computing resources of the target container according to a preset computing resource value.

[0107] In one embodiment, after step S3, the method further includes: Step S4: Measure the computing resource consumption of the computing package based on the target container, the first container or the second container, where the target container, the first container and the second container are containers for executing the computing task.

[0108] It should be noted that in the process of the intelligent computing center executing computing tasks, the target container will be expanded or reduced in capacity, and the computing resources of the intelligent computing center will be consumed in the process of expansion and reduction. For example, the relevant data of the computing task is deployed to the first container, and the computing resources of the intelligent computing center need to be used in this process. However, since this process is not the execution of computing tasks, when measuring computing power consumption, only the computing power resources consumed by the target container, the first container or the second container to execute the computing task are measured to improve the accuracy of the measurement.

[0109] In the present invention, the computing power resource consumption of the computing power package is measured based on the target container, the first container or the second container. The target container, the first container and the second container are containers for executing computing power tasks, which can improve the measurement accuracy of computing power resources, thereby reducing the user's usage cost and further realizing the widespread application of inclusive computing power.

[0110] Specifically, the computing resource consumption of the computing package measured based on the target container, the first container, or the second container includes: Obtaining a target ratio of processor resources occupied by the target container, the first container, or the second container, and a total time of occupying a processor, where the processor is a processor of an intelligent computing center; Obtaining a computing power value of the processor, where the computing power value is used to represent computing power resources provided by the processor per unit time; The computing power resource consumption is calculated based on the target ratio, the total time and the computing power value.

[0111] For example, the computing power value of the processor is 3 degrees / hour, the computing task is executed by the target container for a total of 2 hours, and the target proportion of the processor resources occupied during the execution of the computing task is 1 / 3. In this case, the computing power value of the processor, the total time occupied by the processor, and the target proportion are directly multiplied to obtain the computing power value of the computing package consumed by the execution of the computing task is 2 degrees. Among them, "degree" is the unit computing resource.

[0112] Furthermore, when a computing task is executed by multiple containers, it is necessary to sum the computing resources consumed by different containers to obtain the total computing resource consumption for executing the computing task.

[0113] In some implementations, obtaining a target ratio of processor resources occupied by the target container, the first container, or the second container includes: Obtain at least one of a first half-precision floating-point computing power, a first video memory bandwidth, and a first video memory capacity of the processor occupied by the target container, the first container, or the second container; The target ratio is calculated based on at least one of the first half-precision floating-point computing power, the first video memory bandwidth, and the first video memory capacity.

[0114] In the present invention, the target ratio is calculated by using at least one of the first half-precision floating-point computing power number, the first video memory bandwidth and the first video memory capacity, thereby improving the accuracy of the target ratio, thereby improving the accuracy of the calculated computing power consumption, reducing the user's usage cost, and realizing the widespread application of inclusive computing power.

[0115] In some implementations, calculating the target ratio based on at least one of the first half-precision floating-point computing power, the first video memory bandwidth, and the first video memory capacity includes: Obtaining at least one of a preset half-precision floating-point computing power, a preset video memory bandwidth, and a preset video memory capacity of the processor; Calculating a first ratio of the first half-precision floating-point computing power to the preset half-precision floating-point computing power, and / or calculating a second ratio of the first video memory bandwidth to the preset video memory bandwidth, and / or calculating a third ratio of the first video memory capacity to the preset video memory capacity; The target ratio is calculated based on the first ratio, the second ratio, and the third ratio.

[0116] Among them, when only one of the first half-precision floating-point computing power, the first video memory bandwidth and the first video memory capacity is obtained, the corresponding first ratio, second ratio or third ratio is directly set as the target ratio.

[0117] When at least two of the first half-precision floating-point computing power, the first video memory bandwidth, and the first video memory capacity are obtained, at least two of the corresponding first ratio, second ratio, and third ratio are weighted to obtain the target ratio.

[0118] See also Figure 3 , Figure 3 It is a structural diagram of a device for providing computing power resources through a computing power package provided by an intelligent computing center of the present invention, such as Figure 3 As shown, the device 300 for providing computing resources through a computing power package in an intelligent computing center includes: The determination module 301 is used to determine the computing power package corresponding to the computing power task to be processed when computing power resources need to be called. The computing power task is a task that needs to call computing power resources for processing. The computing power package is used to measure the computing power of the intelligent computing center according to the computing amount; The calling module 302 is used to call computing power resources based on the application program interface corresponding to the computing power package. The application program interface includes a code program corresponding to the operation of executing the computing power task. The computing power resources called by the application program interface match the operating environment of the computing power task.

[0119] In one embodiment, the computing power package corresponds to multiple application programming interfaces, and the calling module 302 includes: A determination unit, used to determine a target task type corresponding to the computing task; a query unit, configured to query a target application program interface corresponding to the target task type based on a preset mapping table, wherein the preset mapping table includes a plurality of task types and the plurality of application program interfaces, and a mapping relationship between different task types and different application program interfaces, and the target application program interface is one of the plurality of application program interfaces; The calling unit calls computing resources based on the target application program interface.

[0120] In one embodiment, the multiple task types include at least one of a model training task type, an inference task type, and a data fine-tuning task type, and each task type corresponds to at least one application programming interface.

[0121] In one embodiment, after the calling module 302, the intelligent computing center provides computing resources through the computing power package 300 further includes: A processing module, applying computing resources based on the target application program interface call to process the computing task; The computing task is a computing task processed by the target container, and the processing module includes at least one of the following: an expansion submodule, configured to expand the target container based on the target application program interface when the load data is greater than or equal to a first preset threshold, the load data being the load data of the target container; The shrinking submodule is used to shrink the target container based on the target application program interface when the load data is less than or equal to a second preset threshold.

[0122] In one embodiment, the expansion submodule includes: A first acquisition unit, configured to acquire first computing resource information of a first server when the load data is greater than or equal to the first preset threshold, the first server being a server to which the target container belongs; A first creation unit, configured to create a first container based on the computing power resources of the first server through the target application program interface when the first computing power resource information satisfies the load data; A first deployment unit, configured to deploy the relevant data of the computing task to the first container through the target application program interface; A first execution unit is used to execute the computing task based on the first container.

[0123] In one embodiment, the expansion submodule further includes: A second acquisition unit is used to acquire network information and second computing resource information of a second server when the first computing resource information does not satisfy the load data, where the second server and the first server are in the same area or in different areas; A second creation unit, configured to create a second container based on the computing resources of the second server when the second computing resource information satisfies the load data and the network information satisfies a preset network condition; A second deployment unit, configured to deploy the relevant data of the computing task to the second container; A second execution unit is used to execute the computing task based on the second container.

[0124] In one embodiment, after the processing module, the device 300 for providing computing resources through a computing power package by the intelligent computing center further includes: A metering module is used to measure the computing resource consumption of the computing package based on a target container, a first container or a second container, where the target container, the first container and the second container are containers for executing the computing task.

[0125] The device for providing computing power resources through a computing power package provided by the intelligent computing center of the present invention is capable of realizing the various processes of each embodiment of the method for providing computing power resources through a computing power package of the above-mentioned intelligent computing center. The technical features correspond one to one and can achieve the same technical effect. To avoid repetition, they will not be described here.

[0126] It should be noted that the device for providing computing resources through a computing power package in the intelligent computing center of the present invention can be a device, or a component, integrated circuit, or chip in an electronic device.

[0127] The present invention also provides an electronic device, see Figure 4 , Figure 4 4 is a schematic diagram of a structure of an electronic device provided by the present invention, the electronic device includes a memory 401, a processor 402 and a program or instruction stored in the memory 401 and running thereon, and when the program or instruction is executed by the processor 402, it can be realized Figure 1 Any steps in the corresponding method embodiment of the intelligent computing center providing computing resources through computing power packages and achieving the same beneficial effects will not be repeated here.

[0128] The processor 402 may be a CPU, an ASIC, an FPGA or a GPU.

[0129] A person skilled in the art can understand that all or part of the steps of the method embodiment for implementing the above-mentioned intelligent computing center to provide computing resources through computing power packages can be completed by hardware related to program instructions, and the program can be stored in a readable medium.

[0130] The present invention also provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above Figure 1The corresponding intelligent computing center provides computing resources through computing power packages. Any steps in the embodiment can achieve the same technical effect. To avoid repetition, they are not repeated here. The storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0131] The present invention also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the above Figure 1 The corresponding intelligent computing center provides computing resources through computing power packages. Each process of the method implementation example can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0132] The terms "first", "second", etc. in the present invention are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. In addition, the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, 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 that are not clearly listed or inherent to these processes, methods, products or devices. In addition, "and / or" is used in the present invention to represent at least one of the connected objects, such as A and / or B and / or C, which means including 7 situations including single A, single B, single C, and both A and B exist, both B and C exist, both A and C exist, and both A, B and C exist.

[0133] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0134] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a second terminal device, etc.) to execute the methods of various embodiments of the present invention.

[0135] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can make many forms of expansion without departing from the scope of protection of the purpose of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for an intelligent computing center to provide computing resources through a computing package, characterized in that: include: Step S1: when computing power resources need to be called, determine the computing power package corresponding to the computing power task to be processed, where the computing power task is a task that needs to be processed by calling computing power resources, and the computing power package is used to measure the computing power of the intelligent computing center according to the computing amount; Step S2: Call computing resources based on the application program interface corresponding to the computing package, where the application program interface includes a code program corresponding to the operation of executing the computing task, and the computing resources called by the application program interface match the operating environment of the computing task.

2. The method according to claim 1, characterized in that The computing power package corresponds to a plurality of application program interfaces, and step S2 includes: Step S21, determining the target task type corresponding to the computing task; Step S22: querying a target application program interface corresponding to the target task type based on a preset mapping table, wherein the preset mapping table includes a plurality of task types and the plurality of application program interfaces, and a mapping relationship between different task types and different application program interfaces, and the target application program interface is one of the plurality of application program interfaces; Step S23: calling computing resources based on the target application program interface.

3. The method according to claim 2, characterized in that The multiple task types include at least one of a model training task type, an inference task type, and a data fine-tuning task type, and each task type corresponds to at least one application programming interface.

4. The method according to claim 2, characterized in that After step S2, the method further includes: Step S3: Processing the computing task based on the computing resources called by the target application program interface; The computing task is a computing task processed by the target container, and step S3 includes at least one of the following: Step S31: when the load data is greater than or equal to a first preset threshold, expand the target container based on the target application program interface, and the load data is the load data of the target container; Step S32: When the load data is less than or equal to a second preset threshold, shrink the target container based on the target application program interface.

5. The method according to claim 4, characterized in that The step S31 comprises: Step S311: When the load data is greater than or equal to the first preset threshold, obtain first computing resource information of a first server, where the first server is the server to which the target container belongs; Step S312: When the first computing power resource information satisfies the load data, create a first container based on the computing power resources of the first server through the target application program interface; Step S313: deploy the relevant data of the computing task to the first container through the target application program interface; Step S314: Execute the computing task based on the first container.

6. The method according to claim 5, characterized in that The step S31 further includes: Step S315: When the first computing power resource information does not satisfy the load data, obtain network information and second computing power resource information of a second server, where the second server and the first server are in the same area or in different areas; Step S316: When the second computing power resource information satisfies the load data and the network information satisfies the preset network condition, create a second container based on the computing power resources of the second server; Step S317: deploy the relevant data of the computing task to the second container; Step S318: Execute the computing task based on the second container.

7. The method according to any one of claims 4 to 6, characterized in that After step S3, the method further includes: Step S4: Measure the computing resource consumption of the computing package based on the target container, the first container or the second container, where the target container, the first container and the second container are containers for executing the computing task.

8. A device for providing computing resources through computing packages in an intelligent computing center, characterized in that: include: A determination module, used to determine a computing power package corresponding to a computing power task to be processed when computing power resources need to be called, wherein the computing power task is a task that needs to be processed by calling computing power resources, and the computing power package is used to measure the computing power of the intelligent computing center according to the computing amount; A calling module is used to call computing power resources based on the application program interface corresponding to the computing power package. The application program interface includes a code program corresponding to the operation of executing the computing power task. The computing power resources called by the application program interface match the operating environment of the computing power task.

9. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of the method for providing computing resources through a computing power package by an intelligent computing center as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for an intelligent computing center to provide computing resources through a computing power package as described in any one of claims 1 to 7.

11. A computer program product, characterized in that It includes computer instructions, which, when executed by a processor, implement the steps of the method for an intelligent computing center to provide computing resources through a computing power package as described in any one of claims 1 to 7.

Images