Method and device for measuring temperature sensitivity and energy consumption correlation degree of digital intelligent base

Abstract

The invention relates to a digital intelligent base temperature sensitivity and energy consumption correlation degree measuring method and device, and the method comprises the steps: determining a target digital intelligent base typical task corresponding to each preset digital intelligent base in at least one preset digital intelligent base; executing the target digital intelligent base typical task to obtain a temperature parameter and an energy consumption parameter corresponding to the target digital intelligent base typical task, and obtaining a workload parameter corresponding to the target digital intelligent base typical task; and on the basis of the workload parameters, multivariate linear fitting operation is carried out on the temperature parameters and the energy consumption parameters, so that the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation degree corresponding to the energy consumption parameters can be obtained, and therefore, the green, energy-saving and environmental protection capability measurement of environments such as a digital intelligent base and a computing power center can be quantified more comprehensively and uniformly. Therefore, the problem of secondary energy consumption caused by the refrigeration requirement due to the fact that the prior art and the measurement method are lack of considering the operating environment temperature sensitivity while calculating power energy consumption measurement evaluation of digital intelligent base application and the like is solved.

Description

Technical Field

[0001] The present application relates to the field of intelligent devices and energy consumption management technology, and in particular to a method and device for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base. Background Art

[0002] Currently, the energy demand of data centers continues to grow. The rapid innovation of technologies represented by artificial intelligence has promoted the construction of large-scale computing power infrastructure. The electricity required for these infrastructures far exceeds that required by traditional data centers. However, existing technologies and measurement methods lack the ability to measure and evaluate computing power energy consumption such as digital base applications while taking into account the secondary energy consumption caused by cooling requirements due to the temperature sensitivity of the operating environment, which needs to be urgently addressed. Summary of the invention

[0003] The present application provides a method and device for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base, so as to solve the problem that the existing technology and measurement methods lack the ability to measure and evaluate the computing power energy consumption of digital intelligence base applications while taking into account the secondary energy consumption caused by the cooling requirements due to the temperature sensitivity of the operating environment.

[0004] The first aspect of the present application provides a method for measuring the correlation between the temperature sensitivity and energy consumption of a digital intelligence base, including the following steps: determining a target digital intelligence base typical task corresponding to each preset digital intelligence base in at least one preset digital intelligence base; executing the target digital intelligence base typical task to obtain the temperature parameters and energy consumption parameters corresponding to the target digital intelligence base typical task, and obtaining the workload parameters corresponding to the target digital intelligence base typical task; based on the workload parameters, performing a multivariate linear fitting operation on the temperature parameters and the energy consumption parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters.

[0005] Optionally, in one embodiment of the present application, determining the target digital intelligence base typical tasks corresponding to each preset digital intelligence base in at least one preset digital intelligence base includes: identifying the digital intelligence base type corresponding to each preset digital intelligence base; selecting the target digital intelligence base typical tasks corresponding to each preset digital intelligence base according to the digital intelligence base type, wherein the at least one preset digital intelligence base includes at least one of a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

[0006] Optionally, in one embodiment of the present application, executing the typical task of the target digital intelligence base to obtain the temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligence base, and obtaining the workload parameters corresponding to the typical task of the target digital intelligence base, includes: based on the typical task of the target digital intelligence base, obtaining a test data set of a preset digital intelligence base corresponding to the typical task of the target digital intelligence base, wherein the typical task of the target digital intelligence base includes any one of a preset large model typical task, a preset big data processing typical task, and a preset large model-big data hybrid task; executing the typical task of the target digital intelligence base according to the test data set to obtain the temperature parameters and power information corresponding to the typical task of the target digital intelligence base, and using the power information to calculate the energy consumption parameters corresponding to the typical task of the target digital intelligence base; obtaining the workload parameters generated in the process of executing the typical task of the target digital intelligence base corresponding to each preset digital intelligence base, wherein the workload parameters include at least one of the total number of tokens, the number of transactions processed, and the number of task batches.

[0007] Optionally, in one embodiment of the present application, after calculating the energy consumption correlation metric corresponding to each preset digital and intelligent base, it also includes: establishing a target binary parameter measurement indicator corresponding to each preset digital and intelligent base based on the temperature sensitivity and the energy consumption correlation metric; and optimizing the operating performance and operating environment of each preset digital and intelligent base through the target binary parameter measurement indicator.

[0008] Optionally, in one embodiment of the present application, the multivariate linear fitting calculation expression corresponding to the multivariate linear fitting operation is:

[0009] W=α*T+β*E+γ

[0010] Wherein, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

[0011] The second aspect of the present application provides a device for measuring the correlation between the temperature sensitivity and energy consumption of a digital and intelligent base, including: a determination module, used to determine a target digital and intelligent base typical task corresponding to each preset digital and intelligent base in at least one preset digital and intelligent base; an execution module, used to execute the target digital and intelligent base typical task to obtain the temperature parameters and energy consumption parameters corresponding to the target digital and intelligent base typical task, and obtain the workload parameters corresponding to the target digital and intelligent base typical task; a multivariate linear fitting module, used to perform a multivariate linear fitting operation on the temperature parameters and the energy consumption parameters based on the workload parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters.

[0012] Optionally, in one embodiment of the present application, the determination module includes: a type analysis unit, used to identify the type of digital intelligence base corresponding to each preset digital intelligence base; a selection unit, used to select the target digital intelligence base typical task corresponding to each preset digital intelligence base according to the digital intelligence base type, wherein at least one preset digital intelligence base includes at least one of a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

[0013] Optionally, in one embodiment of the present application, the execution module includes: a first acquisition unit, used to acquire a test data set of a preset digital intelligence base corresponding to the typical task of the target digital intelligence base based on the typical task of the target digital intelligence base, wherein the typical task of the target digital intelligence base includes any one of a preset large model typical task, a preset big data processing typical task, and a preset large model-big data hybrid task; an operation unit, used to execute the typical task of the target digital intelligence base according to the test data set to obtain the temperature parameters and power information corresponding to the typical task of the target digital intelligence base, and use the power information to calculate the energy consumption parameters corresponding to the typical task of the target digital intelligence base; a second acquisition unit, used to acquire the workload parameters generated in the process of executing the typical task of the target digital intelligence base corresponding to each preset digital intelligence base, wherein the workload parameters include at least one of the total number of tokens, the number of transaction processing, and the number of task batches.

[0014] Optionally, in one embodiment of the present application, it also includes: an establishment module, used to establish a target binary parameter measurement indicator corresponding to each preset digital and intelligent base based on the temperature sensitivity and the energy consumption correlation measurement after calculating the energy consumption correlation measurement corresponding to each preset digital and intelligent base; a measurement module, used to optimize the operating performance and operating environment of each preset digital and intelligent base through the target binary parameter measurement indicator.

[0015] Optionally, in one embodiment of the present application, the multivariate linear fitting calculation expression corresponding to the multivariate linear fitting operation is:

[0016] W=α*T+β*E+γ

[0017] Wherein, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

[0018] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for measuring the correlation between the temperature sensitivity and energy consumption of a digital intelligence base as described in the above embodiment.

[0019] The fourth aspect embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above-mentioned method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base.

[0020] The fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the above-mentioned method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base.

[0021] Therefore, the embodiments of the present application have the following beneficial effects:

[0022] The embodiments of the present application can be performed by determining the typical tasks of the target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base; executing the typical tasks of the target digital intelligence base to obtain the temperature parameters and energy consumption parameters corresponding to the typical tasks of the target digital intelligence base, and obtaining the workload parameters corresponding to the typical tasks of the target digital intelligence base; based on the workload parameters, performing multivariate linear fitting operations on the temperature parameters and the energy consumption parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters. The present application analyzes the energy consumption levels of different digital intelligence bases and designs a standard energy consumption test method to propose a scientific strategy to quantify the correlation between temperature sensitivity and energy consumption, thereby more comprehensively and balancedly quantifying the green, energy-saving, and environmentally friendly capabilities of digital intelligence bases, computing power centers, and other environments, providing technical indicator support for the design of intelligent infrastructure and improving the overall performance of the system. As a result, the existing technologies and measurement methods lack the ability to evaluate the computing power energy consumption metrics such as digital intelligence base applications while taking into account the secondary energy consumption caused by the refrigeration requirements caused by the temperature sensitivity of the operating environment.

[0023] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0025] Figure 1 A flowchart of a method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base provided according to an embodiment of the present application;

[0026] Figure 2 A schematic diagram of the logical architecture of a method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base provided in one embodiment of the present application;

[0027] Figure 3 A schematic diagram of a temperature sensitivity and energy consumption correlation metric calculation logic provided for one embodiment of the present application;

[0028] Figure 4 This is an example diagram of a device for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base according to an embodiment of the present application;

[0029] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0030] Among them, 10-a measurement device for the correlation between the temperature sensitivity and energy consumption of the digital intelligence base; 100-a determination module, 200-an execution module, 300-a multivariate linear fitting module; 501-a memory, 502-a processor, 503-a communication interface. DETAILED DESCRIPTION

[0031] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0032] The following describes the measurement method and device of the correlation between the temperature sensitivity and energy consumption of the digital intelligence base of the embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the above background technology, the present application provides a measurement method for the correlation between the temperature sensitivity and energy consumption of the digital intelligence base. In this method, by determining the typical task of the target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base; executing the typical task of the target digital intelligence base to obtain the temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligence base, and obtaining the workload parameters corresponding to the typical task of the target digital intelligence base; based on the workload parameters, performing multivariate linear fitting operations on the temperature parameters and the energy consumption parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters. The present application analyzes the energy consumption levels of different digital intelligence bases and designs a standard energy consumption test method to propose a scientific strategy for quantifying the correlation between temperature sensitivity and energy consumption, thereby more comprehensively and balancedly quantifying the green, energy-saving, and environmentally friendly capabilities of digital intelligence bases, computing power centers, and other environments, providing technical indicator support for the design of intelligent infrastructure and improving the overall performance of the system. This solves the problem that existing technologies and measurement methods lack the ability to measure and evaluate computing power energy consumption in digital base applications, while also taking into account the secondary energy consumption caused by refrigeration requirements due to the temperature sensitivity of the operating environment.

[0033] Specifically, Figure 1 A flowchart of a method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base provided in an embodiment of the present application.

[0034] like Figure 1 As shown, the method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base includes the following steps:

[0035] In step S101, a typical task of a target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base is determined.

[0036] The embodiments of the present application can first select a suitable typical task of a digital intelligence base for each preset digital intelligence base, thereby providing technical and data support for the implementation of energy consumption assessment of the digital intelligence base.

[0037] Optionally, in one embodiment of the present application, determining a target digital intelligence base typical task corresponding to each preset digital intelligence base in at least one preset digital intelligence base includes: identifying the digital intelligence base type corresponding to each preset digital intelligence base; selecting the target digital intelligence base typical task corresponding to each preset digital intelligence base according to the digital intelligence base type, wherein at least one preset digital intelligence base includes at least one of a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

[0038] In the actual implementation process, Figure 2 As shown, the digital intelligence base facilities in the embodiments of the present application mainly include IT infrastructure such as AI big models, databases, big data management related software and computing platforms.

[0039] It should be noted that the above-mentioned AI large model refers to a deep learning model with ultra-large-scale parameters and computing power, which can be widely used in fields such as natural language processing, can handle complex tasks and has the ability to understand and generate data.

[0040] The embodiments of the present application can determine the digital intelligence base type corresponding to each digital intelligence base, so as to select and execute appropriate typical tasks according to different digital intelligence base types, and provide corresponding parameter data for energy consumption analysis.

[0041] In step S102, a typical task of the target digital intelligent base is executed to obtain temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligent base, and to obtain workload parameters corresponding to the typical task of the target digital intelligent base.

[0042] Furthermore, the embodiments of the present application can execute typical tasks of the target digital and intelligent base, and monitor relevant parameters such as temperature and energy consumption generated by the digital and intelligent base when executing typical tasks; in addition, the embodiments of the present application can also obtain workload parameters corresponding to typical tasks of the target digital and intelligent base, thereby providing reliable data guidance and basis for subsequent optimization analysis of the digital and intelligent base.

[0043] Optionally, in one embodiment of the present application, typical tasks of the target digital intelligence base are executed to obtain temperature parameters and energy consumption parameters corresponding to the typical tasks of the target digital intelligence base, and workload parameters corresponding to the typical tasks of the target digital intelligence base are obtained, including: based on the typical tasks of the target digital intelligence base, a test data set of a preset digital intelligence base corresponding to the typical tasks of the target digital intelligence base is obtained, wherein the typical tasks of the target digital intelligence base include any one of preset large model typical tasks, preset big data processing typical tasks, and preset large model-big data hybrid tasks; the typical tasks of the target digital intelligence base are executed according to the test data set to obtain temperature parameters and power information corresponding to the typical tasks of the target digital intelligence base, and the energy consumption parameters corresponding to the typical tasks of the target digital intelligence base are calculated using the power information; the workload parameters generated in the process of executing the typical tasks of the target digital intelligence base corresponding to each preset digital intelligence base are obtained, wherein the workload parameters include at least one of the total number of tokens, the number of transactions processed, and the number of task batches.

[0044] It should be noted that the typical tasks of the digital and intelligent base in the embodiments of the present application include typical tasks of large models, typical tasks of big data processing, and mixed tasks of the two. They provide the data set required for testing (i.e., test data set) for the digital and intelligent base facilities, and use them as input to enable the digital and intelligent base facilities to generate temperature parameters and power information that can perform energy consumption analysis, and calculate the corresponding energy consumption parameters through the power information.

[0045] Typical tasks for large models include data sets related to the functions of large models, such as the Alpaca data set and the GSM8K data set. The Alpaca data set contains text data in multiple languages, which can be used to perform a series of natural language processing tasks, such as sentiment analysis and text classification; the GSM8K data set contains thousands of math problems to test the model's logical reasoning and problem-solving capabilities; the above data sets can be used to evaluate the performance of the model in various tasks and how different types of data affect energy consumption and reasoning performance.

[0046] Typical big data processing tasks include simulating complex query operations related to business in databases and other big data related software to test the performance of related databases and big data related software, such as benchmark tasks such as TPC-H, TPC-IoT, and TPC-AI. Among them, TPC-H simulates complex query and large-scale data processing scenarios, consisting of a set of business-oriented ad hoc queries and concurrent data modifications. By predefining multiple complex SQL operations, it uses indicators such as QphH@Size (number of queries per hour), $ / kQphH@Size (cost per thousand queries), and system availability date to test the performance of the database; TPC-IoT is a benchmark designed specifically for measuring the performance of IoT gateway systems, mainly using indicators such as IoTps (number of IoT transactions processed per second), $ / kIoTps (cost per thousand IoT transactions), and system availability date for measurement; TPCx-AI is a benchmark for evaluating the performance of artificial intelligence and machine learning applications, mainly using indicators such as AIUCpm (number of AI user queries processed per minute), $ / AIUCpm (cost of processing each AI user query), and system availability date for evaluation.

[0047] Afterwards, the embodiments of the present application also need to obtain the workload parameters generated during the execution of the target digital intelligence base typical tasks corresponding to each preset digital intelligence base. The workload parameters are indicators for measuring the overall workload of a certain digital intelligence base, which can be the number of transactions processed or the amount of calculations performed. In the actual execution process, for the AI ​​big model, the workload in the embodiments of the present application can be the total number of tokens in the process of executing the target digital intelligence base typical tasks corresponding to each preset digital intelligence base; for databases or big data software, the workload can be the number of transactions processed or the number of task batches in a certain test.

[0048] In step S103, based on the workload parameter, a multivariate linear fitting operation is performed on the temperature parameter and the energy consumption parameter to obtain a temperature sensitivity corresponding to the temperature parameter and an energy consumption correlation metric corresponding to the energy consumption parameter.

[0049] Furthermore, the embodiments of the present application can perform multivariate linear regression based on the temperature parameters and the energy consumption parameters to obtain linear fitting parameters, which are temperature sensitivity and energy consumption correlation metrics, respectively.

[0050] Optionally, in one embodiment of the present application, the multivariate linear fitting calculation expression corresponding to the multivariate linear fitting operation is:

[0051] W=α*T+β*E+γ

[0052] Wherein, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

[0053] In the specific implementation process, Figure 3 As shown, the embodiment of the present application can obtain workload data W when different digital intelligence base devices select appropriate typical tasks for testing, and obtain the above-mentioned temperature parameters T and power information P through corresponding sensor devices, and calculate the corresponding energy consumption parameter E through the power information P; secondly, the embodiment of the present application can perform multivariate linear fitting operations on temperature parameters and energy consumption parameters based on workload parameters, as shown in the following formula:

[0054] W=α*T+β*E+γ

[0055] Wherein, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

[0056] It should be noted that in the specific implementation process, for some applications, if it is necessary to focus on one or more specific tasks in the task group, bilinear fitting can be performed on the workload, temperature parameters and power information of one or more specific tasks respectively to form a binary pair of temperature sensitivity and energy consumption correlation, thereby supporting finer-grained measurement.

[0057] Therefore, the embodiment of the present application calculates the bilinear fitting parameters of workload, temperature and energy consumption, and uses the linear fitting parameters between the workload parameters and the temperature parameters in the bilinear fitting parameters as temperature sensitivity analysis indicators, which reflect the linear influence of temperature on workload when the energy consumption remains unchanged; at the same time, the linear fitting parameters between the workload parameters and the energy consumption parameters in the bilinear fitting parameters are used as energy consumption correlation measurement analysis indicators, which reflect the linear influence of energy consumption on workload when the temperature remains unchanged.

[0058] Optionally, in one embodiment of the present application, after calculating the energy consumption correlation measurement corresponding to each preset digital and intelligent base, it also includes: establishing a target binary parameter measurement indicator corresponding to each preset digital and intelligent base based on temperature sensitivity and energy consumption correlation measurement; optimizing the operating performance and operating environment of each preset digital and intelligent base through the target binary parameter measurement indicator.

[0059] As an achievable method, the embodiments of the present application can also establish a combined indicator target (α, β) (i.e., a binary parameter measurement indicator) corresponding to each preset digital intelligence base based on temperature sensitivity and energy consumption correlation metrics, so as to measure and optimize the measured operating performance (such as equipment energy consumption performance) and operating environment (such as ambient temperature and cooling factors) of each preset digital intelligence base through the target binary parameter measurement indicator.

[0060] Therefore, the embodiments of the present application analyze the temperature and energy consumption data generated when the digital intelligence base performs typical tasks to obtain the relationship between workload, work efficiency, temperature and energy consumption. For equipment that is greatly affected by temperature, cooling measures can be considered; equipment that is highly correlated with energy consumption can improve efficiency by optimizing energy consumption allocation, and thus, by analyzing the binary indicators of temperature sensitivity and energy consumption correlation measurement, different equipment can be selected to execute different tasks, and directional reference can also be provided for optimizing equipment performance.

[0061] In addition, the present application can also construct a corresponding digital intelligence base temperature sensitivity and energy consumption correlation measurement system according to the measurement method of the digital intelligence base temperature sensitivity and energy consumption correlation. The following is an explanation and introduction of the digital intelligence base temperature sensitivity and energy consumption correlation measurement system of the present application.

[0062] The measurement system of the correlation between temperature sensitivity and energy consumption of the digital intelligent base in the present application mainly includes two parts: an input module and an analysis module.

[0063] Among them, the input module includes digital and intelligent base equipment and typical tasks of the digital and intelligent base. It is responsible for monitoring the temperature, energy consumption and other related parameters generated by the digital and intelligent base when performing typical tasks, and transmitting these data to the analysis module for processing and analysis.

[0064] The analysis module performs linear fitting on the workload according to the temperature and energy consumption, and obtains the linear fitting parameters through multiple linear regression, which are the correlation metrics of temperature sensitivity and energy consumption, and use them as indicators to measure equipment performance and environmental factors.

[0065] According to the measurement method of the correlation between the temperature sensitivity and energy consumption of the digital intelligence base proposed in the embodiment of the present application, by determining the typical task of the target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base; executing the typical task of the target digital intelligence base to obtain the temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligence base, and obtaining the workload parameters corresponding to the typical task of the target digital intelligence base; based on the workload parameters, performing multivariate linear fitting operations on the temperature parameters and the energy consumption parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters. This application analyzes the energy consumption levels of different digital intelligence bases and designs a standard energy consumption test method to propose a scientific strategy to quantify the correlation between temperature sensitivity and energy consumption, thereby more comprehensively and balancedly quantifying the green, energy-saving, and environmentally friendly capabilities of digital intelligence bases, computing power centers, and other environments, providing technical indicator support for the design of intelligent infrastructure and improving the overall performance of the system.

[0066] Secondly, a device for measuring the correlation between the temperature sensitivity and energy consumption of a digital intelligence base proposed in an embodiment of the present application is described with reference to the accompanying drawings.

[0067] Figure 4 It is a block diagram of a device for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base according to an embodiment of the present application.

[0068] like Figure 4 As shown, the device 10 for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base includes: a determination module 100 , an execution module 200 and a multivariate linear fitting module 300 .

[0069] Among them, the determination module 100 is used to determine the typical tasks of the target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base.

[0070] The execution module 200 is used to execute typical tasks of the target digital and intelligent base to obtain temperature parameters and energy consumption parameters corresponding to the typical tasks of the target digital and intelligent base, and obtain workload parameters corresponding to the typical tasks of the target digital and intelligent base.

[0071] The multivariate linear fitting module 300 is used to perform a multivariate linear fitting operation on the temperature parameter and the energy consumption parameter based on the workload parameter to obtain the temperature sensitivity corresponding to the temperature parameter and the energy consumption correlation measurement corresponding to the energy consumption parameter.

[0072] Optionally, in one embodiment of the present application, the determination module 100 includes: a type analysis unit and a selection unit.

[0073] Among them, the type analysis unit is used to identify the digital intelligence base type corresponding to each preset digital intelligence base.

[0074] A selection unit is used to select a target digital intelligence base typical task corresponding to each preset digital intelligence base according to the type of the digital intelligence base, wherein at least one preset digital intelligence base includes at least one of a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

[0075] Optionally, in one embodiment of the present application, the execution module 200 includes: a first acquisition unit, a running unit and a second acquisition unit.

[0076] Among them, the first acquisition unit is used to obtain the test data set of the preset digital intelligence base corresponding to the typical tasks of the target digital intelligence base based on the typical tasks of the target digital intelligence base, wherein the typical tasks of the target digital intelligence base include any one of the preset large model typical tasks, the preset big data processing typical tasks and the preset large model-big data hybrid tasks.

[0077] An operating unit is used to execute typical tasks of the target digital intelligence base according to a test data set to obtain temperature parameters and power information corresponding to the typical tasks of the target digital intelligence base, and use the power information to calculate energy consumption parameters corresponding to the typical tasks of the target digital intelligence base.

[0078] The second acquisition unit is used to obtain the workload parameters generated during the execution of typical tasks of the target digital intelligence base corresponding to each preset digital intelligence base, wherein the workload parameters include at least one of the total number of tokens, the number of transactions processed, and the number of task batches.

[0079] Optionally, in one embodiment of the present application, the device 10 for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base in the embodiment of the present application further includes: an establishment module and a measurement module.

[0080] Among them, a module is established to establish a target binary parameter measurement indicator corresponding to each preset digital intelligence base based on temperature sensitivity and energy consumption correlation measurement after calculating the energy consumption correlation measurement corresponding to each preset digital intelligence base.

[0081] The measurement module is used to optimize the operating performance and operating environment of each preset digital intelligence base through target binary parameter measurement indicators.

[0082] Optionally, in one embodiment of the present application, the multivariate linear fitting calculation expression corresponding to the multivariate linear fitting operation is:

[0083] W=α*T+β*E+γ

[0084] Among them, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

[0085] It should be noted that the above explanation of the embodiment of the method for measuring the correlation between the temperature sensitivity of the digital and intelligent base and the energy consumption is also applicable to the device for measuring the correlation between the temperature sensitivity of the digital and intelligent base and the energy consumption of this embodiment, and will not be repeated here.

[0086] According to the measurement device of the correlation between the temperature sensitivity and energy consumption of the digital intelligence base proposed in the embodiment of the present application, it includes a determination module for determining the typical task of the target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base; an execution module for executing the typical task of the target digital intelligence base to obtain the temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligence base, and obtain the workload parameters corresponding to the typical task of the target digital intelligence base; a multivariate linear fitting module for performing multivariate linear fitting operations on the temperature parameters and the energy consumption parameters based on the workload parameters to obtain the temperature sensitivity corresponding to the temperature parameters and the energy consumption correlation measurement corresponding to the energy consumption parameters. This application analyzes the energy consumption levels of different digital intelligence bases and designs a standard energy consumption test method to propose a scientific strategy for quantifying the correlation between temperature sensitivity and energy consumption, thereby more comprehensively and balancedly quantifying the green, energy-saving, and environmentally friendly capabilities of digital intelligence bases, computing power centers, and other environments, providing technical indicator support for the design of intelligent infrastructure and improving the overall performance of the system.

[0087] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0088] A memory 501 , a processor 502 , and a computer program stored in the memory 501 and executable on the processor 502 .

[0089] When the processor 502 executes the program, the method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base provided in the above embodiment is implemented.

[0090] Furthermore, the electronic device further comprises:

[0091] The communication interface 503 is used for communication between the memory 501 and the processor 502 .

[0092] The memory 501 is used to store computer programs that can be executed on the processor 502 .

[0093] The memory 501 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0094] If the memory 501, the processor 502 and the communication interface 503 are implemented independently, the communication interface 503, the memory 501 and the processor 502 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0095] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can communicate with each other through an internal interface.

[0096] The processor 502 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0097] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base.

[0098] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for measuring the correlation between the temperature sensitivity and energy consumption of the digital intelligence base.

[0099] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0100] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0101] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0102] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0103] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0104] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0105] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0106] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base, characterized in that: The following steps are involved: Determine a target digital intelligence base typical task corresponding to each preset digital intelligence base in at least one preset digital intelligence base; Execute the target digital intelligence base typical task to obtain the temperature parameter and energy consumption parameter corresponding to the target digital intelligence base typical task, and obtain the workload parameter corresponding to the target digital intelligence base typical task; Based on the workload parameter, a multivariate linear fitting operation is performed on the temperature parameter and the energy consumption parameter to obtain a temperature sensitivity corresponding to the temperature parameter and an energy consumption correlation metric corresponding to the energy consumption parameter.

2. The method according to claim 1, characterized in that The determining of a typical task of a target digital intelligence base corresponding to each preset digital intelligence base in at least one preset digital intelligence base includes: Identify the type of digital intelligence base corresponding to each preset digital intelligence base; According to the type of the digital intelligence base, a target digital intelligence base typical task corresponding to each preset digital intelligence base is selected, wherein at least one of the preset digital intelligence base includes at least one of a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

3. The method according to claim 1, characterized in that The executing the target digital intelligence base typical task to obtain the temperature parameter and energy consumption parameter corresponding to the target digital intelligence base typical task, and obtaining the workload parameter corresponding to the target digital intelligence base typical task, includes: Based on the target digital intelligence base typical task, obtain a test data set of a preset digital intelligence base corresponding to the target digital intelligence base typical task, wherein the target digital intelligence base typical task includes any one of a preset large model typical task, a preset big data processing typical task, and a preset large model-big data hybrid task; Execute the target digital intelligence base typical task according to the test data set to obtain the temperature parameter and power information corresponding to the target digital intelligence base typical task, and calculate the energy consumption parameter corresponding to the target digital intelligence base typical task using the power information; Obtain the workload parameters generated during the execution of the typical tasks of the target digital intelligence base corresponding to each preset digital intelligence base, wherein the workload parameters include at least one of the total number of tokens, the number of transactions processed, and the number of task batches.

4. The method according to claim 1, characterized in that: After calculating the energy consumption correlation metric corresponding to each preset digital intelligence base, the method further includes: Based on the temperature sensitivity and the energy consumption correlation metric, establishing a target binary parameter measurement indicator corresponding to each preset digital intelligence base; The operating performance and operating environment of each preset digital intelligence base are optimized through the target binary parameter measurement index.

5. The method according to claim 1, characterized in that: The multivariate linear fitting calculation expression corresponding to the multivariate linear fitting operation is: W=α*T+β*E+γ Wherein, W represents the workload parameter; α represents the temperature sensitivity; T represents the temperature parameter; β represents the energy consumption correlation metric; E represents the energy consumption parameter; γ represents the multivariate linear fitting compensation constant.

6. A device for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base, characterized in that: include: A determination module, used to determine a target digital intelligence base typical task corresponding to each preset digital intelligence base in at least one preset digital intelligence base; An execution module, used to execute the typical task of the target digital intelligent base to obtain the temperature parameters and energy consumption parameters corresponding to the typical task of the target digital intelligent base, and obtain the workload parameters corresponding to the typical task of the target digital intelligent base; A multivariate linear fitting module is used to perform a multivariate linear fitting operation on the temperature parameter and the energy consumption parameter based on the workload parameter to obtain a temperature sensitivity corresponding to the temperature parameter and an energy consumption correlation metric corresponding to the energy consumption parameter.

7. The device according to claim 6, characterized in that The determination module comprises: A type analysis unit, used to identify the type of digital intelligence base corresponding to each preset digital intelligence base; A selection unit is used to select a target digital intelligence base typical task corresponding to each preset digital intelligence base according to the type of the digital intelligence base, wherein at least one of the preset digital intelligence base includes a target AI big model, a target database, a target big data management software and a computing platform IT infrastructure.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement a method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed to implement the method for measuring the correlation between temperature sensitivity and energy consumption of a digital intelligence base as described in any one of claims 1 to 5.

Images