Smart energy management system for data center

Abstract

The embodiment of the invention provides a smart energy management system for a data center, and belongs to the technical field of energy optimization of a power grid. The management system comprises an integrated acquisition layer used for acquiring the power supply utilization efficiency, the carbon utilization efficiency and the water resource utilization efficiency of a management area; the service support layer is used for analyzing the power supply utilization efficiency, the carbon utilization efficiency and the water resource utilization efficiency to determine the overall state of the current management area; and the management application layer is used for adjusting system parameters of the management area according to the overall state feedback. Compared with a general energy supply strategy in the prior art, the management system provided by the invention improves the energy utilization efficiency due to the combination of the real-time overall state of the management area.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid energy optimization, and in particular to a smart energy management system for a data center. Background Art

[0002] In the data centers of modern power grids, the number of server devices is huge and dense, and the stable operation of these devices is crucial for data collection, processing and analysis of the entire power grid. In order to ensure that the server can operate at the optimal operating temperature and avoid performance degradation or damage due to overheating, it must be equipped with precise temperature control equipment. However, the strategies adopted by many data centers in temperature control management are relatively extensive and lack refined management, which leads to a series of problems such as energy waste and shortened equipment life.

[0003] Existing temperature control strategies often adjust the cooling system based on fixed temperature set points, without fully considering the actual needs of different areas, different time periods, and different load conditions within the data center. This relatively general and logically fixed approach ignores the dynamic changes in the internal environment of the data center, resulting in excessive use of cooling resources in some cases and insufficient use in other cases. In addition, the start, stop, and adjustment of temperature control equipment usually rely on simple timing control or single temperature feedback, without realizing intelligent linkage with factors such as server load and external environmental conditions, thus failing to achieve the optimal energy efficiency ratio. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a smart energy management system for a data center, which can improve the energy utilization efficiency of the data center.

[0005] In order to achieve the above object, an embodiment of the present invention provides a smart energy management system for a data center, including:

[0006] An integrated data collection layer for collecting power efficiency, carbon efficiency, and water efficiency in the management area;

[0007] A service support layer, used for analyzing the power utilization efficiency, carbon utilization efficiency and water resource utilization efficiency to determine the overall status of the current management area;

[0008] The management application layer is used to adjust the system parameters of the management area according to the overall status feedback.

[0009] Optionally, the integrated acquisition layer is used for:

[0010] According to formula (1), the power efficiency of the data center is calculated:

[0011] PUE = P total / PIT , (1)

[0012] Wherein, PUE is the power usage efficiency, P total is the total energy consumption of the data center, P IT Energy consumption of IT equipment.

[0013] Optionally, the integrated acquisition layer is used for:

[0014] The carbon utilization efficiency is calculated according to formula (2):

[0015] CUE=E 排放量 / P IT , (2)

[0016] Wherein, CUE is the carbon utilization efficiency, E 排放量 is the total carbon dioxide emissions from various sources of the data center, P IT Energy consumption of IT equipment.

[0017] Optionally, the integrated acquisition layer is used for:

[0018] According to formula (3), the water resource utilization efficiency of the data center is calculated:

[0019] WUE=L 总耗水 / P IT , (3)

[0020] Among them, L 总耗水 is the annual water consumption of the data center, P IT Energy consumption of IT equipment.

[0021] Optionally, the integrated acquisition layer is used for:

[0022] Performing data cleaning operations, data fusion operations, and data denoising operations on the power utilization efficiency, carbon utilization efficiency, and water resource utilization efficiency;

[0023] Connect to multiple types of data management system platforms to obtain the power utilization efficiency, carbon utilization efficiency and water resource utilization efficiency from the DCIM system, environmental monitoring system, power monitoring system and BA system of each data management system platform.

[0024] Optionally, the service support layer is used to:

[0025] Displaying the overall status through a display interface;

[0026] Determining whether the overall status triggers an alarm signal;

[0027] In the case where it is determined that the overall state triggers an alarm signal, basic information of the triggered overall state is recorded.

[0028] Optionally, the system parameters include temperature, humidity and load distribution of the management area;

[0029] The management application layer is used to adjust the temperature, humidity and load distribution according to the overall status feedback.

[0030] Optionally, the management application layer is used to:

[0031] Get the temperature performance value of the data center;

[0032] Determining whether the temperature performance value is greater than a preset temperature threshold;

[0033] When it is determined that the temperature performance value is less than or equal to the temperature threshold, the optimization value is determined according to formulas (4) to (6):

[0034]

[0035] LF=P IT / P UPS , (5)

[0036] P 制冷 =P IT +P dis / LF, (6)

[0037] in, is the optimized value, P 算力 is the computing power of the data center, P UPS is the UPS power consumption, P IT is the energy consumption of IT equipment in the data center, P dis is the power consumption of the power distribution equipment, and Optimize is an O(·) function.

[0038] Optionally, the management application layer is used to:

[0039] According to formula (7) to formula (10), the O(·) function is optimized.

[0040]

[0041]

[0042] Where O(P) is the value of the O(·) function, K is the number of indices of the O(·) function, and w i is the weight of the ith indicator, O i (P) is the value of the i-th indicator, To update the gradient, c i is the constraint of the i-th indicator.

[0043] Through the above technical solution, the embodiment of the present invention provides a smart energy management system for data centers. The management system collects power efficiency, carbon efficiency and water resource efficiency for analysis, and dynamically adjusts the system parameters of the management area based on the analysis results. Compared with the general energy supply strategy in the prior art, the management system provided by the present invention improves the efficiency of energy utilization by combining the real-time overall status of the management area.

[0044] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0046] Figure 1 is a structural block diagram of a smart energy management system for a data center according to an embodiment of the present invention;

[0047] Figure 2 is a logic block diagram of a unified interface method according to an embodiment of the present invention;

[0048] Figure 3 is a flowchart of a specific method for optimizing an O(·) function according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0050] like Figure 1 FIG. 1 is a block diagram of a smart energy management system for a data center according to an embodiment of the present invention. Figure 1 In the example, the management system may include an integrated collection layer 1, a service support layer 2 and a management application layer 3.

[0051] Among them, the integrated collection layer 1 can be used to collect the power efficiency, carbon efficiency and water efficiency of the management area. The service support layer 2 can be used to analyze the power efficiency, carbon efficiency and water efficiency to determine the overall status of the current management area. The management application layer 3 can be used to adjust the system parameters of the management area according to the overall status feedback.

[0052] In this embodiment, the specific collection method of the power utilization efficiency, carbon utilization efficiency and water resource utilization efficiency can be various forms known to those skilled in the art. In one example of the present invention, the power utilization efficiency can be calculated using the following formula (1):

[0053] PUE = P total / P IT , (1)

[0054] Among them, PUE is the power usage efficiency, P total is the total energy consumption of the data center, P IT Energy consumption of IT equipment.

[0055] The carbon utilization efficiency can be calculated using the following formula (2):

[0056] CUE=E 排放量 / P IT , (2)

[0057] Among them, CUE is carbon utilization efficiency, E 排放量 is the total carbon dioxide emissions from various sources of the data center, P IT Energy consumption of IT equipment.

[0058] The water resource utilization efficiency can be calculated using the following formula (3):

[0059] WUE=L 总耗水 / P IT , (3)

[0060] Among them, L 总耗水 is the annual water consumption of the data center, P IT Energy consumption of IT equipment.

[0061] In this embodiment, the original collected data of power efficiency, carbon efficiency and water resource efficiency are relatively redundant. Therefore, when the integrated collection layer 1 collects power efficiency, carbon efficiency and water resource efficiency, the original collected data can be cleaned, fused and de-noised. At the same time, in order to ensure the number of samples of the original collected data, the integrated collection layer 1 can also be connected to multiple types of data management system platforms at the same time to obtain power efficiency, carbon efficiency and water resource efficiency from the DCIM system, dynamic environment monitoring system, power monitoring system and BA system of each data management system platform. Specifically, the integrated collection layer 1 can be integrated with its collection server or communication protocol to read the currently digitized time series data. In combination with the current status of each data center, an edge collection service is established for subsystems that may not be collected. Realize data upload in seconds or minutes. Data collection must be able to meet the needs of the system and ensure that it does not affect the performance of the business system. Therefore, corresponding policies should be formulated when extracting data, including extraction methods, extraction timing strategies, extraction cycles and other contents. More specifically, the integrated acquisition layer 1 can be a centralized data platform that aggregates data from various data centers, performs data cleaning, fusion, and denoising, and stores the data in a database to form master data, providing a data basis for the data center smart energy management prototype system. In order to complete the data upload work more efficiently, in this implementation, the processed data can be classified and corresponded to the data dedicated interface respectively. The unified interface method is as follows Figure 2 As shown, various types of data are analyzed and sorted, and after data standardization, unified standard data is formed to support the system data.

[0062] The service support layer 2 can be used to analyze the power efficiency, carbon efficiency and water efficiency to determine the overall status of the current management area. Furthermore, in order to facilitate prompting the operation and maintenance personnel, the service support layer 2 can also display the overall status through a display interface, and determine whether the overall status triggers an alarm signal. If it does, the basic information of the overall status (the values ​​of various indicators) can be recorded.

[0063] In an example of the present invention, the service support layer 2 may include a monitoring display module, an alarm analysis module and an indicator evaluation module, thereby jointly forming an efficient and stable service support system.

[0064] The monitoring and display module is mainly responsible for monitoring the real-time integrated data center data and displaying it in a visual way. Through this module, you can intuitively view the system's operating status, resource usage, and potential problems.

[0065] In order to improve the efficiency of data storage and query, in this example, the integrated collected data can be stored in a high-performance time series database. Core structured data such as storage device data, inspection data, and spatial data can be stored in a relational database. To facilitate subsequent staff queries, a customized dashboard can be built when storing data to display key performance indicators. Data is dynamically updated and displayed in the form of charts, reports, etc., and the system status is fully displayed, thereby improving the operation and maintenance efficiency of operation and maintenance personnel.

[0066] The alarm analysis module is responsible for real-time analysis of monitoring data and generates alarm notifications according to preset rules. During actual operation and maintenance, the alarm analysis module can automatically detect abnormal situations and promptly notify relevant personnel to handle them, ensuring that system problems are responded to quickly.

[0067] Specifically, the alarm analysis module can set reasonable alarm thresholds and trigger conditions according to business needs and system performance requirements. For example, the PUE value exceeds 1.5. When the monitoring data triggers the alarm rule, an alarm notification is sent to the relevant responsible person through various methods such as email, SMS, and corporate WeChat to ensure that the alarm information can be communicated and processed in a timely manner. In addition, in order to improve the judgment accuracy of the alarm analysis module, an alarm aggregation method can also be used to merge multiple related alarms into one alarm event to reduce alarm storms and improve alarm processing efficiency.

[0068] The indicator evaluation module is used to comprehensively evaluate the various indicators of the system, reflecting the operating status and performance level of the system in a quantitative way. Through the continuous evaluation of indicators, potential problems of the system can be discovered in time and optimized.

[0069] In this embodiment, the management application layer 3 may further include a joint optimization module. The intelligent joint module may use artificial intelligence and machine learning methods to deeply analyze the system's operating data, automatically propose optimization suggestions and perform optimization operations, thereby further improving the system's joint level and achieving automated optimization. Specifically, the joint optimization module may use a machine learning algorithm to mine and analyze historical monitoring data to discover system performance bottlenecks and potential problems. Based on the data analysis results, specific optimization suggestions are generated. For some optimization operations with lower risks, they can be directly executed through automated scripts or tools to reduce manual intervention and improve optimization efficiency.

[0070] The management application layer 3 can also be used to adjust the system parameters of the management area according to the overall state feedback. The system parameters may include the temperature, humidity and load distribution of the management area. Accordingly, the management application layer can be further used to adjust the temperature, humidity and load distribution according to the overall state feedback. Specifically, in one example of the present invention, the management application layer 1 can obtain the temperature performance value of the data center and determine whether the temperature performance value is greater than a preset temperature threshold. In the case of determining that the temperature performance value is less than or equal to the temperature threshold, the management application layer 3 can determine the optimization value according to formulas (4) to (6),

[0071]

[0072] LF=P IT / P UPS , (5)

[0073] P 制冷 =P IT +P dis / LF, (6)

[0074] in, is the optimal value, P 算力 is the computing power of the data center, P UPS is the UPS power consumption, P IT is the energy consumption of IT equipment in the data center, P dis is the power consumption of the power distribution equipment, and Optimize is an O(·) function.

[0075] In this example, for the O(·) function, the O(·) function can be further optimized according to formula (7) to formula (10), and optimizing the i-th power consumption is equal to minimizing O i (P). However, when optimizing these K (4) power consumptions simultaneously, it is easy for the optimal solution of one power consumption to be the suboptimal solution of another power consumption. Therefore, this method combines multiple power consumptions into one power consumption.

[0076]

[0077] Using the stochastic gradient descent method, a sample is randomly selected at each iteration, the O(P) gradient of the sample to the parameter is calculated, and the parameter is updated. Since only one sample is used for update each time, the calculation speed will be greatly accelerated.

[0078]

[0079] Where O(P) is the value of the O(·) function, K is the number of indices of the O(·) function, and w i is the weight of the ith indicator, O i (P) is the value of the i-th indicator, To update the gradient, c i is the constraint of the i-th indicator. First, ignore the factors of multiple power consumption effects and only solve the single power consumption optimization. Then, introduce multiple power consumption effects to ensure that under the premise of single power consumption optimization, multiple power consumption effects are satisfied as much as possible through appropriate adjustments, and generate an effective solution from the feasible set with all constraints.

[0080] More specifically, in this example, the specific method for optimizing the O(·) function may include: Figure 3 In the steps shown in Figure 3 In the example, the specific method for optimizing the O(·) function may include the following steps:

[0081] In step S1, the parameters, weights w and weight boundaries c are initialized. The weights may be initialized by uniform distribution;

[0082] In step S2, for multiple optimization objectives P, determine the i-th index O i (P);

[0083] In step S3, according to the weight w and each index O i (P) determines the value of the O(·) function O(P).

[0084] In step S4, it is determined whether the value O(P) of the currently calculated O(·) function meets the energy consumption requirement of the data center;

[0085] In step S5, when it is determined that the value O(P) of the currently calculated O(·) function does not meet the energy consumption requirement of the data center, the difference between the value O(P) of the O(·) function of the previous iteration and the value O(P) of the O(·) function of the current iteration is calculated;

[0086] In step S6, it is determined whether the difference is less than or equal to a preset difference threshold;

[0087] In step S7, when it is determined that the difference is greater than the threshold, the weight parameter is updated by the random gradient descent method, the new weight is calculated by the quadratic programming method, and the process returns to step S3. When updating the weight parameter, the same power consumption (index O) can be preferably updated. i (P)) to improve the accuracy of parameter iteration.

[0088] In step S8, when it is determined that the difference is less than or equal to the difference threshold or when it is determined that the value O(P) of the currently calculated O(·) function meets the energy consumption requirement of the data center, the current weight is output as the optimal solution.

[0089] Through the above technical solution, the embodiment of the present invention provides a smart energy management system for data centers. The management system collects power efficiency, carbon efficiency and water resource efficiency for analysis, and dynamically adjusts the system parameters of the management area based on the analysis results. Compared with the general energy supply strategy in the prior art, the management system provided by the present invention improves the efficiency of energy utilization by combining the real-time overall status of the management area.

[0090] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0091] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0092] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0094] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0095] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0096] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules 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, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic 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. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0097] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0098] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A smart energy management system for data centers, characterized in that: The management system comprises: An integrated data collection layer for collecting power efficiency, carbon efficiency, and water efficiency in the management area; A service support layer, used for analyzing the power utilization efficiency, carbon utilization efficiency and water resource utilization efficiency to determine the overall status of the current management area; The management application layer is used to adjust the system parameters of the management area according to the overall status feedback.

2. The management system according to claim 1, characterized in that: The integrated acquisition layer is used for: According to formula (1), the power efficiency of the data center is calculated: PUE=P total / P IT , (1) Wherein, PUE is the power usage efficiency, P total is the total energy consumption of the data center, P IT Energy consumption of IT equipment.

3. The management system according to claim 1, characterized in that: The integrated acquisition layer is used for: The carbon utilization efficiency is calculated according to formula (2): CUE=E 排放量 / P IT , (2) Wherein, CUE is the carbon utilization efficiency, E 排放量 is the total carbon dioxide emissions from various sources of the data center, P IT Energy consumption of IT equipment.

4. The management system according to claim 1, characterized in that: The integrated acquisition layer is used for: According to formula (3), the water resource utilization efficiency of the data center is calculated: WUE=L 总耗水 / P IT , (3) Among them, L 总耗水 is the annual water consumption of the data center, P IT Energy consumption of IT equipment.

5. The management system according to claim 1, characterized in that: The integrated acquisition layer is used for: Performing data cleaning operations, data fusion operations, and data denoising operations on the power utilization efficiency, carbon utilization efficiency, and water resource utilization efficiency; Connect to multiple types of data management system platforms to obtain the power utilization efficiency, carbon utilization efficiency and water resource utilization efficiency from the DCIM system, environmental monitoring system, power monitoring system and BA system of each data management system platform.

6. The management system according to claim 1, characterized in that: The service support layer is used for: Displaying the overall status through a display interface; Determining whether the overall status triggers an alarm signal; In the case where it is determined that the overall state triggers an alarm signal, basic information of the triggered overall state is recorded.

7. The management system according to claim 1, characterized in that: The system parameters include temperature, humidity and load distribution of the management area; The management application layer is used to adjust the temperature, humidity and load distribution according to the overall status feedback.

8. The management system according to claim 1, characterized in that: The management application layer is used to: Get the temperature performance value of the data center; Determining whether the temperature performance value is greater than a preset temperature threshold; When it is determined that the temperature performance value is less than or equal to the temperature threshold, the optimization value is determined according to formulas (4) to (6): LF=P IT / P UPS , (5) P 制冷 =P IT +P dis / LF, (6) in, is the optimized value, P 算力 is the computing power of the data center, P UPS is the UPS power consumption, P IT is the energy consumption of IT equipment in the data center, P dis is the power consumption of the power distribution equipment, and Optimize is an O(·) function.

9. The management system according to claim 8, characterized in that: The management application layer is used to: According to formula (7) to formula (10), the O(·) function is optimized. Where O(P) is the value of the O(·) function, K is the number of indices of the O(·) function, and w i is the weight of the ith indicator, O i (P) is the value of the i-th indicator, To update the gradient, c i is the constraint of the i-th indicator.

Images