Virtual computing power pre-distribution method and device for intelligent computing center cloud platform

Abstract

The invention provides a virtual computing power pre-distribution method and device for an intelligent computing center cloud platform, and relates to the technical field of intelligent computing centers, intelligent computing centers, computing power infrastructures and intelligent computing clouds, and the method comprises the steps: S1, receiving computing power resource application information which comprises type information of computing power operation tasks, calculating power resource demand information of the calculating power operation task; s2, under the condition that the type information meets a preset condition, creating a running environment of the container based on a preset computing power resource quantity; s3, under the condition that the computing power operation task is received, a first number of virtual acceleration cards are distributed to the container based on the type information and the computing power resource demand information, the virtual acceleration cards are virtual acceleration cards of the intelligent computing center cloud platform, and the virtual acceleration cards are used for providing computing power resources; and S4, executing the computing power operation task through the container based on the operation environment and the first number of virtual acceleration cards. The computing power resource utilization rate is greatly improved.

Description

Technical Field

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

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

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

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

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

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

[0007] In the process of developing model code in an intelligent computing center, it is necessary to first create a container for running the model code and configure the running environment, and then import the model code into the container to facilitate users to develop the model code. In the prior art, after the container is created and the running environment is configured, there is a situation where users do not actually use computing power resources when developing the model code. At this time, the container occupies the computing power resources of the intelligent computing center but does not actually work, resulting in a very low utilization rate of the computing power resources of the intelligent computing center.

[0008] It can be seen that there is a problem of very low utilization rate of the computing power resources of the intelligent computing center in the prior art. Summary of the Invention

[0009] The present invention provides a method and device for virtual computing power pre-allocation of an intelligent computing center cloud platform to solve the problem of very low utilization rate of the computing power resources of the intelligent computing center in the prior art.

[0010] To solve the above problems, the present invention is implemented as follows: In a first aspect, the present invention provides a method for pre - allocating virtual computing power in an intelligent computing center cloud platform, including: Step S1, receiving computing power resource application information, where the computing power resource application information includes type information of a computing power operation task and computing power resource requirement information of the computing power operation task; Step S2, when the type information meets a preset condition, creating a running environment for a container based on a preset amount of computing power resources; Step S3, when receiving the computing power operation task, allocating a first number of virtual acceleration cards to the container based on the type information and the computing power resource requirement information, where the virtual acceleration cards are virtual acceleration cards of the intelligent computing center cloud platform and are used to provide computing power resources; Step S4, executing the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards.

[0011] In one embodiment, the preset condition includes one of the following: The type information indicates that the computing power operation task is a long - term type task; The type information indicates that the computing power operation task is a step - growth type task; The type information indicates that the computing power operation task is a spare - time release type task.

[0012] In one embodiment, step S3 includes: Step S31, creating a virtual mapping relationship table based on the computing power resource requirement information, where the virtual mapping relationship table is the relationship between the computing power operation task and N virtual acceleration cards, and the computing power resources provided by the N virtual acceleration cards match the computing power resource requirement information, and N is a positive integer greater than or equal to 1; Step S32, when receiving the computing power operation task, determining the first number of virtual acceleration cards based on the type information and the computing power resource requirement information; Step S33, allocating the first number of virtual acceleration cards among the N virtual acceleration cards to the container based on the virtual mapping relationship table, where the first number of virtual acceleration cards are the virtual acceleration cards in an idle state in the virtual mapping relationship table.

[0013] In one embodiment, step S32 includes: [[ID= Step S322, when the type information indicates that the computing power operation task is a step - growth type task, determine that the first quantity of virtual acceleration cards is M virtual acceleration cards, where M is a positive integer greater than or equal to 1 and less than or equal to N.

[0014] In one embodiment, the method further includes: Step S34, set the first quantity of virtual acceleration cards in the virtual mapping relation table to the occupied state.

[0015] In one embodiment, the step S4 includes: Step S41, when the type information indicates that the computing power operation task is a step - growth type task, obtain the load parameter of the container; Step S42, when the load parameter of the container matches the computing power resources provided by the first quantity of virtual acceleration cards, execute the computing power operation task through the container based on the running environment and the first quantity of virtual acceleration cards.

[0016] In one embodiment, the method further includes: Step S43, when the load parameter of the container does not match the computing power resources provided by the first quantity of virtual acceleration cards, allocate a second quantity of virtual acceleration cards for the container based on the computing power resource demand information, where the second quantity is greater than the first quantity, and the computing power resources provided by the second quantity of virtual acceleration cards are less than or equal to the quantity of the required computing power resources characterized by the computing power resource demand information; Step S44, execute the computing power operation task through the container based on the running environment and the second quantity of virtual acceleration cards; Step S45, set the second quantity of virtual acceleration cards in the virtual mapping relation table to the occupied state.

[0017] In one embodiment, the method further includes: Step S46, obtain the usage parameter of the second quantity of virtual acceleration cards; Step S47, when there is a first virtual acceleration card among the second quantity of virtual acceleration cards whose usage parameter indicates no usage within the first time period, release the computing power resources provided by the first virtual acceleration card, and set the first virtual acceleration card in the virtual mapping relation table to the idle state.

[0018] In one embodiment, the step S3 includes: Step S31’, when the type information indicates that the computing power operation task is an idle - time release type task, monitor the first quantity of virtual acceleration cards; Step S32’: When it is detected that a virtual acceleration card among the first number of acceleration cards is called, allocate the first number of virtual acceleration cards to the container.

[0019] In one embodiment, the method further includes: Step S5: When the type information indicates that the computing power operation task is a type of task for releasing during idle time, monitor the first number of virtual acceleration cards; Step S6: When it is detected that the first number of virtual acceleration cards are not used within a second time period, release the container and set the first number of virtual acceleration cards in the virtual mapping relation table to the idle state.

[0020] In one embodiment, step S6 includes: Step S61: When it is detected that the first number of virtual acceleration cards are not used within a second time period, save the data in the memory of the first number of virtual acceleration cards to the storage space; Step S62: Release the container and set the first number of virtual acceleration cards in the virtual mapping relation table to the idle state.

[0021] In a second aspect, the present invention further provides a virtual computing power pre-allocation device for an intelligent computing center cloud platform, including: A receiving module, configured to receive computing power resource application information, where the computing power resource application information includes type information of a computing power operation task and computing power resource requirement information of the computing power operation task; A creating module, configured to create a running environment for a container based on a preset computing power resource amount when the type information meets a preset condition; An allocating module, configured to allocate a first number of virtual acceleration cards to the container based on the type information and the computing power resource requirement information when receiving the computing power operation task, where the virtual acceleration card is a virtual acceleration card of the intelligent computing center cloud platform, and the virtual acceleration card is used to provide computing power resources; An executing module, configured to execute the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards.

[0022] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor, where when the computer program is executed by the processor, the steps in the virtual computing power pre-allocation method for the intelligent computing center cloud platform as described in the first aspect above are implemented.

[0023] Fourthly, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the virtual computing power pre-allocation method of the intelligent computing center cloud platform as described in the first aspect above are implemented.

[0024] Fifthly, the present invention further provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps in the virtual computing power pre-allocation method of the intelligent computing center cloud platform as described in the first aspect above are implemented.

[0025] In the present invention, step S1: Receive computing power resource application information, where the computing power resource application information includes the type information of the computing power operation task and the computing power resource requirement information of the computing power operation task; step S2: Based on a preset amount of computing power resources, create a running environment for the container when the type information meets the preset conditions; step S3: When receiving the computing power operation task, allocate a first number of virtual acceleration cards to the container based on the type information and the computing power resource requirement information, where the virtual acceleration card is a virtual acceleration card of the intelligent computing center cloud platform, and the virtual acceleration card is used to provide computing power resources; step S4: Execute the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards. In this way, creating a running environment for the container based on a preset amount of computing power resources avoids the waste of computing power resources caused by the container occupying computing power resources when the computing power operation task is not executed, and greatly improves the utilization rate of computing power resources; at the same time, by executing the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards, it is avoided that the container is allocated too many virtual acceleration cards, resulting in the waste of computing power resources, and greatly improves the utilization rate of computing power resources. Description of the Drawings

[0026] To more clearly illustrate the technical solutions of the present invention, the drawings required for the description of the present invention will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0027] Figure 1 is a flowchart of a virtual computing power pre-allocation method for an intelligent computing center cloud platform provided by the present invention; Figure 2 is a schematic diagram of the pre-allocation of virtual acceleration cards provided by the present invention; Figure 3 is a structural diagram of a virtual computing power pre-allocation device for an intelligent computing center cloud platform provided by the present invention; Figure 4 is a structural diagram of an electronic device provided by the present invention. Detailed implementation mode

[0028] The technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] The "computing power" referred to in the present invention means: the ability of a computer device or a computing / data center to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to achieve the output of a target result by processing information data, and a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, which mainly provides services to society through computing power infrastructure.

[0030] The "computational power" (Computational Power, CP) referred to in the present invention means: the ability of a data center server to process data and achieve result output, a comprehensive indicator for measuring the computing ability of a data center, including general computing ability, supercomputing ability, and intelligent computing ability. The commonly used measurement unit is the number of floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), and the larger the value, the stronger the comprehensive computing ability. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 + CP 智能 + CP 超级 .

[0031] The "carrying capacity" (Network Power, NP) referred to in the present invention means: the performance of the data transmission ability of computing power facilities, a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., involving network transmission inside and between data centers, and a comprehensive indicator for measuring network transmission scheduling ability.

[0032] The "Storage Power" (SP) described in the present invention refers to: the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon. It is a comprehensive indicator to measure the data storage capacity of a data center, including external storage devices such as storage arrays and built-in storage devices in servers. The commonly used measurement unit for storage capacity is exabyte (EB, 1EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read / write operations per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0033] The "computing power infrastructure" described in the present invention refers to: a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage power, and can realize the centralized computing, storage, transmission, and application of information.

[0034] The "new type of information infrastructure" described in the present invention mainly includes network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, and satellite Internet, computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, and supercomputing centers, and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0035] The "computing power" described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0036] The "general computing power" described in the present invention refers to: the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0037] The "intelligent computing power" described in the present invention refers to: for various artificial intelligence innovation applications, a computing platform is 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), such as natural language processing, machine vision, and so on.

[0038] The "super computing power" described in the present invention mainly refers to: the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system. It is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, gene analysis, etc.

[0039] The "Intelligent Computing Center" described in the present invention refers to a facility that provides the required computing power, data, and algorithms mainly for artificial intelligence applications (such as scenarios like artificial intelligence deep learning model development, model training, and model inference) by using large-scale heterogeneous computing power 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 underlying computing power to top-level application enabling.

[0040] The "Intelligent Computing Center Cloud Platform" described in the present invention, abbreviated as "Intelligent Computing Cloud", refers to a cloud computing platform that comprehensively serves based on the hardware resources and software resources of the intelligent computing center.

[0041] The "Intelligent Computing Center" described in the present invention includes, but is not limited to, the "Intelligent Computing Center".

[0042] The "Intelligent Computing Center" described in the present invention, namely the artificial intelligence computing center, is a type of computing power infrastructure that provides computing power services, data services, and algorithm services required for artificial intelligence applications based on artificial intelligence theory and adopting an artificial intelligence computing architecture.

[0043] The "Computing Power Center" described in the present invention refers to a facility mainly composed of infrastructure such as wind, fire, water, and electricity and IT software and hardware devices, and having computing power, transportation power, and storage power, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0044] The "Supercomputing Center" described in the present invention, namely the supercomputing data center, is a data center based on supercomputers or large-scale computing clusters, 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.

[0045] The "Computing Power Resources" described in the present invention refers to technologies and facilities with information computing, transmission, storage, and application capabilities required for the development of the digital society, including but not limited to computing resources such as CPU and GPU, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and support and guarantee resources such as wind, fire, water, and electricity.

[0046] The "Model" described in the present invention includes, but is not limited to, the "Large Language Model" and the "Multimodal Large Model".

[0047] The "Large Language Model" described in the present invention refers to a large-scale language model (LLM), which is a language model with a relatively large number of parameters, aiming to understand and generate human language, trained with a large amount of text data, and can perform a wide range of tasks including text summarization, translation, sentiment analysis, etc.

[0048] The "Multimodal Large Models" referred to in the present invention means: a model trained by jointly combining multimodal information such as text, images, videos, and audio, including but not limited to multimodal large language models.

[0049] The "computing power operation task" referred to in the present invention means: a specific workload or job executed on computing power resources and requiring a certain amount of computing power support, usually involving scenarios such as complex data processing, numerical calculations, model training, or simulation.

[0050] The "virtual computing power" referred to in the present invention means the computing power resources provided by the virtual acceleration cards of the intelligent computing center cloud platform. The virtual acceleration cards can be obtained by splitting physical acceleration cards, and each virtual acceleration card provides a part of the computing power resources of the physical acceleration card. For example, a physical GPU acceleration card with a size of 80G is split into 4 virtual acceleration cards, and each virtual acceleration card can provide computing power resources with a size of 20G. Among them, the physical acceleration card is the acceleration card deployed on the intelligent computing center cloud platform, which can be a GPU acceleration card or a CPU acceleration card. Among them, existing containerization technologies and virtualization technologies can be used to containerize and virtualize physical acceleration cards to obtain virtual acceleration cards of a certain specification.

[0051] The "pre-allocation" referred to in the present invention means pre-establishing a virtual mapping relationship table between computing power operation tasks and virtual acceleration cards, and when a container needs to actually execute a computing power operation task, then virtually allocate the virtual acceleration cards to the container according to the virtual mapping relationship table. Through pre-allocation, it is possible to avoid the computing power resources being occupied when the container waits for the computing power operation task after creation, so as to greatly improve the utilization rate of computing power resources.

[0052] Please refer to Figure 1 , Figure 1 which is a flowchart of a method for pre-allocating virtual computing power of an intelligent computing center cloud platform provided by the present invention. As Figure 1 shown, it includes the following steps: Step S1: Receive computing power resource application information, where the computing power resource application information includes the type information of the computing power operation task and the computing power resource requirement information of the computing power operation task.

[0053] The above-mentioned computing power resource application information is the information sent by the user to the intelligent computing center cloud platform before invoking the computing power resources of the intelligent computing center cloud platform to execute the computing power operation task. Through the computing power resource application information, the intelligent computing center cloud platform can determine the maximum number of virtual acceleration cards required to provide computing power resources during the execution of the computing power operation task.

[0054] Among them, the computing power resource application information is the information corresponding to the computing power operation task, including the type information of the computing power operation task and the computing power resource demand information. The intelligent computing center cloud platform can determine the number of virtual acceleration cards allocated to the container based on the type information and computing power resource demand information.

[0055] The above type information represents the type of computing task. It should be noted that different types of computing tasks have different requirements for computing resources during execution. It is necessary to dynamically allocate computing resources for different types of computing tasks to reduce idle computing resources and improve computing resource utilization.

[0056] The computing power resource demand information represents the amount of computing power resources required to execute the computing power operation task. In some embodiments, the computing power resource demand information may include the number of virtual accelerator cards. In other embodiments, the computing power demand information may include the size of the required computing power resources. The intelligent computing center cloud platform may determine the number of virtual accelerator cards that are most required to execute the computing power operation task based on the size of the computing power resources.

[0057] Step S2: When the type information meets the preset conditions, create a container operating environment based on the preset computing power resources.

[0058] The above-mentioned preset conditions are used to determine whether the computing task requires pre-allocation, pre-configuring the operating environment with less computing resources. This operating environment is required for the container to execute the computing task. If the type information does not meet the preset conditions, pre-allocation is not performed. In this case, the container is directly created based on the computing resources corresponding to the computing resource requirement information and released after the computing task is completed. If the type information meets the preset conditions, the container's operating environment is created based on the preset computing resources.

[0059] It should be noted that the preset amount of computing power resources is a smaller amount of computing power resources that are pre-set. When creating the container's operating environment through the preset amount of computing power resources, there is no need to immediately allocate all the computing power resources corresponding to the computing power resource demand information. Instead, computing power resources are allocated when the container needs to execute computing power running tasks, thereby avoiding the waste of computing power resources caused by occupying computing power resources when the container does not execute computing power running tasks, and improving the utilization rate of computing power resources.

[0060] In some embodiments, the preset amount of computing power resources may be computing power resources provided by a CPU or a GPU. For example, the preset amount of computing power resources may be computing power resources provided by a 1GB CPU and 1GB memory. The image file is obtained from the image repository using the computing power resources provided by the 1GB CPU and loaded into the 1GB memory to create the container's operating environment.

[0061] In addition, in one embodiment, the preset condition includes one of the following: The type information indicates that the computing power operation task is a long-duration type task; The type information indicates that the computing power operation task is a step-growth type task; The type information indicates that the computing power operation task is an idle-time release type task.

[0062] For example, the type of the computing power operation task can be a long-duration type (i.e., the computing power operation task is a long-duration type task). This type of computing power operation task needs to occupy computing power resources for a long time. Therefore, when allocating computing power resources to the container executing this type of computing power operation task, sufficient virtual acceleration cards can be directly allocated to the container based on the computing power resource demand information.

[0063] Also for example, the type of the computing power operation task can be a step-growth type (i.e., the computing power operation task is a step-growth type task). This type of computing power operation task has a relatively high demand for computing power resources during some periods of the execution process and a relatively low demand during other periods. At this time, the container can be dynamically expanded or contracted according to the load of the container to avoid the container occupying too many computing power resources, thereby improving the utilization rate of computing power resources.

[0064] Also for example, the type of the computing power operation task can be an idle-time release type (the computing power operation task is an idle-time release type task). The execution process of this type of computing power operation task includes an execution phase and an idle phase. During the execution phase, the container executes the computing power operation task based on the computing power resources corresponding to the computing power resource demand information. During the idle phase, the container does not use any computing power resources. At this time, the computing power resources occupied by the container can be released, and these computing power resources can be allocated to other containers, thereby improving the utilization rate of computing power resources.

[0065] Step S3: When receiving the computing power operation task, allocate a first quantity of virtual acceleration cards to the container based on the type information and the computing power resource demand information. The virtual acceleration cards are virtual acceleration cards of the intelligent computing center cloud platform, and the virtual acceleration cards are used to provide computing power resources.

[0066] Step S4: The container executes the computing power operation task based on the operating environment and the first quantity of virtual acceleration cards.

[0067] It should be noted that different types of computing power operation tasks require different amounts of computing power resources at different times during execution. For different types of computing power operation tasks, it is necessary to first determine the first quantity of virtual acceleration cards allocated to the container, and then the container executes the computing power operation task based on the operating environment and the first quantity of virtual acceleration cards, avoiding waste of computing power resources caused by the container being allocated too many virtual acceleration cards, and greatly improving the utilization rate of computing power resources.

[0068] In the present invention, step S1: Receive computing power resource application information, where the computing power resource application information includes the type information of the computing power operation task and the computing power resource requirement information of the computing power operation task; step S2: Based on a preset amount of computing power resources, create an operating environment for the container when the type information meets the preset conditions; step S3: When receiving the computing power operation task, allocate the first quantity of virtual acceleration cards to the container based on the type information and the computing power resource requirement information, where the virtual acceleration cards are virtual acceleration cards of the intelligent computing center cloud platform and are used to provide computing power resources; step S4: The container executes the computing power operation task based on the operating environment and the first quantity of virtual acceleration cards. In this way, creating the operating environment of the container through the preset amount of computing power resources avoids waste of computing power resources caused by the container occupying computing power resources when not executing the computing power operation task, and greatly improves the utilization rate of computing power resources; at the same time, the container executes the computing power operation task based on the operating environment and the first quantity of virtual acceleration cards, avoiding waste of computing power resources caused by the container being allocated too many virtual acceleration cards, and greatly improving the utilization rate of computing power resources.

[0069] In one embodiment, step S3 includes: Step S31: Create a virtual mapping relationship table based on the computing power resource requirement information, where the virtual mapping relationship table is the relationship between the computing power operation task and N virtual acceleration cards, and the computing power resources provided by the N virtual acceleration cards match the computing power resource requirement information, and N is a positive integer greater than or equal to 1; Step S32: When receiving the computing power operation task, determine the first quantity of virtual acceleration cards based on the type information and the computing power resource requirement information; Step S33: Allocate the first quantity of virtual acceleration cards among the N virtual acceleration cards to the container based on the virtual mapping relationship table, where the first quantity of virtual acceleration cards are the virtual acceleration cards in the idle state in the virtual mapping relationship table.

[0070] In the present invention, in step S31, a virtual mapping relation table is created based on the computing power resource demand information. The virtual mapping relation table is a relation table between the computing power operation tasks and N virtual accelerator cards. The computing power resources provided by the N virtual accelerator cards match the computing power resource demand information, and N is a positive integer greater than or equal to 1. In step S32, when the computing power operation task is received, the first number of virtual accelerator cards is determined based on the type information and the computing power resource demand information. The first number of virtual accelerator cards among the N virtual accelerator cards is allocated to the container based on the virtual mapping relation table. The first number of virtual accelerator cards is the virtual accelerator cards in the virtual mapping relation table that are in an idle state. In this way, the first number of virtual accelerator cards is obtained from the N virtual accelerator cards through the virtual mapping relation table and allocated to the container, so that the container can execute the computing power operation task based on the allocated first number of virtual accelerator cards.

[0071] The N virtual accelerator cards in the above virtual mapping relation table are the maximum number of virtual accelerator cards required during the execution of the computing power operation task. By establishing a virtual mapping relation table between the computing power execution task and the N virtual accelerator cards, some or all of the N virtual accelerator cards can be allocated to the container executing the computing power operation task.

[0072] It should be noted that the N virtual accelerator cards can correspond to one or more physical accelerator cards.

[0073] Further, as Figure 2 shown, during the process of creating a virtual mapping relation table for the computing power operation task and the N virtual accelerator cards, since there are idle stages during the execution of different computing power operation tasks and the virtual accelerator cards are not actually occupied during the idle stages, a virtual accelerator card can be pre-allocated to different computing power operation tasks at the same time, so as to provide computing power services for more users. Among them, as Figure 2 shown, for the computing power operation task 1, the pre-allocated virtual accelerator cards are virtual accelerator card 1, virtual accelerator card 2, and virtual accelerator card 3; for the computing power operation task 2, the pre-allocated virtual accelerator cards are virtual accelerator card 3, virtual accelerator card 4, and virtual accelerator card 5. Among them, virtual accelerator card 3 is pre-allocated to two computing power operation tasks at the same time, and the containers corresponding to the two computing power operation tasks can both occupy virtual accelerator card 3 to execute the computing power operation task.

[0074] For example, the intelligent computing center cloud platform is equipped with 100 virtual accelerator cards. By pre-allocating some virtual accelerator cards and creating virtual mapping relation tables with different computing power operation tasks at the same time, computing power resources can be provided for computing power operation tasks that require a total of 130 virtual accelerator cards, greatly improving the utilization rate of computing power resources.

[0075] The computing power resources provided by the above N virtual acceleration cards match the computing power resource demand information. Specifically, the computing power resources provided by the N virtual acceleration cards are greater than the computing power resources required as characterized by the computing power resource demand information. For example, if the computing power resource demand information indicates a demand for computing power resources of 50 GB in size, and each virtual acceleration card can provide at most 20 GB in size, then it is determined that 3 virtual acceleration cards will provide the computing power resources. The maximum computing power resources that 3 virtual acceleration cards can provide is 60 GB in size, which is greater than the required 50 GB in size of the computing power resources.

[0076] In some embodiments, computing power resources with an over-allocation quota of a preset ratio can be added. For example, if the preset ratio is set to 20%, when the computing power resource demand information indicates a demand for computing power resources of 35 GB in size, and each virtual acceleration card can provide at most 20 GB in size, after adding the computing power resources with the over-allocation quota of the preset ratio, the required computing power resource size is 42 GB. At this time, it is determined that 3 virtual acceleration cards will provide the computing power resources, avoiding the situation of insufficient container computing power resources caused by inaccurate computing power resource demand information in the computing power resource application information.

[0077] In the process of allocating the first quantity of virtual acceleration cards among the N virtual acceleration cards to the container based on the virtual mapping relationship table, it is necessary to first determine that there are the first quantity of virtual acceleration cards in the N virtual acceleration cards that are in an idle state. In the case where there are the first quantity of virtual acceleration cards idle, the first quantity of virtual acceleration cards is allocated to the container; if the number of virtual acceleration cards in the N virtual acceleration cards that are in an idle state is less than the first quantity, it is possible to wait for the status of other virtual acceleration cards to change to an idle state before allocation, or directly allocate the virtual acceleration cards in the idle state to the container, and monitor other virtual acceleration cards after allocation for timely expansion.

[0078] In one embodiment, step S32 includes: Step S321, in the case where the type information indicates that the computing power operation task is a long-duration type task or an idle-time release type task, determine that the first quantity of virtual acceleration cards is the N virtual acceleration cards; Step S322, in the case where the type information indicates that the computing power operation task is a step-growth type task, determine that the first quantity of virtual acceleration cards is M virtual acceleration cards, where M is a positive integer greater than or equal to 1 and less than or equal to N.

[0079] In the present invention, when the type information indicates that the computing power operation task is a long-term type task or an idle-time release type task, it is determined that the first number of virtual acceleration cards is the N virtual acceleration cards; when the type information indicates that the computing power operation task is a step-growth type task, it is determined that the first number of virtual acceleration cards is M virtual acceleration cards, where M is a positive integer greater than or equal to 1 and less than or equal to N. In this way, different numbers of virtual acceleration cards are determined for different types of computing power operation tasks, avoiding waste of computing power resources caused by allocating too many virtual acceleration cards to containers.

[0080] Among them, for long-term type tasks or idle-time release type tasks, a relatively large amount of computing power resources are usually required during the execution of this type of computing power operation task. Therefore, when allocating virtual acceleration cards, all virtual acceleration cards are directly allocated to avoid the situation where the execution of the computing power operation task is abnormal due to insufficient computing power resources.

[0081] For step-growth type tasks, during the execution of this type of computing power operation task, the computing power resources required by the container change. At this time, some of the N virtual acceleration cards (i.e., M virtual acceleration cards) are first allocated, and then the container is scaled up or down according to the actual usage situation of the container, avoiding waste of computing power resources and abnormal execution of the computing power operation task due to insufficient computing power resources.

[0082] In some embodiments, the M virtual acceleration cards can be 1 virtual acceleration card, that is, the minimum number of virtual acceleration cards is set to execute the computing power operation task, and the container is scaled up when more computing power resources are required subsequently, so as to maximize the utilization rate of computing power resources.

[0083] In one embodiment, the method further includes: Step S34: Set the first number of virtual acceleration cards in the virtual mapping relationship table to the occupied state.

[0084] In the present invention, setting the first number of virtual acceleration cards in the virtual mapping relationship table to the occupied state makes this part of virtual acceleration cards not be occupied by other containers, and facilitates subsequent scaling up or down of the container based on the virtual mapping relationship table.

[0085] In one embodiment, the step S4 includes: Step S41: When the type information indicates that the computing power operation task is a step-growth type task, obtain the load parameter of the container; Step S42: When the load parameter of the container matches the computing power resources provided by the first number of virtual acceleration cards, execute the computing power operation task by the container based on the operating environment and the first number of virtual acceleration cards.

[0086] When the type information indicates that the computing power operation task is a step - growth type task, during the execution of this type of computing power operation task, the computing power resources required by the container change, and it is necessary to determine whether to scale the container up or down according to the load condition of the container.

[0087] Specifically, in the present invention, step S41: When the type information indicates that the computing power operation task is a step - growth type task, obtain the load parameter of the container; step S42: When the load parameter of the container matches the computing power resources provided by the first number of virtual acceleration cards, execute the computing power operation task through the container based on the operating environment and the first number of virtual acceleration cards. In this way, when the load parameter of the container matches the computing power resources provided by the first number of virtual acceleration cards, there is no need to scale the container up or down, and directly use the first number of virtual acceleration cards to execute the computing power operation task, avoiding waste of computing power resources and abnormal execution of the computing power operation task due to insufficient computing power resources.

[0088] In one embodiment, the method further includes: Step S43: When the load parameter of the container does not match the computing power resources provided by the first number of virtual acceleration cards, allocate a second number of virtual acceleration cards to the container based on the computing power resource requirement information, where the second number is greater than the first number, and the computing power resources provided by the second number of virtual acceleration cards are less than or equal to the quantity of the required computing power resources characterized by the computing power resource requirement information; Step S44: Execute the computing power operation task through the container based on the operating environment and the second number of virtual acceleration cards; Step S45: Set the second number of virtual acceleration cards in the virtual mapping relation table to the occupied state.

[0089] In the present invention, step S43: When the load parameter of the container does not match the computing power resources provided by the first number of virtual acceleration cards, allocate a second number of virtual acceleration cards to the container based on the computing power resource requirement information, where the second number is greater than the first number, and the computing power resources provided by the second number of virtual acceleration cards are less than or equal to the quantity of the required computing power resources characterized by the computing power resource requirement information; step S44: Execute the computing power operation task through the container based on the operating environment and the second number of virtual acceleration cards; step S45: Set the second number of virtual acceleration cards in the virtual mapping relation table to the occupied state. In this way, when the load parameter of the container does not match the computing power resources provided by the first number of virtual acceleration cards, the container is scaled up by allocating a second number of virtual acceleration cards to the container, so that the container can use sufficient computing power resources to execute the computing power operation task.

[0090] It should be noted that by setting the second quantity of virtual acceleration cards in the virtual mapping relationship table to the occupied state, the update of the virtual acceleration card status in the virtual mapping relationship table is realized, and it is convenient for subsequent containers to perform scaling based on the virtual mapping relationship table.

[0091] In some embodiments, the second quantity of virtual acceleration cards can be determined in the following manner: Monitor the N virtual acceleration cards through the monitoring layer of the intelligent computing center cloud platform; When a second virtual acceleration card among the N virtual acceleration cards is called, allocate the second quantity of virtual acceleration cards to the container. The second quantity of virtual acceleration cards includes the second virtual acceleration card and the first quantity of virtual acceleration cards. The second virtual acceleration card is the virtual acceleration card other than the first quantity of virtual acceleration cards among the N virtual acceleration cards.

[0092] The above-mentioned monitoring layer of the intelligent computing center cloud platform is used to monitor each virtual acceleration card in the intelligent computing center cloud platform. Specifically, a monitoring layer for monitoring the usage of virtual acceleration cards can be encapsulated on the driver layer of the virtual acceleration card, so that the call situation of each virtual acceleration card can be determined through the monitoring layer.

[0093] In one embodiment, the method further includes: Step S46: Obtain the usage parameter of the second quantity of virtual acceleration cards; Step S47: When there is a first virtual acceleration card among the second quantity of virtual acceleration cards whose usage parameter indicates that it is not used within the first time period, release the computing power resources provided by the first virtual acceleration card and set the first virtual acceleration card in the virtual mapping relationship table to the idle state.

[0094] It should be noted that since the computing power resources required by the step-growth type tasks are different in different time periods and there are situations of increase or decrease, it is not only necessary to expand the container, but also necessary to scale down the container to avoid the situation where the container occupies the computing power resources without using them, and improve the utilization rate of the computing power resources.

[0095] In the present invention, step S46: obtain the usage parameters of the virtual acceleration cards providing the second quantity; step S47: when there is a situation where the usage parameters of the first virtual acceleration card among the second quantity of virtual acceleration cards indicate that it is not used within the first time period, release the computing power resources provided by the first virtual acceleration card, and set the first virtual acceleration card in the virtual mapping relation table to the idle state. In this way, by determining the usage of each virtual acceleration card, when there is a first virtual acceleration card that is not used within the first time period, the computing power resources provided by the first virtual acceleration card are released, and the first virtual acceleration card is set to the idle state, so that the first virtual acceleration card can be allocated to other containers to execute other computing power operation tasks, greatly improving the utilization rate of computing power resources.

[0096] In addition, by setting the first virtual acceleration card in the virtual mapping relation table to the idle state, it can be determined that the first virtual acceleration card is in the idle state during subsequent scaling, and then the first virtual acceleration card can be used for scaling up.

[0097] In one embodiment, step S3 includes: Step S31': when the type information indicates that the computing power operation task is an idle-release type task, monitor the first quantity of virtual acceleration cards; Step S32': when it is monitored that there is a virtual acceleration card being called among the first quantity of acceleration cards, allocate the first quantity of virtual acceleration cards to the container.

[0098] It should be noted that the idle-release type task requires the computing power resources provided by N virtual acceleration cards during the execution stage of the computing power operation task and does not require computing power resources during the idle stage. Therefore, it is necessary to determine the usage of the N virtual acceleration cards to perform scaling up and release in a timely manner, thereby improving the utilization rate of computing power resources.

[0099] In the present invention, step S31': when the type information indicates that the computing power operation task is an idle-release type task, monitor the first quantity of virtual acceleration cards; step S32': when it is monitored that there is a virtual acceleration card being called among the first quantity of acceleration cards, allocate the first quantity of virtual acceleration cards to the container. In this way, by monitoring the virtual acceleration cards to determine whether the container needs computing power resources to execute the computing power operation task, when there is a virtual acceleration card being called, it is considered that the container needs computing power resources to execute the computing power operation task. At this time, all the virtual acceleration cards (i.e., the first quantity of virtual acceleration cards) can be allocated to the container so that the container can execute the computing power operation task based on the computing power resources provided by the first quantity of virtual acceleration cards.

[0100] In one embodiment, the method further includes: Step S5: When the type information indicates that the computing power operation task is an idle release type task, monitor the first number of virtual acceleration cards; Step S6: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state.

[0101] In the present invention, in Step S5: When the type information indicates that the computing power operation task is an idle release type task, monitor the first number of virtual acceleration cards; in Step S6: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state. In this way, by monitoring the first number of virtual acceleration cards to determine whether the container needs computing power resources to execute the computing power operation task, when the first number of virtual acceleration cards are not used within the second time period, it is considered that the container is in the idle stage at this time and does not need computing power resources. At this time, release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state, thereby avoiding the container occupying computing power resources for a long time without use, and greatly improving the utilization rate of computing power resources.

[0102] In one embodiment, Step S6 includes: Step S61: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, save the data in the memory of the first number of virtual acceleration cards to the storage space; Step S62: Release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state.

[0103] In the present invention, in Step S61: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, save the data in the memory of the first number of virtual acceleration cards to the storage space; in Step S62: Release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state. In this way, first save the data in the memory of the first number of virtual acceleration cards to the storage space, then release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state, so that the relevant data for executing the computing power operation task are all stored in the storage space. When the container needs to execute the computing power operation task again later, the relevant data can be quickly obtained directly from the storage space, realizing fast loading and improving the efficiency of executing the computing power operation task.

[0104] Please refer to Figure 3 , Figure 3This is the structural diagram of a virtual computing power pre - allocation device for an intelligent computing center cloud platform provided by the present invention. As Figure 3 shown, the virtual computing power pre - allocation device 300 of the intelligent computing center cloud platform includes: A receiving module 301, configured to receive computing power resource application information, where the computing power resource application information includes type information of a computing power operation task and computing power resource requirement information of the computing power operation task; A creating module 302, configured to create a running environment for a container based on a preset amount of computing power resources when the type information meets a preset condition; An allocating module 303, configured to allocate a first number of virtual acceleration cards to the container based on the type information and the computing power resource requirement information when receiving the computing power operation task. The virtual acceleration card is a virtual acceleration card of the intelligent computing center cloud platform, and the virtual acceleration card is used to provide computing power resources; An executing module 304, configured to execute the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards.

[0105] In one embodiment, the preset condition includes one of the following: The type information indicates that the computing power operation task is a long - term type task; The type information indicates that the computing power operation task is a step - growth type task; The type information indicates that the computing power operation task is an idle - time release type task.

[0106] In one embodiment, the allocating module 303 includes: A creating sub - module, configured to create a virtual mapping relationship table based on the computing power resource requirement information. The virtual mapping relationship table is the relationship between the computing power operation task and N virtual acceleration cards, and the computing power resources provided by the N virtual acceleration cards match the computing power resource requirement information. N is a positive integer greater than or equal to 1; A determining sub - module, configured to determine the first number of virtual acceleration cards based on the type information and the computing power resource requirement information when receiving the computing power operation task; A first allocating sub - module, configured to allocate the first number of virtual acceleration cards among the N virtual acceleration cards to the container based on the virtual mapping relationship table. The first number of virtual acceleration cards are the virtual acceleration cards in an idle state in the virtual mapping relationship table.

[0107] In one embodiment, the determining sub - module includes: A first determination unit, configured to determine the first quantity of virtual acceleration cards as the N virtual acceleration cards when the type information indicates that the computing power operation task is a long-duration type task or an idle-time release type task; A second determination unit, configured to determine the first quantity of virtual acceleration cards as M virtual acceleration cards when the type information indicates that the computing power operation task is a step-growth type task, where M is a positive integer greater than or equal to 1 and less than or equal to N.

[0108] In one embodiment, the virtual computing power pre-allocation device 300 of the intelligent computing center cloud platform further includes: A first setting sub-module, configured to set the first quantity of virtual acceleration cards in the virtual mapping relationship table to an occupied state.

[0109] In one embodiment, the execution module 304 includes: A first acquisition sub-module, configured to acquire the load parameter of the container when the type information indicates that the computing power operation task is a step-growth type task; A first execution sub-module, configured to execute the computing power operation task through the container based on the running environment and the first quantity of virtual acceleration cards when the load parameter of the container matches the computing power resources provided by the first quantity of virtual acceleration cards.

[0110] In one embodiment, the virtual computing power pre-allocation device 300 of the intelligent computing center cloud platform further includes: A second allocation sub-module, configured to allocate a second quantity of virtual acceleration cards to the container based on the computing power resource demand information when the load parameter of the container does not match the computing power resources provided by the first quantity of virtual acceleration cards, where the second quantity is greater than the first quantity, and the computing power resources provided by the second quantity of virtual acceleration cards are less than or equal to the quantity of the required computing power resources indicated by the computing power resource demand information; A second execution sub-module, configured to execute the computing power operation task through the container based on the running environment and the second quantity of virtual acceleration cards; A second setting sub-module, configured to set the second quantity of virtual acceleration cards in the virtual mapping relationship table to an occupied state.

[0111] In one embodiment, the virtual computing power pre-allocation device 300 of the intelligent computing center cloud platform further includes: A second acquisition sub-module, configured to acquire the usage situation parameter of the second quantity of virtual acceleration cards; A first release sub-module, configured to release the computing power resource device provided by the first virtual acceleration card and set the first virtual acceleration card in the virtual mapping relation table to an idle state when the usage parameter of the first virtual acceleration card among the second quantity of virtual acceleration cards indicates that it has not been used within the first time period.

[0112] In one embodiment, the allocation module 303 includes: A monitoring sub-module, configured to monitor the first quantity of virtual acceleration cards when the type information indicates that the computing power operation task is an idle release type task; A third allocation sub-module, configured to allocate the first quantity of virtual acceleration cards to the container when it is monitored that a virtual acceleration card among the first quantity of acceleration cards is called.

[0113] In one embodiment, the virtual computing power pre-allocation device 300 of the intelligent computing center cloud platform further includes: A monitoring module, configured to monitor the first quantity of virtual acceleration cards when the type information indicates that the computing power operation task is an idle release type task; A release module, configured to release the container and set the first quantity of virtual acceleration cards in the virtual mapping relation table to an idle state when it is monitored that the first quantity of virtual acceleration cards has not been used within the second time period.

[0114] In one embodiment, the release module includes: A saving sub-module, configured to save the data in the memory of the first quantity of virtual acceleration cards to the storage space when it is monitored that the first quantity of virtual acceleration cards has not been used within the second time period; A second release sub-module, configured to release the container and set the first quantity of virtual acceleration cards in the virtual mapping relation table to an idle state.

[0115] The virtual computing power pre-allocation device of the intelligent computing center cloud platform provided by the embodiments of the present invention can implement each process of the virtual computing power pre-allocation method of the above intelligent computing center cloud platform, with the technical features corresponding one by one and achieving the same technical effects. To avoid repetition, it will not be elaborated here.

[0116] It should be noted that the virtual computing power pre-allocation device of the intelligent computing center cloud platform in the embodiments of the present invention can be a device, or a component, an integrated circuit, or a chip in an electronic device.

[0117] The present invention also provides an electronic device. Refer to Figure 4 , Figure 4It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device includes a memory 401, a processor 402, and a program or instruction running on the memory 401. When the program or instruction is executed by the processor 402, it can implement Figure 1 Any steps in the embodiment of the virtual computing power pre-allocation method of the corresponding intelligent computing center cloud platform and achieve the same beneficial effects, which will not be elaborated here.

[0118] Among them, the processor 402 can be a CPU, ASIC, FPGA or GPU.

[0119] Those of ordinary skill in the art can understand that all or part of the steps of implementing the embodiment of the virtual computing power pre-allocation method of the intelligent computing center cloud platform can be completed by hardware related to program instructions, and the program can be stored in a readable medium.

[0120] The present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement the above Figure 1 Any steps in the embodiment of the virtual computing power pre-allocation method of the corresponding intelligent computing center cloud platform and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. The storage medium such as a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disc, etc.

[0121] The present invention also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, it implements the above Figure 1 Each process of the embodiment of the virtual computing power pre-allocation method of the corresponding intelligent computing center cloud platform and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0122] The terms "first", "second", etc. in the present invention are used to distinguish similar objects and do not have to be used to describe a specific order or sequence. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. In addition, in this application, "and / or" is used to represent at least one of the connected objects. For example, A and / or B and / or C represents 7 cases including A alone, B alone, C alone, A and B both present, B and C both present, A and C both present, and A, B, and C all present.

[0123] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising that element.

[0124] From the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on this understanding, the technical solution of the present application, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which may be a mobile phone, computer, server, air conditioner, or a second terminal device, etc.) to execute the methods of the various embodiments of the present application.

[0125] The embodiments of the present application are described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Those of ordinary skill in the art, under the inspiration of the present application and without departing from the spirit of the present application and the scope protected by the claims, can still make many forms, all of which fall within the protection scope of the present application.

Claims

1. A virtual computing power pre-allocation method for an intelligent computing center cloud platform, characterized in that, Including: Step S1: Receive computing power resource application information, where the computing power resource application information includes the type information of the computing power operation task and the computing power resource demand information of the computing power operation task; Step S2: When the type information meets the preset conditions, create a running environment for the container based on the preset computing power resource amount; Step S3: When receiving the computing power operation task, allocate a first number of virtual acceleration cards to the container based on the type information and the computing power resource demand information. The virtual acceleration cards are the virtual acceleration cards of the intelligent computing center cloud platform, and the virtual acceleration cards are used to provide computing power resources; Step S4: Execute the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards; The preset conditions include one of the following: The type information indicates that the computing power operation task is a long-term type task; The type information indicates that the computing power operation task is a step-growth type task; The type information indicates that the computing power operation task is an idle-time release type task.

2. The method according to claim 1, wherein Step S3 includes: Step S31: Create a virtual mapping relationship table based on the computing power resource demand information. The virtual mapping relationship table is a relationship table between the computing power operation task and N virtual acceleration cards. The computing power resources provided by the N virtual acceleration cards match the computing power resource demand information, and N is a positive integer greater than or equal to 1; Step S32: When receiving the computing power operation task, determine the first number of virtual acceleration cards based on the type information and the computing power resource demand information; Step S33: Allocate the first number of virtual acceleration cards among the N virtual acceleration cards to the container based on the virtual mapping relationship table. The first number of virtual acceleration cards are the virtual acceleration cards in the virtual mapping relationship table that are in an idle state.

3. The method according to claim 2, wherein Step S32 includes: Step S321: When the type information indicates that the computing power operation task is a long-term type task or an idle-time release type task, determine the first number of virtual acceleration cards as the N virtual acceleration cards; Step S322: When the type information indicates that the computing power operation task is a step-growth type task, determine the first number of virtual acceleration cards as M virtual acceleration cards, where M is a positive integer greater than or equal to 1 and less than or equal to N.

4. The method according to claim 2, wherein The method further includes: Step S34: Set the first number of virtual acceleration cards in the virtual mapping relationship table to an occupied state.

5. The method according to any one of claims 1 to 4, characterized in that, Step S4 includes: Step S41: When the type information indicates that the computing power operation task is a step-growth type task, obtain the load parameter of the container; Step S42: When the load parameter of the container matches the computing power resources provided by the first number of virtual acceleration cards, execute the computing power operation task through the container based on the running environment and the first number of virtual acceleration cards.

6. The method according to claim 5, wherein The method further includes: Step S43: When the load parameter of the container does not match the computing power resources provided by the first number of virtual acceleration cards, allocate a second number of virtual acceleration cards to the container based on the computing power resource demand information, where the second number is greater than the first number, and the computing power resources provided by the second number of virtual acceleration cards are less than or equal to the quantity of the required computing power resources characterized by the computing power resource demand information; Step S44: Execute the computing power operation task through the container based on the operating environment and the second number of virtual acceleration cards; Step S45: Set the second number of virtual acceleration cards in the virtual mapping relationship table to the occupied state.

7. The method according to claim 6, wherein The method further includes: Step S46: Obtain the usage parameter of the second number of virtual acceleration cards provided; Step S47: When there is a first virtual acceleration card whose usage parameter indicates that it is not used within the first time period among the second number of virtual acceleration cards, release the computing power resources provided by the first virtual acceleration card, and set the first virtual acceleration card in the virtual mapping relationship table to the idle state.

8. The method according to any one of claims 1 to 4, characterized in that, The step S3 includes: Step S31': When the type information indicates that the computing power operation task is an idle-time release type task, monitor the first number of virtual acceleration cards; Step S32': When it is monitored that a virtual acceleration card is called among the first number of acceleration cards, allocate the first number of virtual acceleration cards to the container.

9. The method according to any one of claims 1 to 4, characterized in that The method further includes: Step S5: When the type information indicates that the computing power operation task is an idle-time release type task, monitor the first number of virtual acceleration cards; Step S6: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state.

10. The method according to claim 9, characterized in that, The step S6 includes: Step S61: When it is monitored that the first number of virtual acceleration cards are not used within the second time period, save the data in the memory of the first number of virtual acceleration cards to the storage space; Step S62: Release the container, and set the first number of virtual acceleration cards in the virtual mapping relationship table to the idle state.

11. A virtual computing power pre-allocation device for an intelligent computing center cloud platform, characterized in that, It includes: A receiving module, configured to receive computing power resource application information, where the computing power resource application information includes the type information of the computing power operation task and the computing power resource demand information of the computing power operation task; A creating module, configured to create an operating environment for the container based on a preset computing power resource amount when the type information meets the preset conditions; An allocating module, configured to allocate a first number of virtual acceleration cards to the container based on the type information and the computing power resource demand information when receiving the computing power operation task, where the virtual acceleration cards are virtual acceleration cards of the intelligent computing center cloud platform and are used to provide computing power resources; An executing module, configured to execute the computing power operation task through the container based on the operating environment and the first number of virtual acceleration cards; The preset conditions include one of the following: The type information characterizes that the computing power operation task is a long-duration type task; The type information characterizes that the computing power operation task is a step-growth type task; The type information characterizes that the computing power operation task is an idle-time release type task.

12. An electronic device, characterized in that, Comprising: A processor, a memory, and a program stored on the memory and executable on the processor, and when the program is executed by the processor, the steps of the virtual computing power pre-allocation method of the intelligent computing center cloud platform according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the virtual computing power pre-allocation method of the intelligent computing center cloud platform according to any one of claims 1 to 10 are implemented.

14. A computer program product, characterized in that, Comprising computer instructions, and when the computer instructions are executed by a processor, the steps of the virtual computing power pre-allocation method of the intelligent computing center cloud platform according to any one of claims 1 to 10 are implemented.

Images