Computing power optimization method and device of data center, electronic equipment and storage medium

Abstract

The invention provides a computing power optimization method and device of a data center, electronic equipment and a storage medium. The method comprises the steps of obtaining monitoring data corresponding to each computing power node of the data center in a historical time period; for each computing power node, inputting the monitoring data into a pre-trained prediction model, and outputting a computing power demand of a future time period having a corresponding relationship with the historical time period through the prediction model; based on the computing power demand of each computing power node, the computing power of the data center is optimized, so that the technical problem of unreasonable computing power deployment of the data center in the prior art is solved, and the data security of the data center is ensured.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, electronic device and storage medium for optimizing computing power of a data center. Background Art

[0002] In the digital age, computing power is like the "heart", providing power for the development of technologies such as artificial intelligence, big data, and cloud computing. As an important carrier of computing power, data centers are developing towards ultra-large scale to meet the growing computing and storage needs.

[0003] However, as society's demand for computing power has exploded, the computing power deployed in data centers may not be able to meet the computing power demand. Insufficient computing power may affect the data security and reliability of data centers. On the one hand, due to insufficient computing power, some key tasks may not be processed in time, resulting in an increased risk of data loss or damage. On the other hand, in order to cope with the problem of insufficient computing power, data centers may need to adopt some temporary solutions, such as overload operation, resource reuse, etc. These solutions may increase the complexity of the system and reduce the stability and reliability of the system. Therefore, reasonable computing power node deployment in data centers is very necessary for data security. Summary of the invention

[0004] In view of this, the purpose of this application is to propose a computing power optimization method, device, electronic device and storage medium for a data center to overcome all or part of the deficiencies in the prior art.

[0005] Based on the above-mentioned purpose, the present application provides a method for optimizing the computing power of a data center, including: obtaining monitoring data corresponding to each computing power node of the data center within a historical time period; for each computing power node, inputting the monitoring data into a pre-trained prediction model, and outputting the computing power requirements for a future time period that corresponds to the historical time period through the prediction model; based on the computing power requirements of each computing power node, optimizing the computing power of the data center.

[0006] Optionally, the computing power node includes computing power data; and optimizing the computing power of the data center based on the computing power demand of each computing power node includes: for each computing power node, in response to determining that the difference between the computing power demand and the computing power data is greater than a predetermined value, replacing the computing power node based on the computing power demand.

[0007] Optionally, replacing the computing power node based on the computing power requirement includes: deploying a new computing power node and deleting the computing power node, wherein the computing power data of the new computing power node is greater than the computing power requirement.

[0008] Optionally, before inputting the monitoring data corresponding to the computing power node into a pre-trained prediction model, the method includes: performing a preprocessing operation on the monitoring data.

[0009] Optionally, the preprocessing operation includes a data cleaning operation and a normalization operation.

[0010] Optionally, the training method of the prediction model includes: obtaining historical monitoring data of the computing power nodes of the data center; dividing all historical monitoring data into training set data and test set data according to a preset ratio; initializing the parameters of the prediction model, and iteratively training the prediction model using the training set data; testing the iteratively trained prediction model based on the test set data to obtain a test accuracy, and when the test accuracy is greater than a preset accuracy, a trained prediction model is obtained.

[0011] Optionally, after optimizing the computing power of the data center based on the computing power requirements of each computing power node, the method includes: monitoring the data center; and in response to determining that an abnormal state occurs in the data center, reporting the abnormal state.

[0012] Based on the same inventive concept, the present application also provides a computing power optimization device for a data center, including: an acquisition module, configured to acquire monitoring data corresponding to each computing power node of the data center within a historical time period; an output module, configured to input the monitoring data into a pre-trained prediction model for each computing power node, and output the computing power requirements for a future time period that corresponds to the historical time period through the prediction model; an optimization module, configured to optimize the computing power of the data center based on the computing power requirements of each computing power node.

[0013] Based on the same inventive concept, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the method as described above when executing the computer program.

[0014] Based on the same inventive concept, the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the method as described above.

[0015] From the above, it can be seen that the computing power optimization method, device, electronic device and storage medium of the data center provided by the present application include obtaining monitoring data corresponding to each computing power node of the data center in the historical time period. For each computing power node, the monitoring data is input into a pre-trained prediction model, and the prediction model outputs the computing power demand of the future time period that corresponds to the historical time period, so as to achieve the purpose of accurately predicting the computing power demand. Based on the computing power demand of each computing power node, the computing power of the data center is optimized to ensure the reasonable deployment of the computing power of the data center, thereby ensuring the data security of the data center. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flowchart of a method for optimizing computing power of a data center according to an embodiment of the present application;

[0018] Figure 2 A schematic diagram of the structure of a computing power optimization device for a data center according to an embodiment of the present application;

[0019] Figure 3 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0021] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0022] As mentioned in the background technology section, in the digital age, computing power is like a "heart", providing power for the development of technologies such as artificial intelligence, big data, and cloud computing. As an important carrier of computing power, data centers are developing towards ultra-large scale to meet the growing computing and storage needs. Data centers are buildings where electronic information equipment is centrally placed and their operating environment is provided. A large number of servers, storage devices, network equipment, and other related physical infrastructure are concentrated here, and are equipped with corresponding security equipment and monitoring and management systems. Computing power has become a new quality productivity that promotes social development. As an important carrier of computing power, data centers are developing rapidly and have become a "new infrastructure" to support information construction.

[0023] Smart energy management is a comprehensive energy management system that deeply integrates modern information technology, big data analysis, cloud computing, Internet of Things technology, artificial intelligence algorithms, and advanced sensor technology, aiming to achieve intelligent monitoring, optimization, and management of the entire energy chain from production, transmission, distribution to consumption. This system can not only collect and process massive amounts of energy data in real time, but also mine the value behind the data through advanced data analysis methods, providing support for precise scheduling, efficient configuration, and intelligent decision-making of energy. Under the framework of smart energy management, the production end of energy can achieve more efficient and clean production methods through intelligent technology, such as using intelligent grid connection and scheduling of renewable energy to improve the utilization rate of clean energy. The transmission and distribution links use technologies such as smart grids and smart pipelines to achieve precise transmission and dynamic distribution of energy and reduce transmission losses. On the consumer side, the smart energy management system can monitor and analyze the energy usage behavior of users in real time, and achieve refined management of energy usage through smart meters, smart home systems and other equipment, encouraging users to save energy and improve energy efficiency. In addition, smart energy management also emphasizes cross-system collaboration and integration. By building a unified energy management platform, it integrates multiple energy systems such as electricity, gas, and heat, as well as energy needs in multiple fields such as transportation, construction, and industry, to achieve overall optimization of the energy system. This integrated management method can not only improve the flexibility and reliability of the energy system, but also promote the optimization and transformation of the energy structure and promote the green development of the economy and society.

[0024] The computing power optimization strategy of data centers based on smart energy management involves the deep integration of multiple technical fields, including cloud computing, big data, the Internet of Things, artificial intelligence, etc. The demand for computing power is growing explosively, but the computing power nodes deployed in data centers may not be able to meet the computing power demand. Insufficient computing power may affect the data security and reliability of data centers. On the one hand, due to insufficient computing power, some key tasks may not be processed in time, resulting in an increased risk of data loss or damage. On the other hand, in order to cope with the problem of insufficient computing power, data centers may need to adopt some temporary solutions, such as overload operation and resource reuse, which may increase the complexity of the system and reduce the stability and reliability of the system. The security protection system of data centers faces huge challenges, and how to ensure the security of data has become an urgent problem to be solved. Therefore, reasonable deployment of computing power nodes in data centers is very necessary for data security.

[0025] In view of this, the present application embodiment proposes a method for optimizing the computing power of a data center, referring to Figure 1 , including the following steps:

[0026] Step 101, obtain the monitoring data corresponding to each computing node of the data center within the historical time period.

[0027] In this step, the data center may have been in use for a long time. As the scale of data processing increases, the computing power deployed cannot meet the current computing power requirements. Therefore, it is necessary to optimize the computing power of the data center. Obtain the monitoring data corresponding to each computing power node in the data center during the historical time period. The computing power nodes in the data center are key components with data processing and computing capabilities. They provide efficient and reliable computing services by integrating hardware and software resources. Monitoring data includes but is not limited to key performance indicators such as CPU usage, memory occupancy, disk I / O, and network bandwidth.

[0028] Step 102: for each computing power node, the monitoring data is input into a pre-trained prediction model, and the prediction model outputs the computing power demand for a future time period that corresponds to the historical time period.

[0029] In this step, in order to determine whether the computing power node can meet the computing power demand, it is necessary to predict the computing power demand of the computing power node, so that when the computing power node is far from meeting the computing power demand, the computing power node can be processed accordingly in advance. For each computing power node, the monitoring data is input into the pre-trained prediction model, and the prediction model outputs the computing power demand of the future time period that corresponds to the historical time period. The corresponding relationship can be the relationship between adjacent months of the same year. For example, if the computing power demand for the future time period in December this year needs to be predicted, the historical time period is November this year. The corresponding relationship can also be the relationship between the same months of adjacent years. For example, if the computing power demand for the future time period in December this year is predicted, the historical time period is December last year. The computing power demand of the future time period is predicted by the prediction model, so as to achieve the purpose of accurately predicting the computing power demand.

[0030] Step 103: Optimize the computing power of the data center based on the computing power requirements of each computing power node.

[0031] In this step, after determining the computing power requirements of each computing node in the future time period, the computing power required by each computing node in the future time period is known. The computing power of the existing data center may not meet the current computing power requirements due to the relatively long deployment time. Therefore, it is necessary to optimize the computing power of the data center based on the computing power requirements of each node to ensure the reasonable deployment of the computing power of the data center, thereby ensuring the data security of the data center.

[0032] It should be noted that in order to optimize the computing power of the data center, on the one hand, an adaptive smart energy management system can be deployed in the data center. The smart energy management system can monitor the energy consumption of the data center in real time, collect data and analyze it, integrate the computing power resources with the data of the energy management system, and realize the correlation analysis between computing power and energy consumption. Based on business needs, computing power resource status and energy consumption, a computing power resource scheduling algorithm is designed to realize the dynamic allocation and adjustment of computing power resources. Through simulation tests, historical data analysis and other means, the parameters of the scheduling algorithm are continuously optimized, the cooling system of the data center is evaluated, and its energy consumption, cooling effect and potential optimization space are analyzed. Through virtualization technology, multiple virtual machines are run on the same physical server. Deployment of a smart energy management system realizes real-time monitoring and data collection, and improves the intelligence level of energy management. On the other hand, a comprehensive assessment of the existing computing resources in the data center is conducted, including the usage of hardware resources such as CPU, GPU, and memory, performance bottlenecks, and potential for improvement. Based on business needs, technology development trends, and future expansion plans, the required scale and type of computing resources are predicted and planned. Performance testing tools are used to benchmark key applications, determine the bottlenecks of computing resources, analyze the computing demand characteristics of different business types, develop targeted computing resource allocation plans, consider technology development trends, such as AI and big data, and reserve space for computing resource expansion.

[0033] Through the above scheme, the monitoring data corresponding to each computing power node of the data center in the historical time period is obtained. For each computing power node, the monitoring data is input into the pre-trained prediction model, and the prediction model outputs the computing power demand of the future time period that corresponds to the historical time period, so as to achieve the purpose of accurately predicting the computing power demand. Based on the computing power demand of each computing power node, the computing power of the data center is optimized to ensure the reasonable deployment of the computing power of the data center, thereby ensuring the data security of the data center.

[0034] In some embodiments, the computing power node includes computing power data; the optimizing the computing power of the data center based on the computing power demand of each computing power node includes: for each computing power node, in response to determining that the difference between the computing power demand and the computing power data is greater than a predetermined value, replacing the computing power node based on the computing power demand.

[0035] In this embodiment, the computing power node includes computing power data, and the computing power data can characterize the computing power of the computing power node. For each computing power node, when the difference between the computing power demand and the computing power data is greater than a predetermined value, it means that the computing power that the computing power node can provide is far less than the computing power demand of the computing power node, where the predetermined value is a relatively large value determined based on historical experience. The computing power node is obviously unable to meet the growing computing power demand, so it is necessary to replace the computing power node to ensure that the deployment of the computing power node of the computing power center can meet the computing power demand, thereby ensuring the smooth progress of the computing task and making the computing power deployment of the data center reasonable.

[0036] When the difference between the computing power demand and the computing power data is less than or equal to the predetermined value, it means that the computing power node can complete the computing task. However, when the computing power demand is slightly greater than the computing power data, in order to improve computing efficiency, the scheduling algorithm can be used for task scheduling. List the configuration information of all servers in the data center, including CPU model, number of cores, frequency, GPU model and number, memory size, analyze the load of each server in the past period of time, and identify peak hours and inefficient periods. List the configuration information of all servers in the data center, providing basic data for resource management and optimization. By analyzing the load of each server in the past period of time, it is possible to identify peak hours and inefficient periods, providing a basis for resource scheduling and energy efficiency optimization. According to the actual needs of the data center, a scheduling algorithm that meets the business characteristics is designed. Based on business needs and resource status, a computing power resource scheduling algorithm is developed, and sufficient testing and verification are carried out. The existing cooling system is fully evaluated, a virtualization platform is deployed in the data center, server virtualization is realized, a comprehensive data center monitoring platform is built, and a continuous optimization mechanism for computing power optimization strategies is established based on monitoring data. Considering factors such as load balancing and energy efficiency, we introduced efficient cooling technologies such as intelligent dual-circulation fluorine pump multi-connected air conditioners, replaced or transformed existing cooling equipment, implemented resource pooling management to build a computing resource pool, integrated and uniformly managed the computing resources of virtual machines, developed a resource pool management platform, provided resource application, allocation, monitoring and other functions, realized dynamic scheduling and elastic expansion of resources, and met changes in business needs. We developed and fully tested the computing resource scheduling algorithm to ensure its feasibility and effectiveness in practical applications.

[0037] In some embodiments, replacing the computing power node based on the computing power requirement includes: deploying a new computing power node and deleting the computing power node, wherein the computing power data of the new computing power node is greater than the computing power requirement.

[0038] In this embodiment, since there is a large gap between the computing power provided by the computing power node and the computing power requirement of the computing power node, it means that the deployment of the computing power node is not reasonable. Therefore, a new computing power node is deployed and the original computing power node is deleted. Among them, the computing power data of the new computing power node is greater than the computing power requirement, which means that the new computing power node can ensure the smooth completion of the computing task. The computing power deployment of the data center is optimized so that the data center can successfully complete the computing task.

[0039] In some embodiments, before inputting the monitoring data corresponding to the computing power node into a pre-trained prediction model, the method includes: performing a pre-processing operation on the monitoring data.

[0040] In this embodiment, the directly acquired monitoring data may have low data quality. If the low-quality monitoring data is directly input into the prediction model, on the one hand, the accuracy of the prediction model output result is low, and on the other hand, the processing efficiency of the prediction model on the monitoring data is affected. Therefore, before the monitoring data is input into the prediction model, the monitoring data is preprocessed to improve the data quality of the monitoring data.

[0041] In some embodiments, the preprocessing operation includes a data cleaning operation and a normalization operation.

[0042] In this embodiment, due to sensor failure, data transmission error and other reasons, missing data or abnormal data may appear in the monitoring data, resulting in poor data quality of the monitoring data. In addition, the monitoring data may also have the problem of inconsistent dimensions, which also leads to poor data quality of the monitoring data. The poor data quality of the monitoring data will lead to low accuracy and low efficiency of the prediction results output by the subsequent prediction model. Therefore, in order to solve the problem of missing data or abnormal data in the monitoring data, the preprocessing operation includes a cleaning operation, which performs a cleaning operation on the monitoring data, avoids the above situation, and preliminarily improves the data quality of the monitoring data. In addition, in order to solve the problem of inconsistent data dimensions, the preprocessing operation includes a normalization operation, which performs a normalization operation on the monitoring data, avoids the above problem, and further improves the data quality of the monitoring data.

[0043] In some embodiments, the training method of the prediction model includes: obtaining historical monitoring data of the computing nodes of the data center; dividing all historical monitoring data into training set data and test set data according to a preset ratio; initializing the parameters of the prediction model, and using the training set data to iteratively train the prediction model; testing the iteratively trained prediction model based on the test set data to obtain a test accuracy, and when the test accuracy is greater than the preset accuracy, a trained prediction model is obtained.

[0044] In this embodiment, a data set is constructed by the historical monitoring data of the computing power nodes. When the historical monitoring data of the computing power nodes is relatively small, a data set of normal proportion is constructed by generating an adversarial network; when the historical monitoring data of the computing power nodes is relatively large, the historical monitoring data of the computing power nodes can be directly used for training without using an adversarial network to generate data. It can be divided into training sets and test sets in a ratio of 8:2, and the prediction model can be iteratively trained using training samples, parameter adjustments can be made, and the prediction model can be tested using the test set data to obtain the test accuracy. When the test accuracy is greater than or equal to the preset accuracy, a trained prediction model is obtained. By fully training the prediction model, the accuracy of the output results of the prediction model is ensured.

[0045] In some embodiments, after optimizing the computing power of the data center based on the computing power requirements of each computing power node, the method includes: monitoring the data center; and in response to determining that an abnormal state occurs in the data center, reporting the abnormal state.

[0046] In this embodiment, in order to ensure that the data center can always operate normally, the data center is monitored. The data center can monitor various energy consumption data in real time, including the consumption of key systems such as electricity and cooling. This real-time monitoring capability provides a precise management basis for the data center, enabling operation and maintenance personnel to quickly respond to energy consumption anomalies and adjust operation strategies in a timely manner, thereby effectively avoiding energy waste. Ensure that the system can monitor the energy consumption of the data center in real time, install sensors and smart meters, collect energy consumption data of systems such as electricity and cooling, configure system parameters, ensure the accuracy and real-time nature of data collection, and ensure the accuracy and real-time nature of data collection, which provides a reliable basis for energy consumption analysis and management. Develop a data interface to integrate computing resource usage data and energy consumption data into the same platform, realize the visualization of data, and facilitate operation and maintenance personnel to intuitively understand the operation status of the data center. After installing sensors and smart meters, the data center can collect a large amount of detailed energy consumption data, which provides strong data support for the operation decision-making of the data center. Through in-depth analysis of these data, operation and maintenance personnel can more accurately understand the energy consumption status of the data center, identify peak and trough periods of energy consumption, and the energy consumption correlation between various systems. This will help develop more scientific and reasonable energy management strategies, achieve refined management of energy consumption, and improve energy utilization efficiency. Real-time monitoring and data analysis will help discover bottlenecks and problems in data center energy utilization.

[0047] For example, by monitoring the energy consumption data of the cooling system, it is possible to find out that the cooling efficiency is poor or that the system is overcooled, and then adjust the cooling strategy in time to improve the cooling efficiency and reduce energy consumption. At the same time, monitoring the energy consumption data of the power system can also help operation and maintenance personnel optimize power distribution, avoid power waste, improve overall energy utilization efficiency, and promote green and sustainable development. As a large energy consumer, the energy management of data centers is of great significance to promoting green and sustainable development. Through the implementation of the above measures, data centers can significantly reduce energy consumption, reduce carbon emissions, and contribute to environmental protection. At the same time, this green and sustainable operation model can also help enhance the brand image and market competitiveness of data centers, and attract more customers who pay attention to environmental protection and sustainable development. Real-time monitoring and precise management can timely discover and solve energy consumption abnormalities and avoid unnecessary energy waste. At the same time, optimizing energy management strategies through data analysis can further improve energy utilization efficiency and reduce operating costs. This is undoubtedly an important way to improve economic benefits for data centers.

[0048] In the event of an abnormal state in the data center, the abnormal state will be reported. The abnormal state of the data center includes but is not limited to hardware failure and software failure, among which hardware failure includes equipment failure, line failure and port failure, and software failure includes operating system abnormality and application error. The ability to monitor various energy consumption data of the data center in real time provides a precise management foundation for the data center, allowing operation and maintenance personnel to respond quickly to energy consumption abnormalities, thereby effectively avoiding energy waste and unnecessary cost expenditures. By reporting the monitored abnormal state, it is ensured that the abnormal state of the data center can be corrected in time, ensuring the smooth operation of the data center.

[0049] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to 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 application, and the multiple devices will interact with each other to complete the described method.

[0050] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do 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.

[0051] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a computing power optimization device for a data center.

[0052] refer to Figure 2 , the computing power optimization device of the data center comprises:

[0053] The acquisition module 10 is configured to acquire monitoring data corresponding to each computing node of the data center within a historical time period.

[0054] The output module 20 is configured to input the monitoring data into a pre-trained prediction model for each computing power node, and output the computing power demand for a future time period corresponding to the historical time period through the prediction model.

[0055] The optimization module 30 is configured to optimize the computing power of the data center based on the computing power requirements of each computing power node.

[0056] Through the above device, the monitoring data corresponding to each computing power node of the data center in the historical time period is obtained. For each computing power node, the monitoring data is input into the pre-trained prediction model, and the prediction model outputs the computing power demand of the future time period that corresponds to the historical time period, so as to achieve the purpose of accurately predicting the computing power demand. Based on the computing power demand of each computing power node, the computing power of the data center is optimized to ensure the reasonable deployment of the computing power of the data center, thereby ensuring the data security of the data center.

[0057] In some embodiments, the optimization module 30 is further configured so that the computing power node includes computing power data; for each computing power node, in response to determining that the difference between the computing power requirement and the computing power data is greater than a predetermined value, the computing power node is replaced based on the computing power requirement.

[0058] In some embodiments, the optimization module 30 is further configured to deploy new computing nodes and delete the computing nodes, wherein the computing power data of the new computing power nodes is greater than the computing power requirement.

[0059] In some embodiments, a preprocessing operation module is also included, and the preprocessing operation module is configured to perform a preprocessing operation on the monitoring data corresponding to the computing power node before inputting the monitoring data into a pre-trained prediction model.

[0060] In some embodiments, a preprocessing operation module is further included, and the preprocessing operation module is configured so that the preprocessing operation includes a data cleaning operation and a normalization operation.

[0061] In some embodiments, a training module is also included, which is configured to obtain historical monitoring data of the computing nodes of the data center; divide all historical monitoring data into training set data and test set data according to a preset ratio; initialize the parameters of the prediction model, and use the training set data to iteratively train the prediction model; test the iteratively trained prediction model based on the test set data to obtain the test accuracy, and when the test accuracy is greater than the preset accuracy, a trained prediction model is obtained.

[0062] In some embodiments, a reporting module is also included, which is configured to monitor the data center after optimizing the computing power of the data center based on the computing power requirements of each computing power node; in response to determining that an abnormal state occurs in the data center, the abnormal state is reported.

[0063] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0064] The device of the above-mentioned embodiment is used to implement the computing power optimization method of the corresponding data center in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0065] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application 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, the computing power optimization method of the data center as described in any of the above embodiments is implemented.

[0066] 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 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0067] The processor 1010 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.

[0068] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 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 1020 and are called and executed by the processor 1010.

[0069] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (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.

[0070] The communication interface 1040 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, WIFI, Bluetooth, etc.).

[0071] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0072] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, 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.

[0073] The electronic device of the above-mentioned embodiment is used to implement the computing power optimization method of the corresponding data center in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0074] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application 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 the computing power optimization method of the data center as described in any of the above embodiments.

[0075] 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.

[0076] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the computing power optimization method of the data center described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0077] Based on the same concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a computer program product, including computer program instructions. When the computer program instructions are run on a computer, the computer executes the computing power optimization method of the data center as described in any of the above embodiments, which has the beneficial effects of the corresponding method embodiments and will not be repeated here.

[0078] It should be noted that the embodiments of the present application can be further described in the following manner:

[0079] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0080] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0081] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0082] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0083] A person 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 application is limited to these examples. In line with the concept of the present application, 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 the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0084] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, 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 to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0085] Although the present application has been described in conjunction with specific embodiments of the present application, 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.

[0086] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the protection scope of the present application.

Claims

1. A method for optimizing computing power of a data center, characterized in that: include: Obtain monitoring data corresponding to each computing node in the data center within the historical time period; For each computing power node, the monitoring data is input into a pre-trained prediction model, and the prediction model outputs the computing power demand for a future time period that corresponds to the historical time period; Based on the computing power requirements of each computing power node, the computing power of the data center is optimized.

2. The method according to claim 1, characterized in that The computing power node includes computing power data; The optimizing the computing power of the data center based on the computing power requirement of each computing power node includes: For each computing power node, in response to determining that a difference between the computing power requirement and the computing power data is greater than a predetermined value, the computing power node is replaced based on the computing power requirement.

3. The method according to claim 2, characterized in that The replacing the computing power node based on the computing power requirement includes: Deploy a new computing power node and delete the computing power node, wherein the computing power data of the new computing power node is greater than the computing power requirement.

4. The method according to claim 1, characterized in that: Before inputting the monitoring data corresponding to the computing power node into the pre-trained prediction model, the method includes: A preprocessing operation is performed on the monitoring data.

5. The method according to claim 4, characterized in that The preprocessing operation includes a data cleaning operation and a normalization operation.

6. The method according to claim 1, characterized in that The training method of the prediction model includes: Obtaining historical monitoring data of the computing power nodes of the data center; Divide all historical monitoring data into training set data and test set data according to a preset ratio; Initializing the parameters of the prediction model, and iteratively training the prediction model using the training set data; The prediction model that has been iteratively trained is tested based on the test set data to obtain a test accuracy. When the test accuracy is greater than a preset accuracy, a trained prediction model is obtained.

7. The method according to claim 1, characterized in that After optimizing the computing power of the data center based on the computing power requirements of each computing power node, the method includes: Monitoring the data center; In response to determining that an abnormal state occurs in the data center, the abnormal state is reported.

8. A computing power optimization device for a data center, characterized in that: include: An acquisition module is configured to acquire monitoring data corresponding to each computing power node in the data center within a historical time period; An output module is configured to input the monitoring data into a pre-trained prediction model for each computing power node, and output the computing power demand of a future time period corresponding to the historical time period through the prediction model; The optimization module is configured to optimize the computing power of the data center based on the computing power requirements of each computing power node.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images