Computing power directional scheduling method and device of intelligent computing center

Abstract

The invention provides a computing power directional scheduling method and device of an intelligent computing center. The method comprises the steps that S1, computing power operation task information input by a user is received; s2, receiving node selection information input by a user based on the computing power operation task; and S3, scheduling the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information. After the computing power operation task information input by the user is received, the operation code information and the task configuration information are determined, and the computing power operation task is scheduled to the target computing power node according to the node selection information input by the user. And the target computing power node processes the computing power operation task based on the operation code information and the task configuration information, so that a user can select heterogeneous computing power nodes in a customized manner, and the flexibility of processing the computing power operation task by an intelligent computing center is greatly improved.

Description

Technical Field

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

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

[0003] An "intelligent computing center" refers to a facility that uses large-scale heterogeneous computing power resources, including general computing power and intelligent computing power, to mainly provide 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, 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.

[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 is based on artificial intelligence theory, adopts an 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 centers" and "intelligent computing centers". It 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 target results by processing information data, and a new type of productive force that integrates information computing power, network carrying capacity, and data storage capacity, and mainly provides services to society through computing power infrastructure.

[0007] Since the emergence of intelligent computing centers, the problem of user-defined selection of heterogeneous computing power nodes has been an urgent problem to be solved. Currently, when an intelligent computing center provides computing power resource services to users, users can only request the intelligent computing center to process computing power operation tasks, and users cannot select the heterogeneous computing power nodes for processing computing power operation tasks, resulting in a very poor flexibility in the intelligent computing center's processing of computing power operation tasks. Summary of the Invention

[0008] The present invention provides a method and device for directional scheduling of computing power in an intelligent computing center, which solves the problem of very poor flexibility in the intelligent computing center's processing of computing power operation tasks in the prior art.

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

[0010] In a first aspect, the present invention provides a method for power computing capacity oriented scheduling of an intelligent computing center, the method comprising:

[0011] Step S1, receiving computing capacity operation task information input by a user, the computing capacity operation task information including operation code information and task configuration information, the operation code information being used to process the computing capacity operation task, and the task configuration information including computing capacity resource parameter values for processing the computing capacity operation task;

[0012] Step S2, receiving node selection information input by the user based on the computing capacity operation task, the node selection information being used to determine a target computing capacity node for processing the computing capacity operation task among a plurality of heterogeneous computing capacity nodes, the target computing capacity node being one of the plurality of heterogeneous computing capacity nodes;

[0013] Step S3, scheduling the computing capacity operation task to the target computing capacity node, so that the target computing capacity node processes the computing capacity operation task based on the operation code information and the task configuration information.

[0014] Optionally, the method further comprises:

[0015] Step S4, during the process that the target computing capacity node processes the computing capacity operation task based on the operation code information and the task configuration information, when a task query request input by the user is received, in response to the task query request, displaying a first page;

[0016] Wherein, the task query request is used to query status information of the computing capacity operation task, the first page includes first log information, first resource information, and first node information corresponding to the computing capacity operation task, the first log information being task logs for processing the computing capacity operation task, the first resource information being computing capacity resource usage status information for processing the computing capacity operation task, and the first node information being computing node information for processing the computing capacity operation task.

[0017] Optionally, after step S4, during the process that the target computing capacity node processes the computing capacity operation task based on the operation code information and task configuration information, when a task query request input by the user is received, in response to the task query request, displaying a first page, the method further comprises:

[0018] Step S5, when the target computing capacity node finishes processing the computing capacity operation task based on the operation code information and the task configuration information, displaying a second page;

[0019] Among them, the second page includes second log information, second resource information, and second node information. The second log information is the task log of processing the computing power operation task, the second resource information is the computing power resource usage status information of processing the computing power operation task, and the second node information is the computing node information of processing the computing power operation task.

[0020] Optionally, the step S1 of receiving the computing power operation task information input by the user includes:

[0021] Step S11: Based on the code configuration interface, receive the operation code information input by the user;

[0022] Step S12: When performing security verification on the operation code information and the verification is passed, generate a task configuration interface based on the operation code information;

[0023] Step S13: Based on the task configuration interface, receive the task configuration information input by the user.

[0024] Optionally, the step S13 of receiving the task configuration information input by the user based on the task configuration interface includes:

[0025] Step S131: Receive the mirror file imported by the user or selected in the task configuration interface. The mirror file is used to create and deploy the configuration environment for processing the computing power operation task;

[0026] Step S132: Based on the user input, determine the computing power resource information. The computing power resource information includes the computing power resource parameters for processing the computing power operation task;

[0027] Among them, the task configuration information includes the mirror file and the computing power resource information.

[0028] Optionally, the step S2 of receiving the node selection information input by the user based on the computing power operation task includes:

[0029] Step S21: Based on the user input, determine at least one computing unit and at least one configuration information. The at least one computing unit and the at least one configuration information are in one-to-one correspondence. The computing unit is used to run the operation code in the configuration environment, and the configuration information is used to indicate the configuration parameters of the corresponding computing unit;

[0030] Among them, the target computing power node includes the at least one computing unit, and the node selection information includes the target computing power node and the at least one configuration information.

[0031] Optionally, the first resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the power consumption of the CPU that has been used, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory;

[0032] The second resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the power consumption of the CPU that has been used, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory.

[0033] In a second aspect, the present invention provides a computing power directional scheduling device for an intelligent computing center, and the device includes:

[0034] A first receiving module, configured to receive the computing power operation task information input by a user, where the computing power operation task information includes operation code information and task configuration information, the operation code information is used to process the computing power operation task, and the task configuration information includes the computing power resource parameter values for processing the computing power operation task;

[0035] A second receiving module, configured to receive the node selection information input by the user based on the computing power operation task, where the node selection information is used to determine a target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes, and the target computing power node is one of the multiple heterogeneous computing power nodes;

[0036] A processing module, configured to schedule the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information.

[0037] 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, and when the computer program is executed by the processor, the steps in the method described in the first aspect above are implemented.

[0038] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the method described in the first aspect above are implemented.

[0039] In a fifth aspect, the present invention further provides a computer program product, including computer instructions, and when the computer instructions are executed by a processor, the steps in the method described in the first aspect above are implemented.

[0040] The present invention provides a method and device for power computing task scheduling in an intelligent computing center. The method includes: Step S1, receiving computing power operation task information input by a user, where the computing power operation task information includes operation code information and task configuration information. The operation code information is used to process the computing power operation task, and the task configuration information includes the value of computing power resource parameters for processing the computing power operation task; Step S2, receiving node selection information input by the user based on the computing power operation task. The node selection information is used to determine a target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes, and the target computing power node is one of the multiple heterogeneous computing power nodes; Step S3, scheduling the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information. After receiving the computing power operation task information input by the user, the present invention determines the operation code information and the task configuration information, and according to the node selection information input by the user, schedules the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information, realizing the user's custom selection of heterogeneous computing power nodes and greatly improving the flexibility of the intelligent computing center in processing computing power operation tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] To more clearly illustrate the technical solutions of the present invention, the following will briefly introduce the drawings required for the description of the present invention. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1 is a flowchart of a method for power computing task scheduling in an intelligent computing center provided by the present invention;

[0043] Figure 2 is a structural diagram of a device for power computing task scheduling in an intelligent computing center provided by the present invention;

[0044] Figure 3 is a structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] The following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0046] 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 process information data and achieve the output of a target result, a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly providing services to society through computing power infrastructure.

[0047] 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_general + CP_intelligent + CP_super.

[0048] The "carrying capacity" (Network Power, NP) referred to in the present invention means: the manifestation of the data transmission ability of computing power facilities, a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., involving network transmission inside and between data centers, and a comprehensive indicator for measuring network transmission scheduling ability.

[0049] The "storage capacity" (Storage Power, SP) referred to in the present invention means: the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon, a comprehensive indicator for measuring the data storage ability of a data center, including external storage devices such as storage arrays and server internal storage devices. The commonly used measurement unit for storage capacity is exabyte (EB, 1 EB = 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.

[0050] The "computing power infrastructure" referred to in the present invention means: a new type of information infrastructure integrating information computing power, network carrying capacity, and data storage capacity, which can realize the centralized computing, storage, transmission, and application of information.

[0051] The "new information infrastructure" as described in the present invention refers to: mainly including network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, satellite Internet, etc., computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, supercomputing centers, etc., and new technology facilities such as artificial intelligence, blockchain, quantum computing, etc.

[0052] The "computing power" as described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0053] The "general computing power" as 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.

[0054] The "intelligent computing power" as 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), ASIC (Application Specific Integrated Circuit), such as natural language processing, machine vision, etc.

[0055] The "super computing power" as described in the present invention refers to: mainly 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.

[0056] The "intelligent computing center" as described in the present invention refers to: a facility that mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios like artificial intelligence deep learning model development, model training, and model inference) 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 the underlying computing power to the top-level application enabling.

[0057] The "intelligent computing center" as described in the present invention includes but is not limited to the "intelligent computing center".

[0058] The "intelligent computing center" as described in the present invention, that is, the artificial intelligence computing center, is a type of computing power infrastructure based on artificial intelligence theory, adopting an artificial intelligence computing architecture, and providing computing power services, data services, and algorithm services required for artificial intelligence applications.

[0059] 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, with computing power, transportation power, and storage power, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0060] The "supercomputing center" described in the present invention refers to: that is, a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters, capable of providing functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0061] The "computing power resources" described in the present invention refer 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 CPUs and GPUs, computing power 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.

[0062] The "inclusive computing power" described in the present invention refers to: based on the requirements of equal opportunity and the principle of commercial sustainability, providing appropriate and effective computing power services to all social strata and groups with computing power service needs at an affordable cost.

[0063] The "model" described in the present invention includes but is not limited to "large language models" and "multimodal large models".

[0064] 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.

[0065] The "multimodal large model" (Multimodal Large Models) described in the present invention refers to: a model that jointly trains multimodal information such as text, images, videos, and audio, including but not limited to multimodal large language models.

[0066] The "computing power operation task" described in the present invention refers to specific workloads or jobs that are executed on computing power resources and require a certain amount of computing power support, usually involving scenarios such as complex data processing, numerical calculations, model training, or simulation.

[0067] The "heterogeneous computing power nodes" described in the present invention include: computing tasks or training programs need to be deployed and run in some different entities. These entities will provide some computing power such as CPUs, GPUs, Networks, Storage, etc. Heterogeneous refers to different implementation methods among "virtual machines", "containers", and "bare metals", and these three are all common computing power nodes.

[0068] Please refer to Figure 1 , Figure 1 which is a flowchart of a computing power directed scheduling method for an intelligent computing center provided by the present invention. As Figure 1 shown, it includes the following steps:

[0069] Step S1: Receive the computing power operation task information input by the user. The computing power operation task information includes operation code information and task configuration information. The operation code information is used to process the computing power operation task, and the task configuration information includes the computing power resource parameter values for processing the computing power operation task.

[0070] In the present invention, the execution subject in this embodiment can be the server side. The intelligent computing center can receive the computing power operation task information input by the user or the user creates a new computing power operation task in the intelligent computing center, so as to process the computing power operation task requested by the user. Specifically, after the intelligent computing center receives the computing power operation task information input by the user, it prompts the user to input operation code information and task configuration information on the page. Among them, the operation code information includes the task code for processing the computing power operation task, which varies according to specific application scenarios and requirements. Exemplarily, it can be code written in Python, which is not specifically limited in the present invention. The task configuration information includes the computing power resource parameter values for processing the computing power operation task. Among them, the computing power resource parameters can include computing parameters, storage parameters, and computing power resource parameters, etc. In this embodiment, each computing power resource parameter corresponds to a specific value. For example, the computing parameter in the task configuration information can be the usage rate of the GPU, the usage rate of the memory, etc.

[0071] Step S2: Receive the node selection information input by the user based on the computing power operation task. The node selection information is used to determine the target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes. The target computing power node is one of the multiple heterogeneous computing power nodes.

[0072] In the present invention, the heterogeneous computing power nodes are different computing power nodes among "virtual machines", "containers", and "bare metals", and the user can customize and select the heterogeneous computing power nodes in the present invention.

[0073] Specifically, the user inputs node selection information at the server end according to the computing power to run a task. Among them, the node selection information includes the target computing power node selected by the user from multiple heterogeneous computing power nodes. It should be noted that the target computing power node is one of the multiple heterogeneous computing power nodes, and during the process of the computing power running task, the computing power running task is processed in the target computing power node.

[0074] If the target computing power node selected by the user has been used or cannot meet the processing of the computing power running task, the server will prompt the user to re-select from multiple heterogeneous computing power nodes and re-determine a suitable target computing power node.

[0075] Exemplarily, the code for determining the target computing power node can be as follows:

[0076]

[0077]

[0078] Among them, key is the host name and values is the server ID.

[0079] Step S3: Schedule the computing power running task to the target computing power node so that the target computing power node processes the computing power running task based on the running code information and the task configuration information.

[0080] In the present invention, the computing power running task is scheduled to the target computing power node, and the target computing power node is used as a computing power container or a computing power node, so that the target computing power node processes the computing power running task based on the running code information and the task configuration information input by the user.

[0081] After receiving the computing power running task information input by the user, the present invention determines the running code information and the task configuration information, and schedules the computing power running task to the target computing power node according to the node selection information input by the user, so that the target computing power node processes the computing power running task based on the running code information and the task configuration information, realizing the user's custom selection of heterogeneous computing power nodes and greatly improving the flexibility of the intelligent computing center to process computing power running tasks.

[0082] In some feasible embodiments, optionally, the method further includes:

[0083] Step S4: During the process that the target computing power node processes the computing power running task based on the running code information and the task configuration information, when a task query request input by the user is received, in response to the task query request, display a first page;

[0084] Among them, the task query request is used to query the status information of the computing power operation task. The first page includes the first log information, the first resource information, and the first node information corresponding to the computing power operation task. The first log information is the task log for processing the computing power operation task. The first resource information is the usage status information of the computing power resources for processing the computing power operation task. The first node information is the computing node information for processing the computing power operation task.

[0085] In this embodiment, during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information, the user can query the processing status of the computing power operation task at any time. Specifically, the user can input a task query request in the server, and the server displays the first page on the page according to the task query request.

[0086] Specifically, according to the task query request, the first page is displayed to the user. Among them, the first page includes the first log information, the first resource information, and the first node information corresponding to the computing power operation task. Specifically, the first log information includes the task log for processing the computing power operation task, such as the current processing result of the computing power operation task, the output log, etc. The first log information is the processing log generated by the computing unit, and the processing log may include at least one of the following: task basic information, resource usage, error and exception information, task status, processing time, system information, etc. The first resource information is the current usage status information of the computing power resources for processing the computing power operation task, such as the computing resource utilization rate, system load, etc. The first node information is the processing node that actually processes the computing power operation task, such as the ID of the processing node, the system number, etc.

[0087] Optionally, after step S4, during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information, when a task query request input by the user is received, and in response to the task query request, after the first page is displayed, the method further includes:

[0088] Step S5, when the target computing power node finishes processing the computing power operation task based on the operation code information and the task configuration information, display a second page;

[0089] Among them, the second page includes the second log information, the second resource information, and the second node information. The second log information is the task log for completing the processing of the computing power operation task. The second resource information is the usage status information of the computing power resources for completing the processing of the computing power operation task. The second node information is the computing node information for completing the processing of the computing power operation task.

[0090] In the present invention, after the target computing power node finishes processing the computing power operation task based on the operation code information and the task configuration information, the intelligent computing center displays a second page to the user. The second page is the same as the first page and includes second log information, second resource information, and second node information. Specifically, the second log information is the task log of the completed computing power operation task, such as the processing result of the computing power operation task, output log, etc. Specifically, the second log information is the processing log generated by the computing unit, and the processing log may include at least one of the following: basic task information, resource usage, error and exception information, task status, processing time, system information, etc. The second node information is the processing node that has completed the computing power operation task, such as the ID of the processing node, system number, etc. The second resource information is the overall usage status information of the computing power resources for processing the computing power operation task, such as computing resource utilization rate, system load, etc.

[0091] By displaying the relevant information of the processed computing power operation task after the processing of the computing power operation task is completed, it can help the user quickly and comprehensively understand the relevant computing power resources of the processed computing power operation task, facilitate the user to use it as a reference for the next processing of the computing power operation task, and improve the flexibility of the intelligent computing center in processing the computing power operation task.

[0092] Optionally, step S1 of receiving the computing power operation task information input by the user includes:

[0093] Step S11: Receive the operation code information input by the user based on the code configuration interface;

[0094] Step S12: Generate a task configuration interface based on the operation code information when the security verification of the operation code information is passed;

[0095] Step S13: Receive the task configuration information input by the user based on the task configuration interface.

[0096] In the present invention, after receiving the operation code information input by the user on the display interface, it is necessary to perform security verification on the input operation code information, mainly to verify whether the code is secure. Exemplarily, the method of verifying the operation code information can be to use static analysis tools (such as SonarQube, ESLint, etc.) to scan for security vulnerabilities and bad coding in the code or use dynamic analysis tools (such as OWASP ZAP, Burp Suite) to test the running code to discover potential runtime vulnerabilities, etc. No specific limitation is made in this embodiment.

[0097] In the case where the security verification of the running code information passes, a task configuration interface is generated based on the running code information, where the task configuration interface is used to receive task configuration information input by the user. By verifying the running code information, the security of processing the computing power operation task can be ensured.

[0098] Optionally, step S13, receiving the task configuration information input by the user based on the task configuration interface, includes:

[0099] Step S131, receiving the mirror file imported by the user or selected in the task configuration interface, where the mirror file is used to create and deploy a configuration environment for processing the computing power operation task;

[0100] Step S132, determining computing power resource information based on the user input, where the computing power resource information includes computing power resource parameters for processing the computing power operation task;

[0101] Wherein, the task configuration information includes the mirror file and the computing power resource information.

[0102] In the present invention, the task configuration information includes a mirror file and computing power resource information. After the user selects the above configuration, task configuration information is generated. It should be noted that the task configuration information in the present invention may include one or more of the above, and specific adaptive adjustments are made according to actual situations.

[0103] Specifically, the mirror file is used to determine the mirror, which can be imported by the user himself or selected by the user in the task configuration interface. The mirror can define the application program and all its dependencies, ensure that different computing nodes or services run in exactly the same environment, help the user more clearly define and allocate the required computing resources, and avoid resource waste caused by environmental problems. In this embodiment, the selection of the mirror file, the selection of the number of workers (computing units), and the selection of the resources of a single worker are included. After the user has made all the selections, task configuration information is generated, and then the security verification of the task configuration information is performed.

[0104] The computing power resource information is determined by the user's selection. For example, it is selected in the task configuration interface. The computing power resource information usually includes various parameters, metrics, and configurations related to the construction, management, and performance of the network, such as the uplink transmission rate and the downlink transmission rate, etc.

[0105] Exemplarily, the user selects from multiple candidate mirror files in the task configuration interface to determine the required mirror file. Similarly, the computing power resource information can also be selected by the user in the task configuration interface to determine multiple pieces of information corresponding to the computing power resource information, such as parameters, configurations, etc., which are not specifically limited in this embodiment.

[0106] In the present invention, the computing power resources required by the user can be accurately determined based on the mirror file and computing power resource information input by the user. Therefore, the computing power resources required by the user can be used to process the computing power operation task.

[0107] Optionally, step S2, receiving node selection information input by the user based on the computing power operation task, includes:

[0108] Step S21, based on the user input, determine at least one computing unit and at least one configuration information, the at least one computing unit and the at least one configuration information are in one-to-one correspondence, the computing unit is used to run the running code in the configuration environment, and the configuration information is used to indicate the configuration parameters of the corresponding computing unit;

[0109] Wherein, the target computing power node includes the at least one computing unit, and the node selection information includes the target computing power node and the at least one configuration information.

[0110] In this embodiment, the computing unit can be a worker. Specifically, there are multiple identical or different workers in the task configuration interface. The user can select in the task configuration interface to determine one or more workers. The workers and the configuration information are in one-to-one correspondence and are determined by the user's selection. "Worker" usually refers to a computing unit that executes specific computing or data processing tasks. A worker can be a physical server, a virtual machine, a container, or other computing resources. Workers make full use of available CPU, memory, and storage resources by distributing computing tasks, maximizing the utilization rate of system resources. The configuration information is used to determine the configuration parameters of the corresponding computing unit, such as GPU selection, number of CPUs, amount of memory, etc. The identity information in the identity information of at least one computing unit refers to the number, system parameters, name, etc. corresponding to each computing unit.

[0111] Exemplarily, the user selects from multiple pre-designed computing units in the task configuration interface to determine at least one computing unit. After determining at least one computing unit, select the configuration information corresponding to each computing unit according to the at least one computing unit, such as selecting the CPU, memory, etc. of the computing unit, which is not specifically limited in this embodiment.

[0112] By customizing the determination of the computing unit and the configuration information, the computing power resources required by the user can be used to process the computing power operation task.

[0113] Optionally, the first resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the usage rate of the CPU memory space, the power consumed by the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the usage rate of the disk, and the usage rate of the memory;

[0114] The second resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the usage rate of the CPU memory space, the power consumed by the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the usage rate of the disk, and the usage rate of the memory.

[0115] In this embodiment, the first resource information indicates the calculation loss value of the computing power operation task during the processing of the computing power operation task, the duration when the computing power operation task is processed, the configuration environment parameters for processing the computing power operation task, the computing resource parameters for processing the computing power operation task, and the processing result of the computing power operation task. The configuration environment parameters include the mirror file, and the computing resource parameters include the identity information of the at least one computing unit and the at least one configuration information.

[0116] The second resource information indicates the calculation loss value of the computing power operation task after the processing of the computing power operation task is completed, the duration when the computing power operation task is processed, the configuration environment parameters for processing the computing power operation task, the computing resource parameters for processing the computing power operation task, and the processing result of the computing power operation task. The configuration environment parameters include the mirror file, and the computing resource parameters include the identity information of the at least one computing unit and the at least one configuration information.

[0117] After receiving the computing power operation task information input by the user, the present invention determines the operation code information and the task configuration information, and schedules the computing power operation task to the target computing power node according to the node selection information input by the user, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information, realizing the user's custom selection of heterogeneous computing power nodes, and greatly improving the flexibility of the intelligent computing center in processing computing power operation tasks.

[0118] Please refer to Figure 2 , Figure 2 which is the structural diagram of a computing power directional scheduling device of an intelligent computing center provided by the present invention. As Figure 2 shown, the computing power directional scheduling device 200 of the intelligent computing center includes:

[0119] A first receiving module 210, configured to receive computing power operation task information input by a user, where the computing power operation task information includes operation code information and task configuration information, the operation code information is used to process the computing power operation task, and the task configuration information includes a computing power resource parameter value for processing the computing power operation task;

[0120] A second receiving module 220, configured to receive node selection information input by the user based on the computing power operation task, where the node selection information is used to determine a target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes, and the target computing power node is one of the multiple heterogeneous computing power nodes;

[0121] A processing module 230, configured to schedule the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information.

[0122] Optionally, it further includes:

[0123] A first display module, configured to display a first page in response to a task query request received by the user during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information;

[0124] Wherein, the task query request is used to query the status information of the computing power operation task, the first page includes first log information, first resource information, and first node information corresponding to the computing power operation task, the first log information is the task log for processing the computing power operation task, the first resource information is the computing power resource usage status information for processing the computing power operation task, and the first node information is the computing node information for processing the computing power operation task.

[0125] Optionally, it further includes:

[0126] A second display module, configured to display a second page when the target computing power node finishes processing the computing power operation task based on the operation code information and the task configuration information;

[0127] Wherein, the second page includes second log information, second resource information, and second node information, the second log information is the task log for completing the processing of the computing power operation task, the second resource information is the computing power resource usage status information for completing the processing of the computing power operation task, and the second node information is the computing node information for completing the processing of the computing power operation task.

[0128] Optionally, the first receiving module 210 includes:

[0129] The first receiving sub-module is used to receive the operation code information input by the user based on the code configuration interface;

[0130] The verification sub-module is used to generate a task configuration interface based on the operation code information when the operation code information is verified for security and the verification passes;

[0131] The second receiving sub-module is used to receive the task configuration information input by the user based on the task configuration interface.

[0132] Optionally, the second receiving sub-module includes:

[0133] The receiving unit is used to receive the mirror file imported by the user or selected in the task configuration interface, and the mirror file is used to create and deploy a configuration environment for processing the computing power operation task;

[0134] The determining unit is used to determine the computing power resource information based on the user input, and the computing power resource information includes the computing power resource parameters for processing the computing power operation task;

[0135] Wherein, the task configuration information includes the mirror file and the computing power resource information.

[0136] Optionally, the second receiving module includes:

[0137] The determining sub-module is used to determine at least one computing unit and at least one configuration information based on the user input, the at least one computing unit and the at least one configuration information are in one-to-one correspondence, the computing unit is used to run the operation code in the configuration environment, and the configuration information is used to indicate the configuration parameters of the corresponding computing unit;

[0138] Wherein, the target computing power node includes the at least one computing unit, and the node selection information includes the target computing power node and the at least one configuration information.

[0139] Optionally, the first resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the used power of the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory;

[0140] The second resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the used power of the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory.

[0141] After receiving the computing power operation task information input by the user, the present invention determines the operation code information and the task configuration information, and schedules the computing power operation task to the target computing power node according to the node selection information input by the user, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information, realizing the user's custom selection of heterogeneous computing power nodes and greatly improving the flexibility of the intelligent computing center to process computing power operation tasks.

[0142] An embodiment of the present invention also provides an electronic device. Please refer to Figure 3 , the electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and executable on the processor 301.

[0143] When the program 3021 is executed by the processor 301, it can implement Figure 1 any step in the corresponding method embodiment:

[0144] Step S1: Receive the computing power operation task information input by the user, where the computing power operation task information includes operation code information and task configuration information, the operation code information is used to process the computing power operation task, and the task configuration information includes the computing power resource parameter values for processing the computing power operation task;

[0145] Step S2: Receive the node selection information input by the user based on the computing power operation task, where the node selection information is used to determine the target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes, and the target computing power node is one of the multiple heterogeneous computing power nodes;

[0146] Step S3: Schedule the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information.

[0147] Optionally, the method further includes:

[0148] Step S4: During the process that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information, when a task query request input by the user is received, in response to the task query request, display a first page;

[0149] Among them, the task query request is used to query the status information of the computing power operation task. The first page includes the first log information, the first resource information, and the first node information corresponding to the computing power operation task. The first log information is the task log for processing the computing power operation task. The first resource information is the status information of the computing power resources used for processing the computing power operation task. The first node information is the computing node information for processing the computing power operation task.

[0150] Optionally, in step S4, during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information, when a task query request input by the user is received, after displaying the first page in response to the task query request, the method further includes:

[0151] Step S5: When the target computing power node finishes processing the computing power operation task based on the operation code information and the task configuration information, display a second page;

[0152] Among them, the second page includes second log information, second resource information, and second node information. The second log information is the task log for completing the processing of the computing power operation task. The second resource information is the status information of the computing power resources used for completing the processing of the computing power operation task. The second node information is the computing node information for completing the processing of the computing power operation task.

[0153] Optionally, step S1: Receiving the computing power operation task information input by the user includes:

[0154] Step S11: Based on the code configuration interface, receive the operation code information input by the user;

[0155] Step S12: After performing security verification on the operation code information and passing the verification, generate a task configuration interface based on the operation code information;

[0156] Step S13: Based on the task configuration interface, receive the task configuration information input by the user.

[0157] Optionally, step S13: Based on the task configuration interface, receiving the task configuration information input by the user includes:

[0158] Step S131: Receive the mirror file imported by the user or selected in the task configuration interface. The mirror file is used to create and deploy the configuration environment for processing the computing power operation task;

[0159] Step S132: Based on the user input, determine the computing power resource information. The computing power resource information includes the computing power resource parameters for processing the computing power operation task.

[0160] Among them, the task configuration information includes the mirror file and the computing power resource information.

[0161] Optionally, step S2, receiving node selection information input by the user based on the computing power to run the task, includes:

[0162] Step S21, based on the user input, determining at least one computing unit and at least one configuration information, the at least one computing unit corresponding to the at least one configuration information one by one, the computing unit being used to run the running code in the configuration environment, and the configuration information being used to indicate the configuration parameters of the corresponding computing unit;

[0163] Among them, the target computing power node includes the at least one computing unit, and the node selection information includes the target computing power node and the at least one configuration information.

[0164] Optionally, the first resource information includes at least one of the following: the calculation loss value corresponding to the computing power running task, the processed duration of the computing power running task, the usage rate of the CPU memory space, the power used by the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the usage rate of the disk, and the usage rate of the memory;

[0165] The second resource information includes at least one of the following: the calculation loss value corresponding to the computing power running task, the processed duration of the computing power running task, the usage rate of the CPU memory space, the power used by the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the usage rate of the disk, and the usage rate of the memory.

[0166] After receiving the computing power running task information input by the user, the present invention determines the running code information and the task configuration information, and schedules the computing power running task to the target computing power node according to the node selection information input by the user, so that the target computing power node processes the computing power running task based on the running code information and the task configuration information, realizing the user's custom selection of heterogeneous computing power nodes and greatly improving the flexibility of the intelligent computing center to process computing power running tasks.

[0167] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements each process of the above embodiment of the computing power directional scheduling method of the intelligent computing center and can achieve the same technical effect. To avoid repetition, it will not be described in detail here. Among them, the computer-readable storage medium, such as a read-only memory (ROM for short), a random access memory (RAM for short), a magnetic disk or an optical disc, etc.

[0168] Another embodiment of the present invention further provides a computer program product. The computer program product is stored in a storage medium and is executed by at least one processor to implement each process of the above-described embodiment of the computing power directed scheduling method of the intelligent computing center, and can achieve the same technical effects. To avoid repetition, it will not be described here again.

[0169] It should be noted that in this document, the terms "include", "comprise" or any other variant 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 also elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of another identical element in the process, method, article or device comprising the element.

[0170] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions for causing a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0171] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the present invention and the claims, and all of them belong to the protection scope of the present invention.

Claims

1. A computing power directional scheduling method for an intelligent computing center, characterized in that The method includes: Step S1: Receive the computing power operation task information input by the user. The computing power operation task information includes operation code information and task configuration information. The operation code information is used to process the computing power operation task, and the task configuration information includes the computing power resource parameter values for processing the computing power operation task; Step S2: Receive the node selection information input by the user based on the computing power operation task. The node selection information is used to determine the target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes. The target computing power node is one of the multiple heterogeneous computing power nodes; Step S3: Schedule the computing power operation task to the target computing power node so that the target computing power node processes the computing power operation task based on the operation code information and the task configuration information.

2. The method according to claim 1, characterized in that, The method further includes: Step S4: When a task query request input by the user is received during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information, in response to the task query request, display a first page; Wherein, the task query request is used to query the status information of the computing power operation task. The first page includes the first log information, first resource information, and first node information corresponding to the computing power operation task. The first log information is the task log for processing the computing power operation task, the first resource information is the computing power resource usage status information for processing the computing power operation task, and the first node information is the computing node information for processing the computing power operation task.

3. The method according to claim 2, wherein After step S4: When a task query request input by the user is received during the process of the target computing power node processing the computing power operation task based on the operation code information and the task configuration information, and in response to the task query request, a first page is displayed, the method further includes: Step S5: When the target computing power node finishes processing the computing power operation task based on the operation code information and the task configuration information, display a second page; Wherein, the second page includes second log information, second resource information, and second node information. The second log information is the task log for completing the processing of the computing power operation task, the second resource information is the computing power resource usage status information for completing the processing of the computing power operation task, and the second node information is the computing node information for completing the processing of the computing power operation task.

4. The method according to claim 1, wherein Step S1: Receiving the computing power operation task information input by the user includes: Step S11: Based on the code configuration interface, receive the operation code information input by the user; Step S12: When the operation code information is subjected to security verification and the verification is passed, generate a task configuration interface based on the operation code information; Step S13: Based on the task configuration interface, receive the task configuration information input by the user.

5. The method according to claim 4, wherein Step S13: Based on the task configuration interface, receiving the task configuration information input by the user includes: Step S131: Receive the image file imported by the user or selected in the task configuration interface. The image file is used to create and deploy a configuration environment for processing the computing power operation task. Step S132: Determine the computing power resource information based on the user input. The computing power resource information includes the computing power resource parameters for processing the computing power operation task. Among them, the task configuration information includes the image file and the computing power resource information.

6. The method according to claim 5, wherein In step S2, receive the node selection information input by the user based on the computing power operation task, including: Step S21: Determine at least one computing unit and at least one configuration information based on the user input. The at least one computing unit and the at least one configuration information are in one-to-one correspondence. The computing unit is used to run the running code in the configuration environment, and the configuration information is used to indicate the configuration parameters of the corresponding computing unit. Among them, the target computing power node includes the at least one computing unit, and the node selection information includes the target computing power node and the at least one configuration information.

7. The method according to claim 3, wherein The first resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the used power of the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory. The second resource information includes at least one of the following: the calculation loss value corresponding to the computing power operation task, the processed duration of the computing power operation task, the utilization rate of the CPU memory space, the used power of the CPU, the temperature of the GPU, the utilization rate of the GPU, the utilization rate of the CPU, the utilization rate of the disk, and the utilization rate of the memory.

8. An arithmetic power directional scheduling device for an intelligent computing center, characterized in that, The device includes: A first receiving module, configured to receive the computing power operation task information input by the user. The computing power operation task information includes the running code information and the task configuration information. The running code information is used to process the computing power operation task, and the task configuration information includes the computing power resource parameter values for processing the computing power operation task. A second receiving module, configured to receive the node selection information input by the user based on the computing power operation task. The node selection information is used to determine the target computing power node for processing the computing power operation task among multiple heterogeneous computing power nodes. The target computing power node is one of the multiple heterogeneous computing power nodes. A processing module, configured to schedule the computing power operation task to the target computing power node, so that the target computing power node processes the computing power operation task based on the running code information and the task configuration information.

9. An electronic device, characterized in that, Includes: A processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, including computer instructions which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images