Data center computing power resource and refrigeration equipment energy consumption correlation analysis method and device

Abstract

The invention relates to a correlation analysis method and device for computing power resources of a data center and energy consumption of refrigeration equipment, and the method comprises the steps: obtaining related data of the computing power resources of the data center and the energy consumption of the refrigeration equipment, carrying out the correlation analysis of the preprocessed related data, and obtaining key parameters related to the computing power resources of the data center and the energy consumption of the refrigeration equipment; selecting a model according to the key parameters and the modeling target, training the selected model by using historical data to learn the correlation between the computing power resource and the energy consumption of the refrigeration equipment, and evaluating the performance index of the selected model; according to the modeling target and the constraint condition, an optimization algorithm is selected to carry out joint optimization logic of the data center computing power resources and the refrigeration equipment energy consumption, the optimization algorithm is operated, and distribution of the data center computing power resources and operation parameters of the refrigeration equipment are adjusted through iteration; and obtaining a performance index for evaluating the correlation between the computing power resource of the data center and the energy consumption of the refrigeration equipment. According to the method, the prediction accuracy, generalization ability and interpretability of the model are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and specifically relates to a method and device for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption. Background Art

[0002] There is a close relationship between the computing resources of a data center and the energy consumption of cooling equipment. Generally speaking, the increase in computing resources is often accompanied by an increase in the energy consumption of cooling equipment, but the optimization and management between the two are crucial to the overall energy efficiency of the data center.

[0003] In the existing technology, traditional modeling methods have certain limitations, high costs and long cycles. Physical models are mainly established through experiments and tests, which require a lot of time and cost and cannot fully consider complex logical relationships. Although physical models have high reliability and good result traceability, they cannot fully consider the complex logical relationships of data centers. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a method and device for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption, so as to solve or partially solve the problems mentioned in the background technology.

[0005] Based on the above objectives, in a first aspect, the present invention provides a method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption, comprising:

[0006] Obtain data related to data center computing resources and refrigeration equipment energy consumption, and preprocess the related data, wherein the related data includes data center computing resources usage data, refrigeration equipment energy consumption data, and environmental factor data;

[0007] Performing correlation analysis on the preprocessed relevant data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption; selecting a model based on the key parameters and modeling objectives, using historical data to train the selected model to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluating the performance indicators of the selected model;

[0008] According to the modeling objectives and constraints, an optimization algorithm is selected to perform the joint optimization logic of data center computing resources and refrigeration equipment energy consumption. The optimization algorithm is run to iteratively adjust the allocation of data center computing resources and the operating parameters of refrigeration equipment. When the optimization algorithm converges or reaches the preset number of iterations, performance indicators for evaluating the correlation between data center computing resources and refrigeration equipment energy consumption are obtained.

[0009] As a preferred solution of the data center computing power resource and refrigeration equipment energy consumption correlation analysis method, the data center computing power resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data;

[0010] The preprocessing of the relevant data includes removing outliers and missing values, and normalizing or standardizing the relevant data.

[0011] As a preferred solution for the method of analyzing the correlation between data center computing resources and refrigeration equipment energy consumption, preprocessing the relevant data also includes time window division or lag processing of the data to determine the input layer, hidden layer, and output layer structure of the model, and select activation functions, loss functions, and optimizers.

[0012] As a preferred solution for the method of analyzing the correlation between data center computing resources and refrigeration equipment energy consumption, the model selected according to the key parameters and modeling objectives is a physical model, a mathematical model, a simulation model or a machine learning model;

[0013] The performance indicators for evaluating the selected models include prediction error indicators and stability indicators.

[0014] As a preferred solution for the data center computing power resources and refrigeration equipment energy consumption correlation analysis method, the modeling objectives are energy consumption reduction rate and computing power resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits;

[0015] The method further includes mathematizing the constraint conditions, converting the constraint conditions into mathematical expressions or inequalities, and setting initial parameters of the algorithm according to the constraint conditions;

[0016] The optimization algorithms selected are particle swarm optimization, genetic algorithm, and deep reinforcement learning algorithm.

[0017] As a preferred solution of the data center computing power resources and refrigeration equipment energy consumption correlation analysis method, the iterative adjustment of the data center computing power resources allocation and the refrigeration equipment operating parameters includes:

[0018] In each iteration, a new solution is generated according to the algorithm rules, the objective function value of the new solution is evaluated, and it is screened according to the constraints, the current optimal solution is updated, and the iteration continues until the termination condition is met.

[0019] In a second aspect, the present invention provides a device for analyzing the correlation between computing resources in a data center and energy consumption of refrigeration equipment, comprising:

[0020] A relevant data acquisition processing module is used to acquire relevant data on data center computing resources and refrigeration equipment energy consumption, and pre-process the relevant data, wherein the relevant data includes data center computing resources usage data, refrigeration equipment energy consumption data, and environmental factor data;

[0021] A correlation analysis module is used to perform correlation analysis on the pre-processed related data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption;

[0022] A model selection module, used to select a model according to the key parameters and modeling objectives, train the selected model using historical data to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluate the performance indicators of the selected model;

[0023] The joint optimization module is used to select the optimization algorithm to perform the joint optimization logic of data center computing power resources and refrigeration equipment energy consumption according to the modeling objectives and constraints, run the optimization algorithm, and iteratively adjust the allocation of data center computing power resources and the operating parameters of refrigeration equipment. When the optimization algorithm converges or reaches the preset number of iterations, performance indicators for evaluating the correlation between data center computing power resources and refrigeration equipment energy consumption are obtained.

[0024] As a preferred solution of the data center computing power resource and refrigeration equipment energy consumption correlation analysis device, in the related data acquisition processing module, the data center computing power resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data;

[0025] In the relevant data acquisition processing module, preprocessing the relevant data includes removing abnormal values ​​and missing values, and normalizing or standardizing the relevant data;

[0026] The related data acquisition and processing module is also used to divide the data into time windows or perform lag processing to determine the input layer, hidden layer, and output layer structure of the model, and select the activation function, loss function, and optimizer.

[0027] As a preferred solution for the data center computing power resources and refrigeration equipment energy consumption correlation analysis device, in the model selection module, the model selected according to the key parameters and modeling objectives is a physical model, a mathematical model, a simulation model or a machine learning model; the performance index for evaluating the selected model includes a prediction error index and a stability index;

[0028] The modeling objectives are energy consumption reduction rate and computing resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits;

[0029] Also included is a constraint condition conversion module, which is used to mathematize the constraint condition, convert the constraint condition into a mathematical expression or an inequality, and set the initial parameters of the algorithm according to the constraint condition;

[0030] In the joint optimization module, the selected optimization algorithms are particle swarm optimization, genetic algorithm, and deep reinforcement learning algorithm.

[0031] As a preferred solution for the device for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption, in the joint optimization module, a new solution is generated according to the algorithm rules, the objective function value of the new solution is evaluated, and it is screened according to the constraints, the current optimal solution is updated, and iteration continues until the termination condition is met.

[0032] From the above, it can be seen that the technical solution provided by the present invention obtains relevant data on data center computing power resources and refrigeration equipment energy consumption, and pre-processes the relevant data, wherein the relevant data includes data center computing power resource usage data, refrigeration equipment energy consumption data and environmental factor data; performs correlation analysis on the pre-processed relevant data to obtain key parameters related to data center computing power resources and refrigeration equipment energy consumption; selects a model based on the key parameters and modeling objectives, trains the selected model using historical data to learn the correlation between computing power resources and refrigeration equipment energy consumption, and evaluates the performance indicators of the selected model; selects an optimization algorithm based on the modeling objectives and constraints to perform a joint optimization logic of data center computing power resources and refrigeration equipment energy consumption, runs the optimization algorithm, and iteratively adjusts the allocation of data center computing power resources and the operating parameters of refrigeration equipment. When the optimization algorithm converges or reaches a preset number of iterations, performance indicators for evaluating the correlation between data center computing power resources and refrigeration equipment energy consumption are obtained. The present invention has a significant impact on improving model performance by adjusting the model structure or adding feature variables. By deeply analyzing the current performance of the model, the deficiencies of the model can be identified, and then the model structure can be adjusted or new feature variables can be introduced in a targeted manner; the present invention significantly improves the prediction accuracy, generalization ability and interpretability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 A schematic diagram of a flow chart of a method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption provided by an embodiment of the present invention;

[0035] Figure 2An architecture diagram of a device for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption provided by an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0038] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The words "include" or "comprise" and the like used in the embodiments of the present invention mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but do not exclude other elements or objects.

[0039] There is a close relationship between the computing resources of a data center and the energy consumption of cooling equipment. Generally speaking, the increase in computing resources is often accompanied by an increase in the energy consumption of cooling equipment, but the optimization and management between the two are crucial to the overall energy efficiency of the data center.

[0040] First of all, computing resources are the core of data centers, which support various computing, storage and processing tasks. With the improvement of computing resources in data centers, such as the increase in the number of servers and the enhancement of computing power, the energy consumption of data centers will also increase accordingly. Among them, the energy consumption of IT equipment (such as servers, switches, etc.) accounts for most of the total energy consumption of data centers. These devices will generate a lot of heat during all-weather operation, and refrigeration equipment is needed to maintain a suitable operating temperature.

[0041] Secondly, refrigeration equipment plays a vital role in data centers. They provide a suitable temperature and humidity environment for data centers through cooling systems to ensure that IT equipment can operate stably and efficiently. However, refrigeration equipment is also one of the parts with higher energy consumption in data centers. With the improvement of computing resources and the increase in the density of IT equipment, the load of refrigeration equipment will also increase accordingly, resulting in an increase in energy consumption. In the overall energy consumption of data centers, there is a dynamic balance between computing resources and the energy consumption of refrigeration equipment. On the one hand, the improvement of computing resources requires more refrigeration equipment to maintain a suitable operating environment; on the other hand, the increase in energy consumption of refrigeration equipment will also have an adverse effect on the operating costs and environmental impact of data centers.

[0042] The modeling methods in the existing technology have certain limitations, high costs, and long cycles. The physical model is mainly established through experiments and tests, which requires a lot of time and cost, and cannot fully consider complex logical relationships. Although the physical model has high reliability and good result traceability, it cannot fully consider the complex logical relationships in the data center, which limits its application in complex systems. Data modeling and solver selection are difficult. Mathematical models rely on data analysis and mathematical methods, but it is more difficult to select data modeling and solvers. Error analysis is complex, and the error analysis of mathematical models is relatively complex, which increases the uncertainty in practical applications. The limitations of static and dynamic models. Static models are based on fixed attributes and constraints and may not be able to adapt to real-time changing environments. Although dynamic models are based on real-time monitoring data, there are still challenges in real-time calculation and adjustment.

[0043] In view of this, in order to solve the problem that the modeling methods in the prior art have certain limitations, high cost, and long cycle, the physical model is mainly established through experiments and tests, which requires a lot of time and cost, and cannot fully consider the complex logical relationships of the data center. The embodiment of the present invention provides a method and device for analyzing the correlation between computing power resources and refrigeration equipment energy consumption in a data center. The following is the specific content of the embodiment of the present invention.

[0044] See also Figure 1 The embodiment of the present invention provides a method for analyzing the correlation between computing power resources in a data center and energy consumption of refrigeration equipment, comprising the following steps:

[0045] S1. Obtain data related to data center computing resources and refrigeration equipment energy consumption, and pre-process the data, wherein the data includes data center computing resources usage data, refrigeration equipment energy consumption data, and environmental factor data;

[0046] S2. Performing correlation analysis on the preprocessed relevant data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption; selecting a model based on the key parameters and modeling objectives, using historical data to train the selected model to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluating the performance indicators of the selected model;

[0047] S3. According to the modeling objectives and constraints, select an optimization algorithm to perform the joint optimization logic of the data center computing resources and the energy consumption of the refrigeration equipment, run the optimization algorithm, and iteratively adjust the allocation of the data center computing resources and the operating parameters of the refrigeration equipment. When the optimization algorithm converges or reaches the preset number of iterations, obtain the performance indicators for evaluating the correlation between the data center computing resources and the energy consumption of the refrigeration equipment.

[0048] In this embodiment, in step S1, the data center computing resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data; preprocessing the relevant data includes removing outliers and missing values, and normalizing or standardizing the relevant data. Preprocessing the relevant data also includes time window division or lag processing of the data to determine the input layer, hidden layer, and output layer structure of the model, and selecting activation functions, loss functions, and optimizers.

[0049] Specifically, computing resource usage data and refrigeration equipment energy consumption data are obtained from the data center monitoring system, and environmental factor data are recorded to provide a rich data foundation for modeling. The collected relevant data is cleaned, outliers and missing values ​​are removed, and the data is normalized or standardized to ensure that data of different dimensions are comparable. Dimensionality reduction technology is used to reduce data dimensions. Based on the results of correlation analysis, key feature variables are extracted as modeling input to improve modeling efficiency and model performance. Considering the characteristics of time series, it may be necessary to divide the data into time windows or lag processing, determine the input layer, hidden layer, and output layer structure of the model, and select appropriate activation functions, loss functions, and optimizers.

[0050] In this embodiment, in step S2, the model selected according to the key parameters and modeling objectives is a physical model, a mathematical model, a simulation model or a machine learning model; the performance index for evaluating the selected model includes a prediction error index and a stability index. The modeling objectives are energy consumption reduction rate and computing resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits.

[0051] Specifically, collect data related to the computing power resources and refrigeration equipment energy consumption of the data center, including server energy consumption, air conditioning energy consumption, temperature, and humidity, perform correlation analysis on the preprocessed data, find out the key parameters related to the computing power resources and refrigeration equipment energy consumption, select the appropriate model type, such as physical model, mathematical model, simulation model or machine learning model according to the characteristics of the data and the modeling objectives, use historical data to train the selected model to learn the correlation between computing power resources and refrigeration equipment energy consumption, and evaluate the performance of the model, including prediction error, stability and other indicators.

[0052] Among them, the prediction accuracy is improved by establishing a correlation model between the computing power resources of the data center and the energy consumption of the refrigeration equipment: the key parameters are found through correlation analysis, and the appropriate model type is selected for training, which improves the model's prediction accuracy of the relationship between computing power resources and the energy consumption of refrigeration equipment.

[0053] In this embodiment, in step S3, the constraint conditions are mathematized, the constraint conditions are converted into mathematical expressions or inequalities, and the initial parameters of the algorithm are set according to the constraint conditions; the selected optimization algorithms are particle swarm algorithm, genetic algorithm, and deep reinforcement learning algorithm.

[0054] Specifically, joint optimization calculations include goal setting, constraint analysis, optimization algorithm selection, algorithm implementation, algorithm iteration and optimization, result evaluation and application. It is necessary to clarify the goal of joint optimization, select appropriate optimization algorithms such as particle swarm algorithm, genetic algorithm, and deep reinforcement learning according to the goal and constraints, and implement the joint optimization logic of computing resources and refrigeration equipment energy consumption within the selected optimization algorithm framework. The optimization algorithm is run to continuously adjust the allocation of computing resources and the operating parameters of the refrigeration equipment through iteration. When the algorithm converges or reaches the preset number of iterations, the performance of the final solution is evaluated, including indicators such as energy consumption reduction rate and computing resource utilization rate.

[0055] Among them, by setting clear joint optimization goals, such as energy consumption reduction rate, computing power resource utilization rate, etc., a clear direction is provided for optimization calculation. According to the goals and constraints, appropriate optimization algorithms are selected, such as particle swarm optimization, genetic algorithm, deep reinforcement learning, etc., to improve the efficiency and accuracy of optimization calculation. Under the optimization algorithm framework, the joint optimization logic of computing power resources and refrigeration equipment energy consumption is realized to ensure the coordination and balance between the two in the optimization process. By evaluating the performance of the final solution, the optimization effect can be intuitively understood, providing strong support for practical applications.

[0056] Among them, all constraints on computing resource allocation and refrigeration equipment operation are listed, providing comprehensive constraint information for optimization calculations. The constraints are converted into mathematical expressions or inequalities to facilitate subsequent optimization algorithm processing and improve the accuracy and efficiency of optimization calculations. The initial parameters of the algorithm are set according to the constraints, such as population size, number of iterations, learning rate, etc., to provide reasonable parameter configuration for the operation of the optimization algorithm.

[0057] In a possible embodiment, the model is validated using a validation data set, the prediction performance of the model is evaluated, the risk of overfitting is reduced by using methods such as cross-validation, and the model structure is adjusted or feature variables are added according to the evaluation results. Regularization techniques are used to improve the generalization ability of the model, ensuring the accuracy and reliability of the model. Cross-validation and other methods are used to reduce the risk of overfitting and improve the generalization ability of the model. The model structure is adjusted or feature variables are added according to the evaluation results, and regularization techniques are used to further improve the generalization ability of the model, providing a more robust model for practical applications.

[0058] Among them, L1 regularization tends to produce sparse model parameters, that is, some parameters are zero, which helps to simplify the model and improve its interpretability. L2 regularization tends to keep the model parameters small, thereby preventing the model from being too complex on the training data and improving its performance on unseen data.

[0059] In a possible embodiment, the iterative adjustment of the allocation of computing resources in the data center and the operating parameters of the refrigeration equipment includes: in each iteration process, generating a new solution according to the algorithm rules, evaluating the objective function value of the new solution, and screening it according to the constraint conditions, updating the current optimal solution, and continuing to iterate until the termination condition is met.

[0060] Specifically, in each iteration, a new solution is generated according to the algorithm rules and its objective function value is evaluated, which improves the efficiency of the optimization calculation. The generated solutions are screened according to the constraints to ensure the feasibility and compliance of the optimization results. The current optimal solution is continuously updated and iterated until the termination condition is met to ensure the accuracy and stability of the optimization results. By setting flexible termination conditions, the length and depth of the optimization process can be adjusted according to actual needs.

[0061] In summary, the present invention has a significant impact on the improvement of model performance by adjusting the model structure or adding feature variables. By deeply analyzing the current performance of the model, it is possible to identify in which aspects the model is deficient, and then adjust the model structure or introduce new feature variables in a targeted manner. This adjustment can not only help the model better capture the key information in the data and improve the prediction accuracy of the model, but also make the model more adaptable to different data sets and scenarios, and enhance its generalization ability. Specifically, adjusting the model structure may involve increasing or decreasing the number of layers, the number of neurons, the connection method, etc. of the model to optimize the complexity and fitting ability of the model. Adding feature variables enriches the input of the model by introducing more information related to the prediction target, thereby improving its prediction accuracy. These adjustments are all based on an in-depth understanding and analysis of the data, aiming to make the model more in line with the needs of actual problems. Secondly, the use of regularization technology is also an important means to improve the generalization ability of the model. Regularization technology limits the excessive or too small model parameters by adding a penalty term to the loss function of the model, thereby preventing the model from overfitting or underfitting. L1 regularization tends to produce sparse model parameters, that is, some parameters are zero, which helps to simplify the model and improve its interpretability. L2 regularization tends to keep the model parameters at a smaller value, thereby avoiding the model from being too complex on the training data and improving its performance on unseen data. The present invention adjusts the model structure, increases feature variables, and uses regularization techniques to significantly improve the prediction accuracy, generalization ability, and interpretability of the model. These optimization measures not only make the model perform better on the current data set, but also provide it with stronger adaptability and robustness in future application scenarios. Therefore, these measures are of great significance for improving the overall performance of the model.

[0062] It should be noted that the method of the embodiment of the present invention can be performed by a single device, such as a computer or a server. The method of this embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present invention, and the multiple devices will interact with each other to complete the described method.

[0063] It should be noted that some embodiments of the present invention are described above. In some cases, the actions or steps recorded can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0064] See also Figure 2Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, an embodiment of the present invention further provides a device for analyzing the correlation between computing power resources and refrigeration equipment energy consumption in a data center, including:

[0065] The relevant data acquisition processing module 100 is used to obtain the relevant data of the data center computing power resources and the energy consumption of the refrigeration equipment, and pre-process the relevant data, wherein the relevant data includes the data center computing power resource usage data, the refrigeration equipment energy consumption data and the environmental factor data;

[0066] A correlation analysis module 200 is used to perform correlation analysis on the pre-processed related data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption;

[0067] A model selection module 300 is used to select a model according to the key parameters and the modeling objectives, train the selected model using historical data to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluate the performance indicators of the selected model;

[0068] The joint optimization module 400 is used to select an optimization algorithm to perform the joint optimization logic of the data center computing power resources and the energy consumption of the refrigeration equipment according to the modeling objectives and constraints, run the optimization algorithm, and iteratively adjust the allocation of the data center computing power resources and the operating parameters of the refrigeration equipment. When the optimization algorithm converges or reaches a preset number of iterations, performance indicators for evaluating the correlation between the data center computing power resources and the energy consumption of the refrigeration equipment are obtained.

[0069] In this embodiment, in the relevant data acquisition processing module 100, the data center computing resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data;

[0070] In the relevant data acquisition processing module 100, preprocessing the relevant data includes removing abnormal values ​​and missing values, and normalizing or standardizing the relevant data;

[0071] The related data acquisition processing module 100 is also used to perform time window division or lag processing on the data to determine the input layer, hidden layer, and output layer structure of the model, and select the activation function, loss function, and optimizer.

[0072] In this embodiment, in the model selection module 300, the model selected according to the key parameters and the modeling target is a physical model, a mathematical model, a simulation model or a machine learning model; the performance index for evaluating the selected model includes a prediction error index and a stability index;

[0073] The modeling objectives are energy consumption reduction rate and computing resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits;

[0074] Also included is a constraint condition conversion module 500, which is used to mathematize the constraint condition, convert the constraint condition into a mathematical expression or an inequality, and set the initial parameters of the algorithm according to the constraint condition;

[0075] In the joint optimization module 400, the selected optimization algorithm is a particle swarm algorithm, a genetic algorithm, or a deep reinforcement learning algorithm.

[0076] In this embodiment, the joint optimization module 400 generates a new solution according to the algorithm rules, evaluates the objective function value of the new solution, and screens it according to the constraint conditions, updates the current optimal solution, and continues to iterate until the termination condition is met.

[0077] The device of the above embodiment is used to implement a corresponding method for analyzing the correlation between data center computing power resources and refrigeration equipment energy consumption in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0078] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for analyzing the correlation between computing power resources in a data center and energy consumption of refrigeration equipment as described in any of the above embodiments is implemented.

[0079] Figure 3 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 610, a memory 620, an input / output interface 630, a communication interface 640, and a bus 650. The processor 610, the memory 620, the input / output interface 630, and the communication interface 640 are connected to each other through the bus 650 in the device.

[0080] The processor 610 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0081] The memory 620 may be implemented in the form of ROM (Read On ly Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 620 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 620 and are called and executed by the processor 610.

[0082] The input / output interface 630 is used to connect the input / output module to realize information input and output. The input / output module can be configured as a component in the device (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0083] The communication interface 640 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WI FI, Bluetooth, etc.).

[0084] The bus 650 comprises a pathway for transmitting information between the various components of the device (eg, the processor 610, the memory 620, the input / output interface 630, and the communication interface 640).

[0085] It should be noted that, although the above device only shows the processor 610, the memory 620, the input / output interface 630, the communication interface 640 and the bus 650, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0086] The electronic device of the above embodiment is used to implement a corresponding method for analyzing the correlation between data center computing power resources and refrigeration equipment energy consumption in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0087] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present invention also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute a method for analyzing the correlation between data center computing power resources and refrigeration equipment energy consumption as described in any of the above embodiments.

[0088] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0089] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute a method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0090] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0091] In addition, to simplify the description and discussion, and in order not to obscure the embodiments of the present invention, known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, the devices may be shown in the form of block diagrams to avoid obscuring the embodiments of the present invention, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present invention will be implemented (i.e., these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present invention, it will be apparent to those skilled in the art that embodiments of the present invention may be implemented without these specific details or with variations in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0092] Although the present invention has been described in conjunction with specific embodiments of the present invention, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0093] The embodiments of the present invention are intended to cover all such substitutions, modifications and variations that fall within the scope of the protection claimed. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the present invention.

Claims

1. Correlation analysis method between data center computing resources and refrigeration equipment energy consumption, including: include: Obtain data related to data center computing resources and refrigeration equipment energy consumption, and preprocess the related data, wherein the related data includes data center computing resources usage data, refrigeration equipment energy consumption data, and environmental factor data; Performing correlation analysis on the preprocessed relevant data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption; selecting a model based on the key parameters and modeling objectives, using historical data to train the selected model to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluating the performance indicators of the selected model; According to the modeling objectives and constraints, an optimization algorithm is selected to perform the joint optimization logic of data center computing resources and refrigeration equipment energy consumption. The optimization algorithm is run to iteratively adjust the allocation of data center computing resources and the operating parameters of refrigeration equipment. When the optimization algorithm converges or reaches the preset number of iterations, performance indicators for evaluating the correlation between data center computing resources and refrigeration equipment energy consumption are obtained.

2. The method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 1, wherein: The data center computing resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data; The preprocessing of the relevant data includes removing outliers and missing values, and normalizing or standardizing the relevant data.

3. The method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 2, wherein: Preprocessing the relevant data also includes dividing the data into time windows or performing lag processing to determine the input layer, hidden layer, and output layer structure of the model, and selecting an activation function, a loss function, and an optimizer.

4. The method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 1, wherein: The model selected according to the key parameters and modeling objectives is a physical model, a mathematical model, a simulation model or a machine learning model; The performance indicators for evaluating the selected models include prediction error indicators and stability indicators.

5. The method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 1, wherein: The modeling objectives are energy consumption reduction rate and computing resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits; The method further includes mathematizing the constraint conditions, converting the constraint conditions into mathematical expressions or inequalities, and setting initial parameters of the algorithm according to the constraint conditions; The optimization algorithms selected are particle swarm optimization, genetic algorithm, and deep reinforcement learning algorithm.

6. The method for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 1, wherein: The iterative adjustment of the allocation of computing resources in the data center and the operating parameters of the refrigeration equipment includes: In each iteration, a new solution is generated according to the algorithm rules, the objective function value of the new solution is evaluated, and it is screened according to the constraints, the current optimal solution is updated, and the iteration continues until the termination condition is met.

7. Data center computing resources and refrigeration equipment energy consumption correlation analysis device, including: include: A relevant data acquisition processing module is used to acquire relevant data on data center computing resources and refrigeration equipment energy consumption, and pre-process the relevant data, wherein the relevant data includes data center computing resources usage data, refrigeration equipment energy consumption data, and environmental factor data; A correlation analysis module is used to perform correlation analysis on the pre-processed related data to obtain key parameters related to data center computing resources and refrigeration equipment energy consumption; A model selection module, used to select a model according to the key parameters and modeling objectives, train the selected model using historical data to learn the correlation between computing resources and refrigeration equipment energy consumption, and evaluate the performance indicators of the selected model; The joint optimization module is used to select the optimization algorithm to perform the joint optimization logic of data center computing power resources and refrigeration equipment energy consumption according to the modeling objectives and constraints, run the optimization algorithm, and iteratively adjust the allocation of data center computing power resources and the operating parameters of refrigeration equipment. When the optimization algorithm converges or reaches the preset number of iterations, performance indicators for evaluating the correlation between data center computing power resources and refrigeration equipment energy consumption are obtained.

8. The device for analyzing the correlation between data center computing resources and refrigeration equipment energy consumption according to claim 7, wherein: In the relevant data acquisition processing module, the data center computing resource usage data includes server energy consumption data, the refrigeration equipment energy consumption data includes air conditioning energy consumption data, and the environmental factor data includes temperature and humidity data; In the relevant data acquisition processing module, preprocessing the relevant data includes removing abnormal values ​​and missing values, and normalizing or standardizing the relevant data; The related data acquisition and processing module is also used to divide the data into time windows or perform lag processing to determine the input layer, hidden layer, and output layer structure of the model, and select the activation function, loss function, and optimizer.

9. The device for analyzing the correlation between computing power resources and refrigeration equipment energy consumption in a data center according to claim 7, wherein: In the model selection module, the model selected according to the key parameters and the modeling objectives is a physical model, a mathematical model, a simulation model or a machine learning model; the performance index for evaluating the selected model includes a prediction error index and a stability index; The modeling objectives are energy consumption reduction rate and computing resource utilization rate; the constraints include computing power requirements, equipment capacity, and ambient temperature limits; Also included is a constraint condition conversion module, which is used to mathematize the constraint condition, convert the constraint condition into a mathematical expression or an inequality, and set the initial parameters of the algorithm according to the constraint condition; In the joint optimization module, the selected optimization algorithms are particle swarm optimization, genetic algorithm, and deep reinforcement learning algorithm.

10. The device for analyzing the correlation between computing power resources and refrigeration equipment energy consumption in a data center according to claim 7, wherein: In the joint optimization module, a new solution is generated according to the algorithm rules, the objective function value of the new solution is evaluated, and it is screened according to the constraint conditions, the current optimal solution is updated, and iteration continues until the termination condition is met.

Images