Multi-dimensional performance index-based computing power metering method and device for intelligent computing center cloud platform

Abstract

The invention provides a computing power metering method and device of an intelligent computing center cloud platform based on a multi-dimensional performance index, 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, obtaining a first task type of a first computing power operation task; s2, determining a plurality of resource indexes corresponding to the first task type and a weight coefficient corresponding to each resource index; s3, collecting a parameter value corresponding to each resource index in each of the N periods; s4, calculating a computing power value corresponding to the first computing power operation task based on the weight coefficient corresponding to each resource index and the parameter value corresponding to each resource index in N periods; and S5, generating a first verification data packet based on the first task identifier of the first computing power operation task, the weight coefficient corresponding to each resource index and the parameter value corresponding to each resource index in N periods. The computing power credibility of the intelligent computing center cloud platform can be greatly improved.

Description

Technical Field

[0001] This invention relates to the fields of intelligent computing centers, smart computing centers, computing infrastructure, and smart cloud technologies, specifically to a method and apparatus for measuring computing power on an intelligent computing center cloud platform based on multi-dimensional performance indicators. Background Technology

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

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

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

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

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

[0007] In current intelligent computing centers, accelerator cards are deployed to provide computing power, enabling the execution of computing tasks. Existing technologies require the measurement of the computing power consumed in executing these tasks to provide users with more accurate computing power services and improve billing accuracy. However, in current technologies, the measurement of computing power consumed by these tasks is merely a final value provided by the intelligent computing center. This final value may be modified or tampered with, resulting in low reliability of the computing power measurement for these tasks.

[0008] It is evident that since the emergence of intelligent computing centers, the reliability of computing power measurement for computing power operation tasks has been very low, and this problem has always been an urgent issue to be solved in this field. Summary of the Invention

[0009] This invention provides a method and apparatus for measuring computing power in an intelligent computing center cloud platform based on multi-dimensional performance indicators, in order to solve the problem of low reliability of computing power measurement for computing power operation tasks in the existing technology.

[0010] To solve the above problems, the present invention is implemented as follows: In a first aspect, the present invention provides a method for measuring computing power in an intelligent computing center cloud platform based on multi-dimensional performance indicators, comprising: Step S1: Obtain the first task type of the task running with the first computing power; Step S2: Determine multiple resource indicators corresponding to the first task type, and the weight coefficient corresponding to each resource indicator; Step S3: Collect the parameter values ​​corresponding to each resource indicator in each of the N cycles, where the N cycles are the cycles in the operation of the first computing power running task, and N is a positive integer greater than 1. Step S4: Calculate the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods. Step S5: Generate a first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator for the N periods.

[0011] In one embodiment, step S4 includes: Step S41: Based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods, calculate the intermediate computing power value of each period. The intermediate computing power value of the period is the product of the duration of the period and the first weighted value. The first weighted value is calculated by weighting the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the period. Step S42: Calculate the computing power value corresponding to the first computing power running task based on the intermediate computing power values ​​of the N cycles. The computing power value corresponding to the first computing power running task is the sum of the intermediate computing power values ​​of the N cycles.

[0012] In one embodiment, step S5 includes: Step S51: Construct a standardized score vector based on the parameter values ​​corresponding to each resource indicator in the N periods; Step S52: Generate the first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, the standardized score vector, the intermediate computing power value of each period and the first weighted value.

[0013] In one embodiment, the method further includes: Step S6: Receive a query request sent by the first device, wherein the query request includes a second task identifier; Step S7: Determine the second verification data packet based on the second task identifier, wherein the second verification data packet is a data packet including the second task identifier; Step S8: Send the second verification data packet to the first device.

[0014] In one embodiment, step S2 includes: Step S21: Based on the preset mapping relationship and the first task type, determine multiple resource indicators corresponding to the first computing power running task, and the weight coefficient corresponding to each resource indicator. The preset mapping relationship includes the mapping relationship of multiple task types, multiple resource indicators corresponding to each task type, and the weight coefficient corresponding to each resource indicator.

[0015] In one embodiment, the plurality of resource metrics includes at least two of CPU utilization, GPU utilization, read / write bandwidth, and network bandwidth.

[0016] Secondly, the present invention also provides a computing power metering device for an intelligent computing center cloud platform based on multi-dimensional performance indicators, comprising: The acquisition module is used to acquire the first task type of the task running on the first computing power. The first determining module is used to determine multiple resource indicators corresponding to the first task type, and the weight coefficient corresponding to each resource indicator. The acquisition module is used to acquire the parameter values ​​corresponding to each resource indicator in each of the N cycles, where the N cycles are the cycles in the operation of the first computing power running task, and N is a positive integer greater than 1. The calculation module is used to calculate the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods. The generation module is used to generate a first verification data packet based on the first task identifier of the task running the first computing power, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator for the N periods.

[0017] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, it implements the steps in the computing power measurement method based on multi-dimensional performance indicators of the intelligent computing center cloud platform as described in the first aspect above.

[0018] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the computing power measurement method for the intelligent computing center cloud platform based on multi-dimensional performance indicators as described in the first aspect above.

[0019] Fifthly, the present invention also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps in the computing power measurement method for the intelligent computing center cloud platform based on multi-dimensional performance indicators as described in the first aspect above.

[0020] This invention provides a method for measuring computing power in an intelligent computing center cloud platform based on multi-dimensional performance indicators, comprising: step S1, obtaining a first task type of a first computing power running task; step S2, determining multiple resource indicators corresponding to the first task type, and a weight coefficient corresponding to each resource indicator; step S3, collecting parameter values ​​corresponding to each resource indicator in each of N periods, wherein the N periods are the periods in the operation process of the first computing power running task, and N is a positive integer greater than 1; step S4, calculating the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter values ​​corresponding to each resource indicator in the N periods; and step S5, generating a first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter values ​​corresponding to each resource indicator in the N periods. In this way, the computing power value corresponding to the first computing power running task is calculated by using the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator over N periods. Then, the first verification data packet is generated by using the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator over N periods. This allows the computing power value of the first computing power running task to be verified through the first verification data packet, thereby greatly improving the credibility of computing power measurement of computing power running tasks in the intelligent computing center cloud platform. Attached Figure Description

[0021] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a computing power measurement method for an intelligent computing center cloud platform based on multi-dimensional performance indicators, provided by the present invention. Figure 2This is a flowchart illustrating how the first device provided by the present invention verifies the computing power value of a computing power operation task; Figure 3 This is a schematic diagram of multiple resource indicators provided by the present invention; Figure 4 This is a structural diagram of a computing power metering device based on multi-dimensional performance indicators for an intelligent computing center cloud platform provided by the present invention; Figure 5 This is a structural diagram of an electronic device provided by the present invention. Detailed Implementation

[0023] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] The “computing power” mentioned in this invention refers to: the ability of computer equipment or computing / data center to process information; the ability of computer hardware and software to work together to perform a certain computing requirement; the computing power to achieve the target result output by processing information data; and a new type of productivity that integrates information computing power, network carrying capacity, and data storage capacity, mainly providing services to society through computing power infrastructure.

[0025] The "computational power" (CP) described in this invention refers to the ability of a data center server to process data and output results. It is a comprehensive indicator of a data center's computing power, encompassing general computing power, supercomputing power, and intelligent computing power. The commonly used unit of measurement is floating-point operations per second (FLOPS, 1 EFLOPS = 10^18 FLOPS), with higher values ​​indicating stronger overall computing power. It is estimated that 1 EFLOPS is approximately the computing power output of 5 Tianhe-2A supercomputers, 500,000 mainstream server CPUs, or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 +CP 智能 +CP 超级 .

[0026] The "Network Power" (NP) mentioned in this invention refers to the performance of data transmission capability of computing facilities, which includes comprehensive capabilities such as network architecture, network bandwidth, transmission latency, intelligent management and scheduling, and involves network transmission within and between data centers. It is a comprehensive indicator for measuring network transmission scheduling capability.

[0027] The "Storage Power" (SP) described in this invention refers to the comprehensive capabilities of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon operation. It is a comprehensive indicator for measuring the data storage capacity of a data center, including external storage devices such as storage arrays and internal storage devices within servers. The commonly used unit of measurement for storage capacity is exabytes (EB, 1EB = 2^60 bytes), while the commonly used unit of measurement for performance is the number of read / write operations per second (IOPS / TB). Disaster recovery ratio is an important indicator of security and reliability.

[0028] The "computing infrastructure" mentioned in this invention refers to a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage capacity, enabling centralized computing, storage, transmission, and application of information.

[0029] The "new information infrastructure" mentioned in this invention refers to network infrastructure such as 5G networks, fiber optic broadband networks, backbone networks, international communication networks, and satellite internet; computing infrastructure such as data centers, general computing centers, intelligent computing centers, and supercomputing centers; and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0030] The “computing power” mentioned in this invention includes: general computing power, intelligent computing power, and supercomputing power.

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

[0032] The "intelligent computing power" mentioned in this invention refers to: a computing platform deployed on a large scale based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit) for various artificial intelligence innovative applications, such as natural language processing and machine vision.

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

[0034] The "intelligent computing center" described in this invention refers to a facility that, through the use of large-scale heterogeneous computing resources, including general-purpose computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.), primarily provides the necessary computing power, data, and algorithms for artificial intelligence applications (such as the development, training, and inference of deep learning models). The intelligent computing center encompasses facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enablement.

[0035] The "intelligent computing center cloud platform" mentioned in this invention, abbreviated as "intelligent computing cloud", refers to a cloud computing platform that integrates hardware and software resources based on an intelligent computing center.

[0036] The "intelligent computing center" mentioned in this invention includes, but is not limited to, "smart computing center".

[0037] The "intelligent computing center" mentioned in this invention, also known as an artificial intelligence computing center, is a type of computing infrastructure that provides computing power services, data services, and algorithm services required for artificial intelligence applications, based on artificial intelligence theory and adopting an artificial intelligence computing architecture.

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

[0039] The "supercomputing center" mentioned in this invention refers to a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters. It can provide large-scale computing, storage and network services and is widely used in aerospace, defense, oil exploration, climate modeling and genome sequencing and other application scenarios.

[0040] The “computing resources” mentioned in this invention refer to the technologies and facilities required for the development of the digital society that have the ability to compute, transmit, store and apply information, including but not limited to computing resources such as CPUs and GPUs, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and supporting and guaranteeing resources such as wind, fire, water and electricity.

[0041] The "model" mentioned in this invention includes, but is not limited to, "large language model" and "multimodal large model".

[0042] The "large language model" mentioned in this invention refers to a large-scale language model (LLM), which is a language model with a large number of parameters. It is designed to understand and generate human language, and is trained with a large amount of text data. It can perform a wide range of tasks, including text summarization, translation, and sentiment analysis.

[0043] The “Multimodal Large Models” mentioned in this invention refer to models that combine multimodal information such as text, images, videos, and audio for training, including but not limited to multimodal large language models.

[0044] The “computing power running task” mentioned in this invention refers to a specific workload or job executed on computing power resources that requires a certain amount of computing power support, usually involving complex data processing, numerical calculation, model training or simulation scenarios.

[0045] The "multi-dimensional performance indicators" mentioned in this invention refer to indicators of multiple types of resource parameters consumed by an intelligent computing center or intelligent computing center cloud platform during the execution of computing power tasks, such as indicators of CPU utilization, GPU utilization, storage read / write bandwidth, and network bandwidth.

[0046] The "acceleration card" described in this invention includes: GPU (Graphics Processing Unit), TPU (Tensor Processing Unit), NPU (Neural Network Processing Unit), etc.

[0047] Please see Figure 1 , Figure 1 This is a flowchart of a computing power measurement method based on multi-dimensional performance indicators for an intelligent computing center cloud platform provided by the present invention, such as... Figure 1 As shown, it includes the following steps: Step S1: Obtain the first task type of the first computing power running task.

[0048] The aforementioned first computing power operation task is a computing power operation task for which the intelligent computing center cloud platform provides computing power for execution. During the execution of the first computing power operation task, the intelligent computing center cloud platform measures the computing power consumed to obtain the computing power consumed in executing the first computing power operation task.

[0049] The first task type mentioned above refers to the task type of the first computing power execution task. It should be noted that there are various types of computing power execution tasks, and different types consume different resources to varying degrees. For example, computing power execution tasks of the model training type mainly consume GPU resources, storage resources, and network resources, while computing power execution tasks of the scientific computing type mainly consume CPU resources and GPU resources. Therefore, it is necessary to determine the task type of the first computing power execution task in order to determine the resource indicators that need to be collected.

[0050] Step S2: Determine the multiple resource indicators corresponding to the first task type, and the weight coefficient corresponding to each resource indicator.

[0051] The aforementioned resource indicators are those consumed during the execution of the first computing power task. It is understood that different task types consume different resources, and therefore require different resource collection methods. In this invention, the first task type is used to determine the multiple resource indicators that need to be collected.

[0052] Furthermore, for a computing power operation task, the proportion of different resources consumed varies. When calculating the computing power value of the computing power operation task, different weight coefficients need to be set for the different resource indicators collected to achieve accurate measurement of the computing power value. Therefore, in this invention, multiple resource indicators and corresponding weight coefficients are determined by the first task type to facilitate the subsequent collection of resource indicators and the calculation of the computing power value of the first computing power operation task.

[0053] Step S3: Collect the parameter values ​​corresponding to each resource indicator in each of the N cycles, where the N cycles are the cycles in the operation process of the first computing power running task, and N is a positive integer greater than 1.

[0054] The aforementioned N cycles represent the execution cycles of the first computing power task. By sequentially collecting resource metrics within each of the N cycles, the parameter values ​​of the resource metrics for the first computing power task throughout its entire execution process are obtained. It should be noted that while each cycle has the same duration, considering the limited storage space of the intelligent computing center cloud platform, the cycle setting needs to be based on the platform's storage capacity. For example, a relatively long cycle can typically be set, such as a cycle Δt of 300 seconds (s).

[0055] The parameter values ​​corresponding to the above resource indicators are the measured values ​​of the resource indicators within the period. It should be noted that the measured values ​​of different resource indicators are different and need to be collected specifically. For example, for utilization-related resource indicators (CPU utilization, GPU utilization, etc.), the collected parameter value is the ratio of the measured value to the maximum value; while for bandwidth-related resource indicators (such as storage bandwidth, network bandwidth, etc.), the collected parameter value is the ratio of the measured value to the theoretical peak value. In this way, the parameter value for each resource indicator is between [0,1], which facilitates the subsequent verification of the parameter values ​​of the resource indicators.

[0056] Step S4: Calculate the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods.

[0057] After obtaining the weight coefficient corresponding to each resource indicator based on the first task type, and collecting the parameter values ​​corresponding to each resource indicator within N periods, the computing power value corresponding to the first computing power running task is calculated based on the weight coefficient corresponding to each resource indicator and the parameter values ​​corresponding to each resource indicator within the N periods. This computing power value is a measurement value of the computing power consumed during the execution of the first computing power running task. The measurement value is obtained through multi-dimensional performance indicators to improve the accuracy of the measurement value.

[0058] In some implementations, the computing power value corresponding to the first computing power running task can be obtained by weighting the weight coefficient and the parameter value corresponding to each resource.

[0059] Step S5: Generate a first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator for the N periods.

[0060] The aforementioned first verification data packet is used to verify the computing power value of the first computing power running task. In this invention, the first verification data packet is generated by the first task identifier, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator. This allows the data in the first verification data packet to be retrieved for verification when it is necessary to verify the computing power value of the first computing power running task, thereby ensuring the credibility of the computing power value of the first computing power running task.

[0061] This invention provides a method for measuring computing power in an intelligent computing center cloud platform based on multi-dimensional performance indicators, comprising: step S1, obtaining a first task type of a first computing power running task; step S2, determining multiple resource indicators corresponding to the first task type, and a weight coefficient corresponding to each resource indicator; step S3, collecting parameter values ​​corresponding to each resource indicator in each of N periods, wherein the N periods are the periods in the operation process of the first computing power running task, and N is a positive integer greater than 1; step S4, calculating the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter values ​​corresponding to each resource indicator in the N periods; and step S5, generating a first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter values ​​corresponding to each resource indicator in the N periods. In this way, the computing power value corresponding to the first computing power running task is calculated by using the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator over N periods. Then, the first verification data packet is generated by using the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator over N periods. This allows the computing power value of the first computing power running task to be verified through the first verification data packet, thereby greatly improving the credibility of computing power measurement of computing power running tasks in the intelligent computing center cloud platform.

[0062] In one embodiment, step S4 includes: Step S41: Based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods, calculate the intermediate computing power value of each period. The intermediate computing power value of the period is the product of the duration of the period and the first weighted value. The first weighted value is calculated by weighting the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the period. Step S42: Calculate the computing power value corresponding to the first computing power running task based on the intermediate computing power values ​​of the N cycles. The computing power value corresponding to the first computing power running task is the sum of the intermediate computing power values ​​of the N cycles.

[0063] In step S4, the computing power value corresponding to the first computing power running task is calculated based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods. In order to accurately calculate the computing power value corresponding to the first computing power running task, the intermediate computing power value of each period is calculated first in this invention, and then the intermediate computing power values ​​of each period are summed to calculate the computing power value consumed in executing the first computing power running task.

[0064] The intermediate computing power value for each period is obtained by combining the duration of that period with the first weighted value. The first weighted value is calculated by weighting the weight coefficients and parameter values ​​corresponding to each resource indicator in that period.

[0065] For example, the calculation process of the first weighted value can be represented by the following formula: CEI i =W cpu ×Score cpu_i +W gpu ×Score gpu_i +W storage ×Score storage_i +W network ×Score network_i ; CEI in the formula i W represents the first weighted value of the i-th period. cpu W gpu W storage and W network These represent the weighting coefficients corresponding to different resource indicators, Score cpu_i The parameter value representing the CPU utilization in the i-th cycle, Score gpu_i The parameter value representing the GPU utilization in the i-th cycle, Score. storage_i The parameter value representing the storage bandwidth in the i-th cycle, Score network_i The parameter value represents the network bandwidth in the i-th period.

[0066] Furthermore, after calculating the first weighted value for each period, the product of the first weighted value and the duration is taken as the intermediate computing power value for that period, which can be expressed by the following formula: Q i =CEI i ×Δt; Q in the formula i Δt represents the intermediate computing power value in the i-th cycle, and Δt represents the duration of the cycle.

[0067] After calculating the intermediate computing power value for each cycle, the intermediate computing power values ​​for each cycle are summed to obtain the computing power value for the first computing power task. This can be expressed by the following formula: ; Q in the formula total This represents the computing power value of the first computing power task, and the unit can be "Comprehensive Computing Power Units per Second" (CEU·s).

[0068] The computing power value of the first computing power running task is calculated using the above formula. This computing power value can be used to directly perform operations such as billing for the first computing power running task.

[0069] In this invention, step S4 includes: Step S41, calculating the intermediate computing power value for each period based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator for the N periods, wherein the intermediate computing power value for each period is the product of the duration of the period and a first weighted value, the first weighted value being calculated by weighting the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator for the period; Step S42, calculating the computing power value corresponding to the first computing power running task based on the intermediate computing power values ​​for the N periods, wherein the computing power value corresponding to the first computing power running task is the sum of the intermediate computing power values ​​for the N periods. Thus, by first calculating the intermediate computing power value for each period and then summing the intermediate computing power values ​​for each period to obtain the computing power value of the first computing power running task, accurate measurement of computing power value from multiple performance indicators is achieved.

[0070] In one embodiment, step S5 includes: Step S51: Construct a standardized score vector based on the parameter values ​​corresponding to each resource indicator in the N periods; Step S52: Generate the first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, the standardized score vector, the intermediate computing power value of each period and the first weighted value.

[0071] The standardized score vector is used to characterize the parameter value corresponding to each resource indicator in each of the N cycles. The original data of the resource indicators during the execution of the first computing power operation task can be obtained through the standardized score vector. Thus, the computing power value of the first computing power operation task and the first weighted value in the calculation process can be verified through the original data, so as to greatly improve the credibility of the computing power value.

[0072] The standardized score vector can be represented as [Score] cpu_i Score gpu_i ,...].

[0073] In this invention, step S5 includes: step S51, constructing a standardized score vector based on the parameter values ​​corresponding to each resource indicator in the N periods; step S52, generating the first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, the standardized score vector, the intermediate computing power value of each period, and the first weighted value. Thus, the computing power value of the first computing power running task, as well as the first weighted value during the calculation process, can be verified through the standardized score vector in the first verification data packet. Similarly, the computing power value of the first computing power running task can be verified through the intermediate computing power value and the first weighted value of each period in the first verification data packet. All data can be mutually verified throughout the process, thereby greatly improving the reliability of the computing power value through the first verification data packet.

[0074] In one embodiment, the method further includes: Step S6: Receive a query request sent by the first device, wherein the query request includes a second task identifier; Step S7: Determine the second verification data packet based on the second task identifier, wherein the second verification data packet is a data packet including the second task identifier; Step S8: Send the second verification data packet to the first device.

[0075] Steps S6 to S8 above are steps for verifying the computing power value of the computing power running task. It should be noted that when the user or auditor believes that the computing power value of the computing power running task is questionable and needs to be verified, the user or auditor can send a query request through the first device to obtain the second verification data packet corresponding to the second computing power running task to be verified.

[0076] The second computing power operation task is a computing power operation task executed by the intelligent computing center cloud platform. After the intelligent computing center cloud platform executes the second computing power operation task, it generates and stores a second verification data packet corresponding to the second computing power operation task to facilitate the verification of the computing power value of the second computing power operation task. The second computing power operation task can be the first computing power operation task or other computing power operation tasks executed by the intelligent computing center cloud platform.

[0077] In this invention, such as Figure 2 As shown, a query request is sent by the first device, which includes a second task identifier, thereby receiving a second verification data packet sent by the intelligent computing center cloud platform. The data included in the second verification data packet is used to calculate the computing power value calculated locally by the first device. The computing power value is verified by checking whether the local computing power value matches the computing power value provided by the intelligent computing center cloud platform.

[0078] In some implementations, when the local computing power value and the computing power value provided by the intelligent computing center cloud platform do not match, the intermediate computing power value for each period is calculated by standardizing the score vector, and then verified in combination with the intermediate computing power value for each period included in the second verification data packet, thereby determining the period in which the computing anomaly exists.

[0079] In one embodiment, step S2 includes: Step S21: Based on the preset mapping relationship and the first task type, determine multiple resource indicators corresponding to the first computing power running task, and the weight coefficient corresponding to each resource indicator. The preset mapping relationship includes the mapping relationship of multiple task types, multiple resource indicators corresponding to each task type, and the weight coefficient corresponding to each resource indicator.

[0080] The aforementioned preset mapping relationships are pre-configured to determine the resource indicators and weight coefficients corresponding to different task types. Specifically, the preset mapping relationships include mapping relationships for different task types, which are used to determine the resource indicators and weight coefficients corresponding to each task type.

[0081] For example, the task types are Artificial Intelligence (AI) model training, big data analysis, and scientific computing. The weighting coefficients for different resource indicators in the AI ​​model training type can be expressed as: W=[CPU:0.05, GPU:0.60, Storage bandwidth:0.15, Network bandwidth:0.20], where CPU utilization has a weighting coefficient of 0.05, GPU utilization has a weighting coefficient of 0.60, storage bandwidth has a weighting coefficient of 0.15, and network bandwidth has a weighting coefficient of 0.20. The weighting coefficients for different resource indicators in the big data analysis type can be expressed as: W=[CPU:0.40, GPU:0.00, Storage bandwidth:0.40, Network bandwidth:0.20], and the weighting coefficients for different resource indicators in the scientific computing type can be expressed as: W=[CPU:0.50, GPU:0.30, Storage bandwidth:0.10, Network bandwidth:0.10].

[0082] In one embodiment, such as Figure 3 As shown, the multiple resource metrics include at least two of CPU utilization, GPU utilization, read / write bandwidth, and network bandwidth.

[0083] Among them, read / write bandwidth refers to storage read / write bandwidth, and its unit is MB / s; network bandwidth is measured in Gbps.

[0084] For example, this invention provides a verification case. Specifically, a user provides an AI training computing task to the intelligent computing center cloud platform. The weight coefficients for this computing task are W=[0.05,0.60,0.15,0.20]. Executing this computing task takes 2 hours, including a total of 240 cycles. The computing power value Q of this computing task is calculated by the intelligent computing center cloud platform. total =1,850,000 CEU·s. The result obtained through verification data is 1,850,000 CEU·s, which matches the calculation result of the intelligent computing center cloud platform. The computing power value of the computing power running task has been verified and is reliable data.

[0085] Please see Figure 4 , Figure 4 This is a structural diagram of a computing power metering device based on multi-dimensional performance indicators for an intelligent computing center cloud platform provided by the present invention, as shown below. Figure 4 As shown, the computing power metering device 400 of the intelligent computing center cloud platform based on multi-dimensional performance indicators includes: Module 401 is used to obtain the first task type of the first computing power running task; The first determining module 402 is used to determine multiple resource indicators corresponding to the first task type, and the weight coefficient corresponding to each resource indicator. The acquisition module 403 is used to acquire the parameter values ​​corresponding to each resource indicator in each of the N cycles, where the N cycles are the cycles in the operation of the first computing power running task, and N is a positive integer greater than 1. Calculation module 404 is used to calculate the computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods. The generation module 405 is used to generate a first verification data packet based on the first task identifier of the task running the first computing power, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator for the N periods.

[0086] In one embodiment, the computing module 404 includes: The first calculation unit is used to calculate the intermediate computing power value of each period based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods. The intermediate computing power value of the period is the product of the duration of the period and the first weighting value. The first weighting value is calculated by weighting the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the period. The second calculation unit is used to calculate the computing power value corresponding to the first computing power running task based on the intermediate computing power values ​​of the N cycles, wherein the computing power value corresponding to the first computing power running task is the sum of the intermediate computing power values ​​of the N cycles.

[0087] In one embodiment, the generation module 405 includes: A construction unit is used to construct a standardized score vector based on the parameter values ​​corresponding to each resource indicator in the N periods; The generation unit is used to generate the first verification data packet based on the first task identifier of the task running the first computing power, the weight coefficient corresponding to each resource indicator, the standardized score vector, the intermediate computing power value of each period, and the first weighted value.

[0088] In one embodiment, the computing power metering device 400 of the intelligent computing center cloud platform based on multi-dimensional performance indicators includes: A receiving module is used to receive a query request sent by a first device, wherein the query request includes a second task identifier; The second determining module is used to determine the second verification data packet based on the second task identifier, wherein the second verification data packet is a data packet including the second task identifier; The sending module is used to send the second verification data packet to the first device.

[0089] In one embodiment, the first determining module 402 includes: The determining unit is used to determine multiple resource indicators corresponding to the first computing power running task and the weight coefficient corresponding to each resource indicator based on a preset mapping relationship and the first task type. The preset mapping relationship includes a mapping relationship between multiple task types, multiple resource indicators corresponding to each task type, and the weight coefficient corresponding to each resource indicator.

[0090] In one embodiment, the plurality of resource metrics includes at least two of CPU utilization, GPU utilization, read / write bandwidth, and network bandwidth.

[0091] The computing power metering device for the intelligent computing center cloud platform based on multi-dimensional performance indicators provided by this invention can realize the various processes of the above-mentioned intelligent computing center cloud platform computing power metering method based on multi-dimensional performance indicators. The technical features are one-to-one and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0092] It should be noted that the computing power metering device based on multi-dimensional performance indicators in the intelligent computing center cloud platform of this invention can be a device, or it can be a component, integrated circuit, or chip in an electronic device.

[0093] The present invention also provides an electronic device, see [link to relevant documentation]. Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device includes a memory 501, a processor 502, and a program or instructions stored in the memory 501 that run on the processor 502. When the program or instructions are executed by the processor 502, they can achieve the following: Figure 1 The corresponding steps in the embodiment of the computing power measurement method based on multi-dimensional performance indicators of the intelligent computing center cloud platform, and the achievement of the same beneficial effect, will not be elaborated here.

[0094] The processor 502 can be a CPU, ASIC, FPGA, or GPU.

[0095] Those skilled in the art will understand that all or part of the steps of the above-described embodiment of the computing power measurement method based on multi-dimensional performance indicators for intelligent computing center cloud platforms can be implemented by hardware related to program instructions, and the program can be stored in a readable medium.

[0096] The present invention also provides a readable storage medium on which a computer program is stored, and which, when executed by a processor, can perform the above-described functions. Figure 1 The corresponding intelligent computing center cloud platform uses any step in the multi-dimensional performance index-based computing power measurement method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here. The storage medium mentioned includes, for example, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0097] The present invention also provides a computer program product, including computer instructions that, when executed by a processor, implement the above-described... Figure 1 The corresponding intelligent computing center cloud platform uses a multi-dimensional performance index-based computing power measurement method, and each process can achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0098] The terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses. Additionally, the use of "and / or" in this application indicates at least one of the connected objects, such as A and / or B and / or C, representing seven possibilities: A alone, B alone, C alone, both A and B present, both B and C present, both A and C present, and A, B, and C present.

[0099] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or second terminal device, etc.) to execute the methods of the various embodiments of this application.

[0101] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for measuring computing power based on multi-dimensional performance indicators of an intelligent computing center cloud platform, characterized in that, The method comprises: Step S1, acquiring a first task type of a first computing power running task; Step S2, determining a plurality of resource indicators corresponding to the first task type, and a weight coefficient corresponding to each resource indicator; Step S3, collecting a parameter value corresponding to each resource indicator in each period in N periods, the N periods being periods in a running process of the first computing power running task, and N being a positive integer greater than 1; Step S4, calculating a computing power value corresponding to the first computing power running task based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the N periods; Step S5, generating a first verification data packet based on a first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, and the parameter value corresponding to each resource indicator in the N periods.

2. The method of claim 1, wherein, The step S4 comprises: Step S41, calculating an intermediate computing power value of each period based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the period, the intermediate computing power value of the period being a sum of a duration of the period and a first weighted value, the first weighted value being obtained by weighted calculation based on the weight coefficient corresponding to each resource indicator and the parameter value corresponding to each resource indicator in the period; Step S42, calculating a computing power value corresponding to the first computing power running task based on the intermediate computing power values of the N periods, the computing power value corresponding to the first computing power running task being a sum of the intermediate computing power values of the N periods.

3. The method of claim 2, wherein, The step S5 comprises: Step S51, constructing a standardized score vector based on the parameter value corresponding to each resource indicator in the N periods; Step S52, generating the first verification data packet based on the first task identifier of the first computing power running task, the weight coefficient corresponding to each resource indicator, the standardized score vector, the intermediate computing power value of each period, and the first weighted value.

4. The method of any one of claims 1 to 3, wherein, The method further comprises: Step S6, receiving a query request sent by a first device, the query request comprising a second task identifier; Step S7, determining a second verification data packet based on the second task identifier, the second verification data packet being a data packet comprising the second task identifier; Step S8, sending the second verification data packet to the first device.

5. The method of any one of claims 1 to 3, wherein, The step S2 comprises: Step S21, determining a plurality of resource indicators corresponding to the first computing power running task and a weight coefficient corresponding to each resource indicator based on a preset mapping relationship and the first task type, the preset mapping relationship comprising a plurality of task types, a plurality of resource indicators corresponding to each task type, and a mapping relationship of a weight coefficient corresponding to each resource indicator.

6. The method of any one of claims 1 to 3, wherein, The plurality of resource indicators comprise at least two of CPU utilization, GPU utilization, read-write bandwidth, and network bandwidth.

7. A supercomputing center cloud platform computing power metering device based on multi-dimensional performance indicators, characterized by, The method comprises: An acquisition module is configured to acquire a first task type of a first computing power running task; A first determination module is configured to determine a plurality of resource indicators corresponding to the first task type, and a weight coefficient corresponding to each resource indicator; The collection module is configured to collect a parameter value corresponding to each resource index in each of N cycles, the N cycles being cycles in a running process of the first computing power running task, and N being a positive integer greater than 1. The calculation module is configured to calculate a computing power value corresponding to the first computing power running task based on a weight coefficient corresponding to each resource index and the parameter value corresponding to each resource index in the N cycles. The generation module is configured to generate a first verification data packet based on a first task identifier of the first computing power running task, the weight coefficient corresponding to each resource index, and the parameter value corresponding to each resource index in the N cycles.

8. An electronic device, comprising: The processor, the memory, and a program stored on the memory and executable on the processor, the program being executed by the processor to implement the steps of the intelligent computing center cloud platform based multi-dimensional performance index computing power metering method according to any one of claims 1 to 6. The computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the steps of the intelligent computing center cloud platform based multi-dimensional performance index computing power metering method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the steps of the intelligent computing center cloud platform based multi-dimensional performance index computing power metering method according to any one of claims 1 to 6.

10. A computer program product, characterised in that, The computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the steps of the intelligent computing center cloud platform based multi-dimensional performance index computing power metering method according to any one of claims 1 to 6.

Images