Trusted metering method and device for large-scale distributed computing power of intelligent computing center cloud platform

Abstract

The application provides a trusted metering method and device for large-scale distributed computing power of an intelligent computing center cloud platform, and relates to the technical fields of intelligent computing centers, intelligent computing centers, computing power infrastructure and intelligent computing cloud technology, and the method comprises the following steps: S1, receiving N pieces of summary data of first periods sent by edge side devices, the N pieces of summary data of first periods comprising parameter values of a plurality of resource parameters collected by the edge side devices in N first periods during the running of a first computing power running task, and the N pieces of summary data of first periods being data stored by a block chain; S2, determining a weight coefficient of each resource parameter in a plurality of resource parameters corresponding to the type of the first computing power running task; and S3, calculating a computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values of the plurality of resource parameters in the N first periods. The application can greatly improve the computing power metering credibility of the computing power running task in the intelligent computing center cloud platform.

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 reliable measurement method and device for large-scale distributed computing power of an intelligent computing center cloud platform. Background Technology

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

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

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

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

[0006] "Computing power" is the core of "intelligent computing center" and "smart computing center". It is the ability of computer equipment or computing / data center to process 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 can distribute these tasks across multiple edge devices in a distributed manner, allowing them to collectively execute the tasks. However, in these existing technologies, the measurement of computing power consumed by these tasks is achieved through the edge devices, while the intelligent computing center only receives a single measurement value from these devices. This 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 reliable measurement method and apparatus for large-scale distributed computing power of an intelligent computing center cloud platform, in order to solve the problem of low reliability of computing power measurement for computing power running tasks in the prior art.

[0010] To solve the above problems, the present invention is implemented as follows:

[0011] In a first aspect, the present invention provides a reliable measurement method for the large-scale distributed computing power of an intelligent computing center cloud platform, comprising:

[0012] Step S1: Receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1.

[0013] Step S2: Determine the weight coefficient of each of the multiple resource parameters corresponding to the type of the first computing power running task;

[0014] Step S3: Calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first cycles.

[0015] In one embodiment, the method further includes:

[0016] Step S4: Generate the first Merkle root data for the first computing power running task based on the first hash value corresponding to the digest data of each first period;

[0017] Step S5: Store the first Merkle root data using blockchain.

[0018] In one embodiment, step S5 includes:

[0019] Step S51: Generate first evidence data based on the identifier of the first computing power running task, the computing power value of the first computing power running task, the start time and end time of the first computing power running task, and the first Merkle root data;

[0020] Step S52: Store the first evidence data based on the blockchain.

[0021] In one embodiment, the first Merkle root data is used to verify with the second Merkle root data, which is data calculated by the first device based on N second hash values. The N second hash values ​​are calculated by the first device based on the parameter values ​​of multiple resource parameters in N second periods. The parameter values ​​of the multiple resource parameters in N second periods are obtained by the first device from the blockchain based on the identifier of the task run by the first computing power.

[0022] In one embodiment, step S2 includes:

[0023] Step S21: Determine the weight coefficient of each resource parameter among the multiple resource parameters corresponding to the type of the first computing power running task based on the preset mapping relationship. The preset mapping relationship includes the mapping relationship of the weight coefficient of each resource parameter among the multiple resource parameters corresponding to each type.

[0024] In one embodiment, the summary data of the i-th first cycle among the N first cycles includes the identifier of the first computing power running task, the identifier of the edge device, the start and end times of the i-th first cycle, the hash value of the i-th first cycle, and the standardized score vector, wherein the standardized score vector of the i-th first cycle is constructed based on the parameter values ​​of multiple resource parameters of the i-th first cycle, i is a positive integer greater than or equal to 1, and i is less than N.

[0025] Secondly, the present invention also provides a trusted metering device for large-scale distributed computing power of an intelligent computing center cloud platform, comprising:

[0026] The receiving module is used to receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1.

[0027] The determining module is used to determine the weight coefficient of each of the plurality of resource parameters corresponding to the type of the first computing power running task;

[0028] The metering module is used to calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first periods.

[0029] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps in the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0030] 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 trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0031] Fifthly, the present invention also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps in the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform as described in the first aspect above.

[0032] This invention provides a reliable method for measuring the large-scale distributed computing power of an intelligent computing center cloud platform, comprising: Step S1, receiving N first-cycle summary data sent by an edge device, wherein the N first-cycle summary data includes parameter values ​​of multiple resource parameters collected by the edge device during the execution of a first computing power task within the N first cycles, and the N first-cycle summary data is data stored on a blockchain, where N is a positive integer greater than 1; Step S2, determining the weight coefficient of each of the multiple resource parameters corresponding to the type of the first computing power task; Step S3, calculating the computing power value of the first computing power task based on the weight coefficient of each resource parameter and the parameter values ​​of the multiple resource parameters in the N first cycles. In this way, by generating summary data from the parameter values ​​of multiple resource parameters collected within the N first cycles and storing the N first-cycle summary data on a blockchain, reliable measurement of computing power can be achieved, greatly improving the reliability of computing power measurement for computing power tasks in the intelligent computing center cloud platform. Attached Figure Description

[0033] 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.

[0034] Figure 1 This is a flowchart of a reliable measurement method for large-scale distributed computing power of an intelligent computing center cloud platform, provided by the present invention.

[0035] Figure 2 This is a schematic diagram of multiple resource parameters provided by the present invention;

[0036] Figure 3 This is a flowchart provided by the present invention for verifying the computing power value of a computing power running task;

[0037] Figure 4 This is a flowchart of a reliable measurement method for large-scale distributed computing power of an intelligent computing center cloud platform applied to edge devices, provided by the present invention.

[0038] Figure 5 This is a flowchart of a reliable measurement method for large-scale distributed computing power applied to a cloud platform of an intelligent computing center for a first device, provided by the present invention.

[0039] Figure 6 This is a structural diagram of a trusted metering device for large-scale distributed computing power of an intelligent computing center cloud platform provided by the present invention;

[0040] Figure 7 This is a structural diagram of a trusted metering device for large-scale distributed computing power of an intelligent computing center cloud platform provided by the present invention;

[0041] Figure 8 This is a structural diagram of a trusted metering device for large-scale distributed computing power of an intelligent computing center cloud platform provided by the present invention;

[0042] Figure 9 This is a structural diagram of an electronic device provided by the present invention. Detailed Implementation

[0043] 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.

[0044] 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.

[0045] 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 .

[0046] 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.

[0047] 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.

[0048] 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.

[0049] 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.

[0050] The “computing power” mentioned in this invention includes: general computing power, intelligent computing power, and supercomputing power.

[0051] 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.

[0052] 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.

[0053] 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.

[0054] 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.

[0055] 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.

[0056] The "intelligent computing center" mentioned in this invention includes, but is not limited to, "smart computing center".

[0057] 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.

[0058] 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.

[0059] 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.

[0060] 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.

[0061] The "model" mentioned in this invention includes, but is not limited to, "large language model" and "multimodal large model".

[0062] 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.

[0063] 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.

[0064] 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.

[0065] The "trusted measurement" described in this invention refers to: storing the resource parameters consumed during the execution of computing power tasks, measuring the computing power value based on the resource parameters, and verifying the computing power value through the stored data, thereby realizing the trustworthiness of the computing power value and achieving the effect of trusted measurement of computing power tasks.

[0066] The "acceleration card" described in this invention includes: GPU (Graphics Processing Unit), TPU (Tensor Processing Unit), NPU (Neural Network Processing Unit), etc.

[0067] It should be noted that reliable measurement of computing power is currently a very challenging "bottleneck" problem in the global digital infrastructure field. Its difficulty far exceeds that of measurement in other fields such as electricity, traffic, and storage. The core reason lies in the multidimensionality, dynamism, and complexity of computing power.

[0068] For example, the difficulty of computing power measurement can be understood by comparing measurement methods in different fields.

[0069] For electricity metering, only one core indicator is needed—how much work was done (kilowatt-hours). Due to the stable and predictable nature of work done, accurate measurement can be achieved with a single electricity meter, and its consistency and reliability can be guaranteed by the laws of physics.

[0070] For network traffic metering, it is only necessary to determine how much data (bytes) was transmitted. Although the transmission may be affected by network conditions, the data packets are countable, the protocol is standard, the metering is relatively direct, and the metering results are reliable.

[0071] As for computing power measurement, the purpose of measurement is "how many effective calculations have been completed". Measuring computing power is like a black box. Data is input into the black box, and the black box outputs results, but the process in between is extremely complex and difficult to measure with a single, universal indicator. Because the measurement process is complex, the more factors that need to be considered in the process, the more difficult it is to achieve reliable measurement.

[0072] Based on the applicant's repeated testing and experiments on the intelligent computing center's cloud platform, the following are some of the core reasons why reliable measurement of computing power is extremely difficult:

[0073] 1. Computing power exhibits "multidimensionality" and "heterogeneity".

[0074] Computing power, since it is not a simple physical quantity, cannot be compared with homogeneous indicators such as electricity or network traffic.

[0075] For example, computing power can be provided by different types of accelerator cards, such as CPUs, GPUs, and NPUs. Different accelerator cards have completely different architectures and are better suited for different tasks. Using "FLOPS" (floating-point operations) to measure CPU performance to measure a GPU that is good at artificial intelligence (AI) inference is as unreasonable as using "load capacity" to measure the performance of a sports car, and it is difficult to achieve an accurate measurement.

[0076] At the same time, due to the different types of computing tasks, such as scientific computing, AI model training, video rendering, and database querying, the "effective computing power" that the same accelerator card can provide can vary greatly when executing different types of computing tasks. For example, an accelerator card that scores highly in matrix operations may perform only moderately when handling complex logical operations and may not achieve the same effect as when handling matrix operations.

[0077] For the reasons mentioned above, the field of intelligent computing center cloud platforms still lacks a standard unit that is universally recognized by the industry as "kilowatt-hour" and can measure "effective computing workload" across platforms and tasks.

[0078] 2. The performance indicators of computing power consumption are "dynamic" and "fluctuating".

[0079] During the execution of computing power tasks, the computing power output by the intelligent computing center cloud platform is not a constant value, but a fluctuating value, which is affected by many dynamic factors.

[0080] For example, the hardware is affected by the software stack and drivers. Different operating systems, drivers, compilers, and mathematical libraries (such as CUDA, oneDNN, and other databases) can greatly affect the actual performance of the hardware. A minor driver update can lead to a significant performance improvement or decrease.

[0081] Computing power is also affected by workload characteristics. Even for the same task, the size, shape, and precision (such as FP32, FP16, INT8 precision) of the input data can lead to huge differences in computing efficiency.

[0082] The influence of the system environment, such as memory bandwidth, PCIe bandwidth, storage I / O, heat dissipation conditions, and other concurrent tasks, can compete for system resources, leading to unstable computing power output.

[0083] Furthermore, the computing power output suffers from the "barrel effect," meaning that the final performance of the intelligent computing center cloud platform depends on the slowest component in the entire system (memory, cache, bus, etc.), rather than just the peak computing power of the accelerator card.

[0084] 3. The "externalities (or intrusiveness)" and "overhead (or consumption)" of the process of measuring computing power.

[0085] Measuring computing power is a complex systemic problem that relies on external devices or services, such as performance counters, instrumentation, and proxies, that are unrelated to the execution of computing power tasks.

[0086] Among them, performance counters are hardware configurations of the intelligent computing center and are an important source of data for resource parameters. However, performance counters themselves incur overhead when collecting or reading parameters, requiring a certain amount of computing power. Moreover, for resource parameters that are too low-level (such as the number of instructions or cache hit rate), it is difficult to directly map them to "effective computing workload," and it is necessary to additionally calculate the relationship between the parameter and "effective computing workload."

[0087] For instrumentation and proxying, a common practice is to run a standard benchmark program (such as MLPerf) as a proxy to estimate the system's computing power. However, this is a sampling method and cannot represent all real-world tasks, so its accuracy is generally low. Moreover, running benchmarks itself incurs overhead, requiring a certain amount of computing power.

[0088] Furthermore, it is difficult to measure computing power online or without loss, meaning that it is impossible to obtain an absolutely accurate computing power value in real time and without interrupting operations, just like reading an electricity meter.

[0089] 4. The susceptibility to fraud and cheating is a core issue of "reliable measurement".

[0090] In scenarios such as computing power leasing and trading, reliable measurement is directly related to economic interests, and therefore faces severe challenges from cheating.

[0091] For example, at the software level, computing power providers can deliberately provide under-optimized drivers or libraries, or obtain false high scores by "tuning" system parameters in benchmark tests, while performance drops sharply in actual customer tasks, and users cannot obtain services corresponding to the real computing power.

[0092] Hardware fraud, such as modified hardware, substandard products (e.g., disguising consumer-grade graphics cards as professional cards), and overselling virtualization resources, can lead to insufficient computing power due to hardware fraud.

[0093] In existing technologies, the verification cost is extremely high. In order to verify whether the computing power provided by the supplier is real, the demand side may need to deploy a complex test environment, which is costly and impractical.

[0094] 5. Fragmentation of standards and ecosystems

[0095] There is a lack of an authoritative body for computing power measurement: Unlike the International Bureau of Weights and Measures (BIPM) which defines "meter" and "second", the computing power field lacks a globally recognized authoritative body to develop and certify measurement standards.

[0096] In summary, the great difficulty of reliable computing power measurement stems from the need to use one or a few simple, static numbers to describe the effective output of a highly complex, multidimensional, dynamic system that is strongly dependent on the environment under a specific task.

[0097] It is evident that computing power metering is not merely a technical issue, but a complex systems engineering project involving hardware architecture, software engineering, metrology, economics, and trust mechanisms. As heterogeneous computing and AI computing power become new productive forces, resolving the issue of reliable computing power metering is crucial for building a fair and efficient computing power market and promoting the development of the digital economy. Currently, the industry is exploring this through standardized benchmark tests such as MLPerf, Trusted Execution Environments (TEEs), and technologies such as blockchain, but there is still a long way to go before achieving the ideal "computing power meter."

[0098] This invention provides a solution for the reliable measurement of computing power. For details, please refer to [link to relevant documentation]. Figure 1 , Figure 1 This is a flowchart of a reliable measurement method for large-scale distributed computing power of an intelligent computing center cloud platform, provided by the present invention. Figure 1 As shown, it includes the following steps:

[0099] Step S1: Receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1.

[0100] The aforementioned edge-side devices are devices used to provide computing power to execute computing tasks. After the user uploads the first computing task to the intelligent computing center cloud platform, the intelligent computing center cloud platform allocates the first computing task to the edge-side devices, and then the edge-side devices provide computing power to execute the first computing task.

[0101] It should be noted that since the first computing power operation task is executed by edge devices, the edge devices need to collect their resource parameters. In some implementations, the process of running the first computing power operation task is divided into N first cycles. The parameter values ​​of multiple resource parameters in each first cycle are collected to generate summary data, which is then sent to the intelligent computing center cloud platform. This allows the intelligent computing center cloud platform to determine the parameter values ​​of multiple resource parameters in each first cycle based on the summary data, and then calculate the computing power value based on the parameter values ​​of multiple resource parameters.

[0102] The N first-period summary data are stored on the blockchain, ensuring their credibility. When subsequent verification of the computing power value of the intelligent computing center cloud platform is required, the stored summary data is retrieved from the blockchain to determine the reliability of the computing power value.

[0103] In this way, since existing technologies do not consider the need for trusted verification, the present invention creatively generates summary data by collecting parameter values ​​of multiple resource parameters in N first cycles, and stores the summary data of N first cycles through blockchain, thereby enabling trusted measurement of computing power.

[0104] In some implementations, the summary data can be represented as: Summary = {TaskID, NodeID, Timestamp_Start, Timestamp_End, S, Hash(Raw_Data)}, where TaskID represents the identifier of the task running by the first computing power, NodeID represents the identifier of the edge device, Timestamp_Start represents the start time of the first cycle, Timestamp_End represents the end time of the first cycle, S represents the normalized score vector, and Hash(Raw_Data) represents the hash value generated by the parameter values ​​of multiple resource parameters.

[0105] In some implementations, there can be one or more edge devices. When there are multiple edge devices, it is necessary to receive summary data from each edge device, calculate the computing power value of each edge device based on the summary data, and then sum them up to obtain the total computing power value of the first computing power running task.

[0106] Step S2: Determine the weight coefficient of each of the multiple resource parameters corresponding to the type of the first computing power running task.

[0107] 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 parameters that need to be collected.

[0108] Furthermore, for a computing power operation task, the proportion of different resources consumed varies. When calculating the computing power value of the task, different weighting coefficients need to be set for the different resource parameters collected to achieve accurate measurement of the computing power value. Therefore, in this invention, multiple resource parameters and corresponding weighting coefficients are determined through a first task type to facilitate the subsequent collection of resource parameters and the calculation of the computing power value of the first computing power operation task.

[0109] In some implementations, multiple resource parameters can be as follows: Figure 2As shown, the parameters include those for computing units, storage units, and network units.

[0110] The computing unit metrics include CPU utilization (CPU_Util), GPU utilization (GPU_Util), and GPU memory utilization (GPU_Mem_Util). CPU utilization refers to the percentage of non-idle time of all CPU cores within the sampling period, reflecting the processor's workload. GPU utilization refers to the percentage of active periods (%) of GPU streaming multiprocessors (SMs) executing computing tasks within the sampling period, and is a key indicator for measuring GPU computing load. GPU memory utilization refers to the ratio (%) of GPU memory currently occupied by computing tasks to the total GPU memory, reflecting the pressure on model size and data caching.

[0111] Storage unit metrics include storage IOPS (IO_IOPS), storage bandwidth (IO_BW), and storage I / O latency (IO_Latency). Storage IOPS refers to the number of read and write operations completed per second, used to measure the storage system's ability to handle random small file read and write requests. Storage bandwidth refers to the amount of data successfully read and written per second (MB / s), used to measure the storage system's throughput ability to handle large file sequential read and write. Storage I / O latency refers to the average time (ms) from initiating an I / O request to receiving a response, directly affecting data loading speed.

[0112] Network unit metrics include parameters such as network bandwidth utilization (Net_BW_Util), network latency (Net_Latency), network jitter (Net_Jitter), and network packet loss rate (Net_PacketLoss). Among them, network bandwidth utilization refers to the ratio (%) of the current actual transmission rate of a node's network interface to its theoretical maximum bandwidth, reflecting the saturation level of the network channel; network latency usually refers to "round-trip time (RTT)," which is the time (ms) required for a data packet to travel from the local node to the target node and back, and is a key factor affecting the efficiency of distributed task collaboration; network jitter refers to the standard deviation (ms) of the variation in network latency, used to measure the stability of network latency, and high jitter will affect the performance of real-time applications; network packet loss rate refers to the percentage (%) of the number of data packets lost during data transmission out of the total number of data packets sent. Packet loss will trigger retransmission, which will seriously reduce the effective bandwidth and increase latency.

[0113] Furthermore, considering the different ranges of various resource parameters, to facilitate the subsequent calculation of computing power values ​​and the storage of summary data, after obtaining the original parameter values ​​of the resource parameters, the original parameter values ​​can be normalized to obtain parameter values ​​within the range of [0,1]. In some implementations, the original parameter values ​​can be mapped to the range of [0,1] using a preset normalization formula (Z-score normalization).

[0114] In some implementations, a standardized score vector can be constructed based on multiple collected resource parameters, where each element in the vector represents the parameter value of a resource parameter. For example, the standardized score vector S can be represented as: S = [s_cpu, s_gpu_util, s_gpu_mem, s_iops, s_io_bw, s_io_lat, s_net_bw, s_net_lat, s_net_jit, s_net_loss], where s_cpu represents the parameter value of CPU utilization, s_gpu_util represents the parameter value of GPU utilization, s_gpu_mem represents the parameter value of GPU memory utilization, s_iops represents the parameter value of storage IOPS, s_io_bw represents storage bandwidth, s_io_lat represents storage I / O latency, s_net_bw represents the parameter value of network bandwidth utilization, s_net_lat represents the parameter value of network latency, s_net_jit represents the parameter value of network jitter, and s_net_loss represents the parameter value of network packet loss rate.

[0115] In some implementations, the weight coefficients of resource parameters corresponding to different task types are different, which can be represented by a weight vector. Each element in the weight vector represents a weight coefficient, and the number of elements in the weight vector corresponds one-to-one with the number of elements in the standardized score vector. In this way, the computing power value can be calculated from the weight vector and the standardized score vector.

[0116] For example, for AI model training tasks, the weight vector W_ai is represented as W_ai = [0.05, 0.50, 0.10, 0.00, 0.05, 0.00, 0.10, 0.10, 0.05, 0.05], meaning that the GPU and network unit metrics are given the highest weights, while the storage unit metrics are given lower weights. For big data analysis tasks, the weight vector W_data is represented as W_data = [0.30, 0.00, 0.00, 0.20, 0.20, 0.10, 0.10, 0.05, 0.05, 0.00], meaning that the CPU's computation and storage throughput are given higher weights. For scientific computing tasks, the weight vector W_hpc is represented as W_hpc = [0.40, 0.20, 0.00, 0.00, 0.10, 0.00, 0.20, [0.10, 0.00, 0.00], which means assigning high weights to CPU and network bandwidth.

[0117] Step S3: Calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first cycles.

[0118] In this invention, the computing power value corresponding to the first computing power running task is calculated based on the weight coefficient corresponding to each resource parameter and the parameter value corresponding to each resource indicator in N first periods. In some embodiments, to improve the accuracy of the calculation, the intermediate computing power value of each first period can be calculated based on the weight coefficient corresponding to each resource parameter and the parameter value corresponding to each resource indicator in the N first periods, and then the intermediate computing power values ​​of each first period are summed to calculate the computing power value consumed in executing the first computing power running task.

[0119] In some implementations, the intermediate computing power value for each first cycle can be calculated using a weight vector and a standardized score vector, and then the intermediate computing power values ​​for each first cycle can be summed to calculate the computing power value consumed in executing the first computing power running task.

[0120] For example, the calculation process of the computing power value Total_Compute for a computing power task can be expressed by the following formula:

[0121] Total_Compute = Σ(all edge-side devices)Σ(all first cycles)(W·S)*Δt;

[0122] Wherein, (W·S) is the dot product of the weight vector and the standardized score vector, representing the "comprehensive computing power consumption rate" of the edge device in this period. Multiplying it by the time window length Δt of the period, we get the computing power value of the edge device. Finally, we sum the computing power values ​​of all edge devices to obtain the computing power value of the first computing power running task.

[0123] This invention provides a reliable method for measuring the large-scale distributed computing power of an intelligent computing center cloud platform, comprising: Step S1, receiving N first-cycle summary data sent by an edge device, wherein the N first-cycle summary data includes parameter values ​​of multiple resource parameters collected by the edge device during the execution of a first computing power task within the N first cycles, and the N first-cycle summary data is data stored on a blockchain, where N is a positive integer greater than 1; Step S2, determining the weight coefficient of each of the multiple resource parameters corresponding to the type of the first computing power task; Step S3, calculating the computing power value of the first computing power task based on the weight coefficient of each resource parameter and the parameter values ​​of the multiple resource parameters in the N first cycles. In this way, by generating summary data from the parameter values ​​of multiple resource parameters collected within the N first cycles and storing the N first-cycle summary data on a blockchain, reliable measurement of computing power can be achieved, greatly improving the reliability of computing power measurement for computing power tasks in the intelligent computing center cloud platform.

[0124] In one embodiment, the method further includes:

[0125] Step S4: Generate the first Merkle root data for the first computing power running task based on the first hash value corresponding to the digest data of each first period;

[0126] Step S5: Store the first Merkle root data using blockchain.

[0127] The aforementioned first Merkle root data is used to verify each digest data and the computing power value of the first computing power running task. Mr. Qi, the first Merkle root data is obtained through the first hash value corresponding to the digest data of each first period. Specifically, it can be that the first hash value of each first period is generated first, and then the first Merkle root data is generated through the first hash value corresponding to the digest data of each first period.

[0128] In some implementations, a Merkle tree can be constructed based on the summary data of all edge devices. Each leaf node in the Merkle tree corresponds to the hash value of a summary data, which can be represented as H_i = Hash(Summary_i). Here, Hash() represents calculating the hash value, and Summary_i represents the i-th summary data.

[0129] Furthermore, the first Merkle root data can be generated based on each leaf node in the Merkle tree.

[0130] In this invention, the method further includes: step S4, generating first Merkle root data for the first computing power running task based on the first hash value corresponding to the digest data of each first period; and step S5, storing the first Merkle root data using a blockchain. Thus, by generating first Merkle root data for the first computing power running task using the first hash value corresponding to the digest data of each first period, and storing the first Merkle root data, the method facilitates verification of each digest data and the computing power value of the first computing power running task using the first Merkle root.

[0131] In one embodiment, step S5 includes:

[0132] Step S51: Generate first evidence data based on the identifier of the first computing power running task, the computing power value of the first computing power running task, the start time and end time of the first computing power running task, and the first Merkle root data;

[0133] Step S52: Store the first evidence data based on the blockchain.

[0134] The aforementioned first evidence-based data is used to preserve relevant data of the first computing power operation task, so as to facilitate subsequent verification of data such as the computing power value of the first computing power operation task. Specifically, the first evidence-based data is generated using the identifier of the first computing power operation task, the computing power value of the first computing power operation task, the start and end times of the first computing power operation task, and the first Merkle root data. Thus, the identifier of the first computing power operation task, the computing power value of the first computing power operation task, the start and end times of the first computing power operation task, and the first Merkle root data can be obtained through the first evidence-based data. Furthermore, the first Merkle root data can be used to verify each summary data and the computing power value of the first computing power operation task, and the start and end times of the first computing power operation task can also be used to verify the data queried by the user.

[0135] In some implementations, the first evidence data Evidence can be represented as: Evidence = {TaskID,Total_Compute, Time_Window_Start, Time_Window_End, MR}, where TaskID represents the identifier of the first computing power running task, Total_Compute represents the computing power value of the first computing power running task, Time_Window_Start and Time_Window_End represent the start and end times of the first computing power running task, respectively, and MR represents the first Merkle root data.

[0136] In one embodiment, the first Merkle root data is used to verify with the second Merkle root data, which is data calculated by the first device based on N second hash values. The N second hash values ​​are calculated by the first device based on the parameter values ​​of multiple resource parameters in N second periods. The parameter values ​​of the multiple resource parameters in N second periods are obtained by the first device from the blockchain based on the identifier of the task run by the first computing power.

[0137] It should be noted that the process of verifying the computing power value of the first computing power task is implemented by the user or the auditing party, such as... Figure 3 As shown, the user or auditor obtains the summary data of N second periods and the first Merkle root data of the first computing power running task through the first device; calculates N second hash values ​​based on the summary data of the N second periods, and calculates the second Merkle root data based on the N second hash values; then generates a verification result based on the first Merkle root data and the second Merkle root data, and determines whether the computing power value of the first computing power running task is reliable through the verification result.

[0138] Specifically, if the first Merkle root data and the second Merkle root data match, the computing power value of the first computing power running task is considered reliable; if the first Merkle root data and the second Merkle root data do not match, the computing power value of the first computing power running task is considered unreliable.

[0139] In one embodiment, step S2 includes:

[0140] Step S21: Determine the weight coefficient of each resource parameter among the multiple resource parameters corresponding to the type of the first computing power running task based on the preset mapping relationship. The preset mapping relationship includes the mapping relationship of the weight coefficient of each resource parameter among the multiple resource parameters corresponding to each type.

[0141] The aforementioned preset mapping relationships are pre-configured to determine the resource parameters 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 parameters and weight coefficients corresponding to each task type.

[0142] In one embodiment, the summary data of the i-th first cycle among the N first cycles includes the identifier of the first computing power running task, the identifier of the edge device, the start and end times of the i-th first cycle, the hash value of the i-th first cycle, and the standardized score vector, wherein the standardized score vector of the i-th first cycle is constructed based on the parameter values ​​of multiple resource parameters of the i-th first cycle, i is a positive integer greater than or equal to 1, and i is less than N.

[0143] The standardized score vector is used to characterize the parameter value corresponding to each resource parameter in each of the N first cycles. The original data of the resource parameters 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.

[0144] For example, taking the execution of a large-scale distributed AI model training task (TaskID: T-1001) on the intelligent computing center cloud platform as an example, the computing power value of this task is verified as follows:

[0145] The intelligent computing center cloud platform ran a large-scale distributed AI model training task, which utilized the GPU nodes of 100 edge devices and lasted for 1 hour.

[0146] Generate summary data. Each edge agent on the edge device generates 1 summary data per minute (i.e., one first cycle Δt=60s), for a total of 100 edge devices * 60 minutes = 6000 summary data.

[0147] The intelligent computing center cloud platform loads the W_ai weight vector, weights and sums 6000 summary data, and calculates the total computing power consumption of the task: Total_Compute = 5800 GPU-Hours.

[0148] To construct a Merkle tree, the intelligent computing center cloud platform uses 6000 summary data as leaf nodes to construct a Merkle tree to obtain the first Merkle root data MR = 0xabc...def.

[0149] On-chain evidence storage. Generate the first evidence storage data Evidence = {T-1001, 5800, t_start, t_end, 0xabc...def}, successfully submit and record it on the blockchain, realizing evidence storage via blockchain.

[0150] User Audit: After the computing power task was completed, the user requested all 6000 digest data from the blockchain. The user recalculated the second Merkle root MR' locally, queried the blockchain, and compared it to find that MR' == MR. Based on this, the user can be confident that the metering result of 5800 GPU-Hours is true and reliable.

[0151] Please see Figure 4 , Figure 4 This is a flowchart of a reliable measurement method for large-scale distributed computing power of an intelligent computing center cloud platform applied to edge devices, provided by the present invention. Figure 4 As shown, it includes the following steps:

[0152] Step S1': Collect parameter values ​​of multiple resource parameters in each of the N cycles, where the N cycles are the cycles in the process of running the first computing power task, and N is a positive integer greater than 1;

[0153] Step S2': Generate summary data for each period based on the parameter values ​​of multiple resource parameters for the N periods;

[0154] Step S3': Store the summary data of the N periods;

[0155] Step S4': Send the summary data of the N periods to the intelligent computing center cloud platform.

[0156] In one embodiment, step S2' includes:

[0157] Step S21': Generate the hash value of each period, wherein the N periods include the first period, and the hash value of the first period is calculated based on the parameter values ​​of multiple resource parameters of the first period;

[0158] Step S22': Generate summary data for each period. The summary data for the first period includes parameter values ​​of multiple resource parameters for the first period and the hash value of the first period.

[0159] In one embodiment, step S2 further includes:

[0160] Step S23': Construct a standardized score vector for each period. The standardized score vector for the first period is constructed based on the parameter values ​​of multiple resource parameters in the first period.

[0161] The summary data for the first period includes the standardized score vector for the first period and the hash value for the first period.

[0162] Please see Figure 5 , Figure 5 This is a flowchart of a reliable measurement method for large-scale distributed computing power applied to a cloud platform of an intelligent computing center for a first device, provided by the present invention. Figure 5 As shown, it includes the following steps:

[0163] Step S1'': Obtain the summary data of N second cycles and the first Merkle root data of the first computing power running task, where N is a positive integer greater than 1;

[0164] Step S2'': Calculate N second hash values ​​based on the N second period summary data;

[0165] Step S3'': Calculate the second Merkle root data based on the N second hash values;

[0166] Step S4'': Generate a verification result based on the first Merkle root data and the second Merkle root data. The verification result is used to characterize whether the computing power value of the first computing power running task is reliable.

[0167] In one implementation, step S1'' includes:

[0168] Step S11'': Receive the N second-period summary data sent by the intelligent computing center, or obtain the N second-period summary data from the blockchain;

[0169] Step S12'': Obtain the first Merkle root data from the blockchain.

[0170] Please see Figure 6 , Figure 6 This is a structural diagram of a trusted metering device for large-scale distributed computing power in an intelligent computing center cloud platform provided by the present invention, as shown below. Figure 6 As shown, the trusted metering device 600 for the large-scale distributed computing power of the intelligent computing center cloud platform includes:

[0171] The receiving module 601 is used to receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power running task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1.

[0172] The determining module 602 is used to determine the weight coefficient of each of the plurality of resource parameters corresponding to the type of the first computing power running task;

[0173] The metering module 603 is used to calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first cycles.

[0174] In one embodiment, the trusted metering device 600 for the large-scale distributed computing power of the intelligent computing center cloud platform further includes:

[0175] The generation module is used to generate the first Merkle root data of the first computing power running task based on the first hash value corresponding to the summary data of each first period;

[0176] The evidence storage module is used to store the first Merkle root data based on blockchain.

[0177] In one embodiment, the evidence storage module includes:

[0178] The generation unit is used to generate first evidence data based on the identifier of the first computing power running task, the computing power value of the first computing power running task, the start time and end time of the first computing power running task, and the first Merkle root data.

[0179] The evidence storage unit is used to store the first evidence storage data based on the blockchain.

[0180] In one embodiment, the first Merkle root data is used to verify with the second Merkle root data, which is data calculated by the first device based on N second hash values. The N second hash values ​​are calculated by the first device based on the parameter values ​​of multiple resource parameters in N second periods. The parameter values ​​of the multiple resource parameters in N second periods are obtained by the first device from the blockchain based on the identifier of the task run by the first computing power.

[0181] In one embodiment, the determining module 602 includes:

[0182] The determining unit is used to determine the weight coefficient of each resource parameter among the plurality of resource parameters corresponding to the type of the first computing power running task based on a preset mapping relationship. The preset mapping relationship includes the mapping relationship of the weight coefficients of each resource parameter among the plurality of resource parameters corresponding to each type.

[0183] In one embodiment, the summary data of the i-th first cycle in the N first cycles includes the identifier of the first computing power running task, the identifier of the edge device, the start and end times of the i-th first cycle, the hash value of the i-th first cycle, and the standardized score vector, wherein the standardized score vector of the i-th first cycle is constructed based on the parameter values ​​of multiple resource parameters of the i-th first cycle, i is a positive integer greater than or equal to 1, and i is less than or equal to N.

[0184] The trusted metering device for large-scale distributed computing power of the intelligent computing center cloud platform provided by this invention is capable of achieving the above. Figure 1 The various processes and technical features of the embodiments of the trusted measurement method for the large-scale distributed computing power of the intelligent computing center cloud platform shown are all corresponding one-to-one and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0185] It should be noted that the trusted metering device for the large-scale distributed computing power of the intelligent computing center cloud platform in this invention can be a device, or it can be a component, integrated circuit, or chip in an electronic device.

[0186] Please see Figure 7 , Figure 7 This is a structural diagram of a trusted metering device for large-scale distributed computing power in an intelligent computing center cloud platform provided by the present invention, as shown below. Figure 7 As shown, the trusted metering device 700 for the large-scale distributed computing power of the intelligent computing center cloud platform includes:

[0187] The acquisition module 701 is used to acquire the parameter values ​​of multiple resource parameters in each of the N cycles, where the N cycles are the cycles in the process of running the first computing power task, and N is a positive integer greater than 1.

[0188] The generation module 702 is used to generate summary data for each period based on the parameter values ​​of multiple resource parameters in the N periods;

[0189] The evidence storage module 703 is used to store the summary data of the N periods;

[0190] The sending module 704 is used to send the summary data of the N periods to the intelligent computing center cloud platform.

[0191] In one embodiment, the generation module 702 includes:

[0192] The first generation unit is used to generate the hash value of each period, wherein the N periods include the first period, and the hash value of the first period is calculated based on the parameter values ​​of multiple resource parameters of the first period.

[0193] The second generation unit is used to generate summary data for each period, wherein the summary data for the first period includes parameter values ​​of multiple resource parameters for the first period and hash value for the first period.

[0194] In one embodiment, the generation module 702 further includes:

[0195] A construction unit is used to construct a standardized score vector for each period, wherein the standardized score vector for the first period is constructed based on the parameter values ​​of multiple resource parameters in the first period;

[0196] The summary data for the first period includes the normalized score vector for the first period and the hash value for the first period.

[0197] The trusted metering device for large-scale distributed computing power of the intelligent computing center cloud platform provided by this invention is capable of achieving the above. Figure 4 The various processes and technical features of the embodiments of the trusted measurement method for the large-scale distributed computing power of the intelligent computing center cloud platform shown are all corresponding one-to-one and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0198] It should be noted that the trusted metering device for the large-scale distributed computing power of the intelligent computing center cloud platform in this invention can be a device, or it can be a component, integrated circuit, or chip in an electronic device.

[0199] Please see Figure 8 , Figure 8 This is a structural diagram of a trusted metering device for large-scale distributed computing power in an intelligent computing center cloud platform provided by the present invention, as shown below. Figure 8 As shown, the trusted metering device 800 for the large-scale distributed computing power of the intelligent computing center cloud platform includes:

[0200] The acquisition module 801 is used to acquire the summary data of N second cycles and the first Merkle root data of the first computing power running task, where N is a positive integer greater than 1;

[0201] The first calculation module 802 is used to calculate N second hash values ​​based on the summary data of the N second periods;

[0202] The second calculation module 803 is used to calculate the second Merkle root data based on the N second hash values;

[0203] The generation module 804 is used to generate a verification result based on the first Merkle root data and the second Merkle root data. The verification result is used to characterize whether the computing power value of the first computing power running task is reliable.

[0204] In one embodiment, the acquisition module 801 includes:

[0205] The first acquisition unit is used to receive the summary data of the N second periods sent by the intelligent computing center, or to acquire the summary data of the N second periods from the blockchain;

[0206] The second acquisition unit is used to acquire the first Merklegen data from the blockchain.

[0207] The trusted metering device for large-scale distributed computing power of the intelligent computing center cloud platform provided by this invention is capable of achieving the above. Figure 5 The various processes and technical features of the embodiments of the trusted measurement method for the large-scale distributed computing power of the intelligent computing center cloud platform shown are all corresponding one-to-one and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0208] It should be noted that the trusted metering device for the large-scale distributed computing power of the intelligent computing center cloud platform in this invention can be a device, or it can be a component, integrated circuit, or chip in an electronic device.

[0209] The present invention also provides an electronic device, see below. Figure 9 , Figure 9This 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 901, a processor 902, and a program or instructions stored in the memory 901 that run on the processor. When the program or instructions are executed by the processor 902, they can achieve the following: Figure 1 The steps in the corresponding embodiment of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform, and the achievement of the same beneficial effects, will not be elaborated here.

[0210] The processor 902 can be a CPU, ASIC, FPGA, or GPU.

[0211] Those skilled in the art will understand that all or part of the steps of the above-described embodiment of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform can be implemented by hardware related to program instructions, and the program can be stored in a readable medium.

[0212] 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 Any step in the corresponding embodiment of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform can achieve the same technical effect, and will not be described again here to avoid repetition. The storage medium mentioned includes, for example, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0213] 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 various processes of the corresponding implementation of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform can achieve the same technical effect, and will not be described again here to avoid repetition.

[0214] 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.

[0215] 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.

[0216] 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.

[0217] 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 reliable measurement method for large-scale distributed computing power of an intelligent computing center cloud platform, characterized in that, include: Step S1: Receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1. Step S2: Determine the weight coefficient of each of the multiple resource parameters corresponding to the type of the first computing power running task; Step S3: Calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first cycles; Step S4: Generate the first Merkle root data for the first computing power running task based on the first hash value corresponding to the digest data of each first period; Step S5: Store the first Merklegen data using blockchain; The first Merkle root data is used to verify with the second Merkle root data, which is data calculated by the first device based on N second hash values. The N second hash values ​​are calculated by the first device based on the parameter values ​​of multiple resource parameters in N second periods. The parameter values ​​of the multiple resource parameters in N second periods are obtained by the first device from the blockchain based on the identifier of the task run by the first computing power. The first device is further configured to generate a verification result based on the first Merkle root data and the second Merkle root data, and the verification result is used to determine whether the computing power value of the first computing power running task is reliable.

2. The method as described in claim 1, characterized in that, Step S5 includes: Step S51: Generate first evidence data based on the identifier of the first computing power running task, the computing power value of the first computing power running task, the start time and end time of the first computing power running task, and the first Merkle root data; Step S52: Store the first evidence data based on the blockchain.

3. The method as described in claim 1 or 2, characterized in that, Step S2 includes: Step S21: Determine the weight coefficient of each resource parameter among the multiple resource parameters corresponding to the type of the first computing power running task based on the preset mapping relationship. The preset mapping relationship includes the mapping relationship of the weight coefficient of each resource parameter among the multiple resource parameters corresponding to each type.

4. The method as described in claim 1 or 2, characterized in that, The summary data of the i-th first cycle in the N first cycles includes the identifier of the first computing power running task, the identifier of the edge device, the start and end times of the i-th first cycle, the hash value of the i-th first cycle, and the standardized score vector. The standardized score vector of the i-th first cycle is constructed based on the parameter values ​​of multiple resource parameters of the i-th first cycle, where i is a positive integer greater than or equal to 1 and less than or equal to N.

5. A trusted metering device for large-scale distributed computing power of an intelligent computing center cloud platform, characterized in that, include: The receiving module is used to receive N first-cycle summary data sent by the edge device. The N first-cycle summary data includes the parameter values ​​of multiple resource parameters collected by the edge device in the N first cycles during the operation of the first computing power task. The N first-cycle summary data is data stored through blockchain, and N is a positive integer greater than 1. The determining module is used to determine the weight coefficient of each of the plurality of resource parameters corresponding to the type of the first computing power running task; The metering module is used to calculate the computing power value of the first computing power running task based on the weight coefficient of each resource parameter and the parameter values ​​of multiple resource parameters in the N first periods. The generation module is used to generate the first Merkle root data of the first computing power running task based on the first hash value corresponding to the summary data of each first period; The evidence storage module is used to store the first Merklegen data based on blockchain. The first Merkle root data is used to verify with the second Merkle root data, which is data calculated by the first device based on N second hash values. The N second hash values ​​are calculated by the first device based on the parameter values ​​of multiple resource parameters in N second periods. The parameter values ​​of the multiple resource parameters in N second periods are obtained by the first device from the blockchain based on the identifier of the task run by the first computing power. The first device is further configured to generate a verification result based on the first Merkle root data and the second Merkle root data, and the verification result is used to determine whether the computing power value of the first computing power running task is reliable.

6. An electronic device, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the trusted measurement method for large-scale distributed computing power of the intelligent computing center cloud platform as described in any one of claims 1 to 4.

8. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the steps of a trusted measurement method for large-scale distributed computing power of an intelligent computing center cloud platform as described in any one of claims 1 to 4.

Images