Server management method and device, equipment, medium and product
By establishing a power consumption model, the actual power consumption of cloud computer servers can be evaluated in real time, solving the problem of inaccurate installed capacity and achieving efficient server management and energy utilization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-30
- Publication Date
- 2026-04-14
AI Technical Summary
In existing technologies, the power consumption management methods of cloud computer servers lead to inaccurate installed capacity, resulting in resource waste and excessive energy consumption, making it difficult to achieve efficient server management.
By establishing a power consumption model, the actual power consumption of the target server can be evaluated in real time. Combined with the relationship between the business layer and the service layer, the number of servers installed and the allocation of services can be dynamically managed.
It improved the accuracy of server management, optimized the energy efficiency of data centers, and reduced operating costs.
Smart Images
Figure CN121857951A_ABST
Abstract
Description
Technical Field
[0001] This disclosure belongs to the field of computer technology, specifically relating to a server management method, apparatus, device, medium, and product. Background Technology
[0002] Cloud computing, as an emerging computing model, has been widely adopted due to its advantages in resource sharing, flexibility, and cost-effectiveness. With the development of cloud computing technology, data centers are expanding in scale, and the number of cloud computing servers is increasing dramatically, leading to huge power consumption and cooling demands.
[0003] In related technologies, power consumption is typically estimated based on the server manufacturer's stated power rating or by calculating power consumption based on the server's CPU utilization, thus managing the number of servers installed. However, this management method results in a significant discrepancy between the obtained server power consumption figures and the actual power consumption data, leading to inaccurate server installation figures. Therefore, it reduces the accuracy of server management. Summary of the Invention
[0004] This disclosure addresses some of the deficiencies mentioned in the background art by providing a server management method, apparatus, device, medium, and product that can improve the accuracy of server management.
[0005] In a first aspect, embodiments of this disclosure provide a server management method, comprising: Obtain target service load data of the target server; wherein, the target service load data represents the association between the service layer and the business layer corresponding to the target server; Based on a pre-established power consumption model, the target actual power consumption of the target server is determined according to the target service load data; wherein, the pre-established power consumption model represents the mapping relationship between the service layer, server layer and power consumption layer of at least one server, and the pre-established power consumption model is used to quantify the actual power consumption of each server under different service scenarios, and the target server is a server among at least one of the servers. The target server is dynamically managed based on its actual power consumption.
[0006] Optionally, the step of dynamically managing the target server based on the target's actual power consumption includes: Obtain the target nominal power of the target server, whereby the target nominal power represents the rated output power of the target server; The number of units installed in the target server is determined based on the target's actual power consumption and the target's nominal power.
[0007] Optionally, the method further includes: Obtain historical business load data and historical actual power consumption for at least one business scenario; Feature extraction is performed on the historical business load data to obtain the historical server performance characteristics corresponding to each historical business load data, and the historical server performance characteristics represent the resource usage relationship between the business layer and the server layer. Based on the performance characteristics of each historical server and the corresponding historical actual power consumption, the power consumption coefficient corresponding to each historical server performance characteristic is determined. The power consumption model is constructed based on at least one of the historical server performance characteristics and the corresponding power consumption coefficient.
[0008] Optionally, server performance characteristics include at least one of the following: CPU utilization, memory utilization, storage performance consumption, and network consumption.
[0009] Optionally, determining the target actual power consumption of the target server based on the target service load data according to the pre-established power consumption model includes: The target service load data is subjected to feature extraction processing to obtain the target server performance characteristics; Based on the pre-established power consumption model, the target power consumption coefficient is determined according to the performance characteristics of the target server; The actual power consumption of the target server is determined based on the target server performance characteristics and the target power consumption coefficient.
[0010] Optionally, the step of dynamically managing the target server based on the target's actual power consumption includes: For any one of the multiple servers, if the performance characteristics of any one of the first servers exceed the corresponding scheduling threshold, the target service in the first server is reallocated to the second server based on a preset scheduling strategy.
[0011] Optionally, the scheduling strategy is used to determine the target service to be scheduled and the second server by combining the performance characteristics of each server with the actual power consumption of the target.
[0012] Optionally, the method further includes: Obtain historical business load data; The historical business load data is processed to obtain first business load data. The data processing includes at least one of the following: data cleaning and data aggregation. Perform business analysis on the first business load data to obtain at least one business scenario.
[0013] Optionally, the method further includes: Using a pre-trained fault prediction model, the server performance characteristics and actual power consumption of the target server are analyzed and processed to obtain the fault prediction results of the target server.
[0014] Optionally, the method further includes: Using a pre-trained evaluation model, the actual power consumption of all servers in the data center to which the target server belongs is evaluated to obtain the power utilization efficiency of the data center.
[0015] In a second aspect, embodiments of this disclosure provide a server management apparatus, comprising: The acquisition module is used to acquire target service load data of the target server; wherein, the target service load data represents the association between the service layer and the business layer corresponding to the target server; The first determining module is used to determine the target actual power consumption of the target server based on the target service load data according to the pre-established power consumption model; wherein, the pre-established power consumption model represents the mapping relationship between the service layer, server layer and power consumption layer of at least one server, and the pre-established power consumption model is used to quantify the actual power consumption of each of the servers under different service scenarios, and the target server is a server among at least one of the servers; The management module is used to dynamically manage the target server based on the actual power consumption of the target.
[0016] In a third aspect, embodiments of this disclosure provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the server management method described above.
[0017] In a fourth aspect, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the above-described server management method.
[0018] In a fifth aspect, embodiments of this disclosure provide a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device executes the above-described server management method.
[0019] This disclosure involves acquiring target service load data for a target server; this target service load data represents the relationship between the service layer and the business layer corresponding to the target server; based on a pre-established power consumption model, the target actual power consumption of the target server is determined according to the target service load data; the pre-established power consumption model represents the mapping relationship between the business layer, server layer, and power consumption layer corresponding to at least one server, and is used to quantify the actual power consumption of each server under different business scenarios; the target server is one of at least one servers; and the target server is dynamically managed based on the target actual power consumption. The power consumption model allows for real-time evaluation of the target server's actual power consumption, enabling the assessment of the target server's installed capacity and business allocation. In other words, by combining the relationship between the cloud computer's business layer characteristics and service layer power consumption, the target server can be managed more accurately and dynamically. Therefore, the accuracy of server management can be improved.
[0020] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description
[0021] Figure 1 This is a flowchart of a server management method provided in this disclosure.
[0022] Figure 2 A flowchart for power consumption model analysis provided in this disclosure.
[0023] Figure 3 Another flowchart for a server management method provided in this disclosure.
[0024] Figure 4 This is a schematic diagram of the structure of a server management device provided in this disclosure.
[0025] Figure 5 This is a hardware block diagram of an electronic device provided in this disclosure.
[0026] Figure 6 This is a schematic diagram of a computer program product provided in this disclosure. Detailed Implementation
[0027] To enable those skilled in the art to better understand the technical solution of this application, the application scenario of this application will be described first below.
[0028] Cloud computing, as an emerging computing model, has been widely adopted due to its advantages in resource sharing, flexibility, and cost-effectiveness. With the development of cloud computing technology, data centers are expanding in scale, and the number of cloud computing servers is increasing dramatically, leading to huge power consumption and cooling demands. Traditional data center management methods often fail to fully consider the dynamic changes in cloud computing workloads, resulting in resource waste and excessive energy consumption. Traditional data center operation and maintenance typically rely on manual monitoring and management, which is not only inefficient but also makes it difficult to ensure optimal energy utilization. To improve the utilization rate of limited rack resources, data centers need to rationally set the number of servers within each rack. As business volume and types increase, servers need to be added or removed from the racks to adapt to business needs. Therefore, how to rationally rack cloud computing servers and efficiently manage and maintain them has become an urgent problem to be solved.
[0029] In related technologies, power consumption is typically estimated based on the server manufacturer's stated power rating or by calculating power consumption based on the server's CPU utilization, thus managing the number of servers installed. However, this management method often results in significant discrepancies between the obtained server power consumption figures and the actual power consumption data. This can easily lead to situations where the rack installation rate is too high, resulting in excessively high actual power consumption and servers overloading, or the server power consumption is lower than the calculated value, leading to low rack installation rate, wasted rack space, and inefficient power supply utilization. Ultimately, this results in inaccurate server installation figures, reducing the accuracy of server management.
[0030] To address the aforementioned technical problems, this disclosure provides an inventive concept: a power consumption model can be used to assess the actual power consumption of a target server in real time, thereby enabling the evaluation of the target server's installed capacity and service allocation. In other words, by combining the relationship between the characteristics of the cloud computer's service layer and the power consumption of the service layer, the target server can be managed more accurately and dynamically. This achieves efficient power consumption management and operation and maintenance of cloud computer servers in data centers, improving energy efficiency and reducing operating costs through refined management and intelligent control.
[0031] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present disclosure and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the drawings, not the entire structure.
[0032] Figure 1 This is a flowchart illustrating a server management method provided in this disclosure. Figure 1 As shown, the method includes: S101: Obtain the target business load data of the target server.
[0033] Specifically, monitoring can be deployed on both the cloud PC client and server sides to continuously collect target business load data from the target server. This target business load data represents the relationship between the service layer and the business layer corresponding to the target server.
[0034] S102: Based on the pre-established power consumption model, determine the target actual power consumption of the target server according to the target service load data.
[0035] Specifically, the target workload data is input into a pre-established power consumption model. The power consumption model analyzes and processes the target workload data to obtain the corresponding actual power consumption. The pre-established power consumption model represents the mapping relationship between the service layer, server layer, and power consumption layer corresponding to at least one server. This model is used to quantify the actual power consumption of each server under different business scenarios, with the target server being one of the at least one servers. This allows for a more accurate acquisition of the true power consumption of the business system.
[0036] The business layer is the core that drives the entire business system, determining how much computing and storage resources are needed to support the system. The growth or decline of business will directly affect the demand for server resources. The server layer is the infrastructure that supports business operations. Depending on business needs, the number or configuration of servers may need to be increased or decreased. The performance and number of servers directly affect power consumption. Power consumption reflects the energy consumption required for server operation. The number, configuration, and usage of servers will affect power consumption data.
[0037] S103: Dynamically manage the target server based on its actual power consumption.
[0038] Specifically, based on the actual power consumption of the target server, the number of installed servers can be calculated, or services can be redistributed to target servers with unbalanced loads by combining business load data, so as to achieve dynamic management of the target server.
[0039] This disclosure involves acquiring target service load data for a target server; this target service load data represents the relationship between the service layer and the business layer corresponding to the target server; based on a pre-established power consumption model, the target actual power consumption of the target server is determined according to the target service load data; the pre-established power consumption model represents the mapping relationship between the business layer, server layer, and power consumption layer corresponding to at least one server, and is used to quantify the actual power consumption of each server under different business scenarios; the target server is one of at least one servers; and the target server is dynamically managed based on the target actual power consumption. The power consumption model allows for real-time evaluation of the target server's actual power consumption, enabling the assessment of the target server's installed capacity and business allocation. In other words, by combining the relationship between the cloud computer's business layer characteristics and service layer power consumption, the target server can be managed more accurately and dynamically. Therefore, the accuracy of server management can be improved.
[0040] In one possible implementation, an exemplary method for dynamically managing a target server based on its actual power consumption includes: Obtain the target nominal power of the target server; based on the actual power consumption and the target nominal power, determine the number of servers installed in the target server.
[0041] Specifically, the target nominal power represents the rated output power of the target server. The quotient of the target actual power consumption and the target nominal power is calculated, and the result is the number of servers installed. In this embodiment, by combining the relationship between cloud computer service load and server power consumption, the true power consumption of the business system can be obtained more accurately. The racks are then configured based on the true power consumption of the business system, improving rack utilization and solving the problem of accuracy in rack installation.
[0042] In one possible implementation, the method further includes: Obtain historical business load data; process the historical business load data to obtain first business load data; perform business analysis on the first business load data to obtain at least one business scenario.
[0043] Specifically, by deploying load monitoring tools in a cloud computing environment, business load data under different business scenarios is collected, and combined with historical business load data, the fluctuation patterns and characteristics of the business load are analyzed. The collected data is stored in a big data platform, and through data processing and analysis, the time-series characteristics and fluctuation patterns of the business load are obtained. Based on the business fluctuation patterns of the cloud computing environment, the business scenarios of cloud computing can be defined into six types: idle operation, light office work, medium office work, heavy office work, network-intensive, and storage-intensive. In this embodiment, data processing includes at least one of the following: data cleaning and data aggregation.
[0044] In one possible implementation, the power consumption model can be constructed using the following method: Obtain historical business load data and historical actual power consumption for at least one business scenario; extract features from the historical business load data to obtain the historical server performance characteristics corresponding to each historical business load data; determine the power consumption coefficient corresponding to each historical server performance characteristic based on each historical server performance characteristic and the corresponding historical actual power consumption; construct a power consumption model based on at least one historical server performance characteristic and the corresponding power consumption coefficient.
[0045] Specifically, six business load scenarios for cloud computers were simulated: idle operation, light office work, medium office work, heavy office work, network-intensive, and storage-intensive. Performance monitoring tools and power consumption monitoring devices were installed on the servers and business systems to collect historical actual power consumption and historical business load data.
[0046] Based on measurement data, a power consumption model is established using regression analysis to describe the relationship between power consumption and business load data, quantifying the server's power consumption under different loads. By analyzing the cloud PC's business load data and the server's actual power consumption, a regression equation is established. This allows for more accurate acquisition of specific business power consumption data for the cloud PC. Figure 2 The power consumption model analysis flowchart provided in this disclosure is as follows: Figure 2 As shown, historical business load data and historical actual power consumption of servers under different business scenarios are collected, and the collected data is standardized to ensure data cleanliness. Feature extraction is performed on the historical business load data to obtain the historical server performance characteristics corresponding to each historical business load data. Among them, the historical server performance characteristics represent the resource usage relationship between the business layer and the server layer.
[0047] Import relevant libraries for regression analysis, assign initial values to historical server performance characteristics and historical actual power consumption samples, create a linear regression model, calibrate the model, and determine the regression coefficients, i.e., power coefficients. In a real-world operating environment, calibrate the power consumption model using real-time monitoring data to improve its accuracy and applicability. Finally, determine the regression equation, i.e., the power consumption model, for subsequent analysis of target actual power consumption. For example, server performance characteristics include at least one of the following: CPU utilization, memory utilization, storage performance consumption, and network consumption.
[0048] An example formula for the regression equation result is shown below: The polynomial regression equation for the server power consumption model is:
[0049] Where Power represents the historical actual power consumption. For CPU utilization, For memory usage, For storage performance consumption, For network consumption, The power factor of the CPU. The power factor for memory. The power factor for storage performance. The power factor of the network consumption. This is the error value.
[0050] For example, in this embodiment, It is determined to be 50. It is determined to be 10. It is determined to be 20. The value is set at 10, meaning the server's base power consumption is 300 watts. For every 1% increase in CPU load, power consumption increases by 50 watts; for every 1% increase in memory load, power consumption increases by 10 watts; for every 1% increase in storage load, power consumption increases by 20 watts; and for every 1% increase in network load, power consumption increases by 10 watts.
[0051] For example, Table 1 is a table showing the relationship between the actual power consumption of the cloud computer target obtained by combining actual business operations, as shown in Table 1.
[0052] Table 1 Actual Power Consumption of the Target
[0053] Based on the above business load data and actual server power consumption, a regression equation is constructed to establish a power consumption model for the cloud computer business layer, server layer, and power consumption layer. The actual power consumption of the cloud computer server is used as the basis for configuring the server rack, which greatly improves the accuracy of the installation.
[0054] In one possible implementation, an exemplary method for determining the target actual power consumption of a target server based on target service load data, using a pre-established power consumption model, includes: Feature extraction processing is performed on the target business load data to obtain the target server performance characteristics; based on the pre-established power consumption model, the target power consumption coefficient is determined according to the target server performance characteristics; based on the target server performance characteristics and the target power consumption coefficient, the target actual power consumption is determined.
[0055] Specifically, based on the target business load data, the corresponding server performance characteristics can be extracted. These characteristics include at least one of the following: CPU utilization, memory utilization, storage performance consumption, and network consumption. By inputting these performance characteristics into the power consumption model and combining them with the corresponding target power consumption coefficients, the actual target power consumption can be calculated and output.
[0056] In one possible implementation, an exemplary method for dynamically managing a target server based on its actual power consumption includes: For any one of the multiple servers, if the performance characteristics of any server on the first server exceed the corresponding scheduling threshold, the target service on the first server will be reallocated to the second server based on the preset scheduling strategy.
[0057] Specifically, based on server performance characteristics and actual target power consumption, an intelligent scheduling strategy is adopted. This strategy monitors various server utilization rates (CPU, memory, network, storage, power consumption) and sets thresholds. When a resource exceeds the set threshold, the intelligent scheduling mechanism is triggered to reallocate tasks, achieving load balancing and optimizing energy consumption. In other words, the scheduling strategy combines the performance characteristics of each server with the actual target power consumption to determine the target service to be scheduled and the second server. Based on server performance characteristics and the scheduling strategy, services on the first server can be rationally distributed across different servers, avoiding single-point overload on the first server.
[0058] Deploy the Zabbix open-source monitoring tool on the servers to monitor the business load data of each server in real time and analyze the server performance characteristics. Set usage thresholds for server performance characteristics. When the performance characteristics of a server exceed the threshold, trigger an intelligent scheduling strategy to redistribute the load. Combining server performance characteristics and target power consumption data, perform virtual machine migration execution, reallocating some requests to other idle servers to balance the load and dynamically adjust task allocation.
[0059] For example, the scheduling policy includes at least one of the following: (1) CPU-intensive tasks: assigned to servers with low CPU utilization.
[0060] (2) Memory-intensive tasks: assigned to servers with sufficient memory.
[0061] (3) Storage-intensive tasks: Assign them to servers with large storage capacity and good I / O performance.
[0062] (4) Network-intensive tasks: assigned to servers with sufficient network bandwidth.
[0063] (5) Power scheduling: Reduce the frequency of services on high-power servers or migrate them to servers with lower power consumption.
[0064] Furthermore, the scheduling strategy can be a dynamic load scheduling algorithm, which optimizes the allocation of server resources based on real-time load prediction results using dynamic scheduling algorithms (such as genetic algorithms, simulated annealing algorithms, etc.).
[0065] For example, suppose there is a cloud computer server A (the first server) whose CPU utilization exceeds a threshold. According to the intelligent scheduling strategy, the scheduling algorithm is triggered to find a server B (the second server) with lower CPU utilization. Some CPU-intensive tasks are migrated from server A to server B. At the same time, the usage of other resources (memory, network, storage) is checked to ensure that the migration will not cause new bottlenecks. If the power consumption of server A is too high, some tasks are migrated to the server with lower power consumption to achieve energy efficiency optimization.
[0066] In one possible implementation, the method further includes: By using a pre-trained fault prediction model, the server performance characteristics and actual power consumption of the target server are analyzed and processed to obtain the fault prediction results of the target server.
[0067] Specifically, the fault prediction model can be trained using supervised or unsupervised learning techniques to analyze and process the server performance characteristics and actual power consumption of the target server, thereby predicting server faults. In this embodiment, the fault prediction model can use algorithms such as random forest and support vector machine.
[0068] In one possible implementation, the method further includes: Using a pre-trained evaluation model, the actual power consumption of all servers in the data center to which the target server belongs is evaluated to obtain the power utilization efficiency of the data center.
[0069] Specifically, an evaluation model can be established to assess the energy consumption of server rooms. In this embodiment, the evaluation model can use a regression model or a neural network to predict the power usage efficiency of the server room. Furthermore, server room energy can be optimized, for example, by designing energy optimization algorithms to dynamically adjust energy procurement and usage strategies based on predicted energy consumption and market energy prices.
[0070] Figure 3 This is another flowchart illustrating a server management method provided in this disclosure. Figure 3 As shown, this method is a way to establish intelligent server management, specifically including: S301: Define cloud computer business scenarios.
[0071] Specifically, by deploying load monitoring tools in the cloud computing environment, business load data under different business scenarios is collected, and combined with historical business load data, the fluctuation patterns and characteristics of business load are analyzed. The collected data is stored in a big data platform, and through data processing and analysis, the time-series characteristics and fluctuation patterns of business load are obtained, thereby determining the cloud computing business scenarios.
[0072] S302: Monitors business load data and actual server power consumption data.
[0073] Specifically, by deploying load monitoring tools in a cloud computing environment, business load data and actual server power consumption data under different business scenarios can be collected.
[0074] S303: Input business load data into the controller.
[0075] S304: Controller Design.
[0076] Specifically, based on real-time measured power consumption data and business load data, a power consumption model predictive controller is designed to regulate the system's operation.
[0077] S305: Power consumption regulation control.
[0078] Specifically, the controller calculates adjustment signals to regulate the system's power consumption. The controller saves power by adjusting the system's load distribution or reducing the operating frequency or voltage of certain components.
[0079] S306: Feedback closed-loop control.
[0080] Specifically, the adjusted power consumption is fed back to the control system for comparison. Based on the difference between the actual and expected power consumption, the controller output, i.e., the power consumption coefficient of the power consumption model, is adjusted to maintain the system in its optimal power consumption state under business requirements. Through the feedback control loop, the operating status of the data center is continuously optimized and adjusted to achieve the goals of energy saving and performance optimization.
[0081] S307: Monitoring and Feedback.
[0082] Specifically, a monitoring system is deployed to monitor power consumption and business parameters in real time, feeding this information back to the control system for timely problem detection and response, and continuous improvement and optimization. Through machine learning and optimization algorithms, intelligent scheduling and optimized control can be implemented for different business scenarios, achieving refined management and intelligent control of the data center. This solves the efficiency problems of power consumption management and operation and maintenance of cloud PC servers.
[0083] Figure 4 This is a schematic diagram of the structure of a server management device provided in this disclosure. Figure 4 As shown, the device 400 includes: a first acquisition module 410, a first determination module 420, and a management module 430.
[0084] The first acquisition module 410 is used to acquire target service load data of the target server; wherein, the target service load data represents the association relationship between the service layer and the business layer corresponding to the target server; The first determining module 420 is used to determine the target actual power consumption of the target server based on the target service load data according to the pre-established power consumption model; wherein, the pre-established power consumption model represents the mapping relationship between the service layer, server layer and power consumption layer of at least one server, and the pre-established power consumption model is used to quantify the actual power consumption of each server in different service scenarios, and the target server is a server among at least one of the servers. The management module 430 is used to dynamically manage the target server based on the actual power consumption of the target.
[0085] Optionally, the management module includes: The acquisition submodule is used to acquire the target nominal power of the target server, wherein the target nominal power represents the rated output power of the target server; The first determining submodule is used to determine the number of units installed in the target server based on the target's actual power consumption and the target's nominal power.
[0086] Optionally, the device further includes: The second acquisition module is used to acquire historical business load data and historical actual power consumption in at least one business scenario. The extraction module is used to extract features from the historical business load data to obtain the historical server performance features corresponding to each historical business load data. The historical server performance features characterize the resource usage relationship between the business layer and the server layer. The second determining module is used to determine the power consumption coefficient corresponding to each historical server performance characteristic based on the historical server performance characteristics and the corresponding historical actual power consumption. A building module is used to build the power consumption model based on at least one of the historical server performance characteristics and the corresponding power consumption coefficient.
[0087] Optionally, the first determining module includes: An extraction submodule is used to perform feature extraction processing on the target business load data to obtain the performance characteristics of the target server. The second determining submodule is used to determine the target power consumption coefficient based on the pre-established power consumption model and the performance characteristics of the target server. The third determining submodule is used to determine the target actual power consumption based on the target server performance characteristics and the target power consumption coefficient.
[0088] Optionally, the management module includes: The allocation submodule is used to reallocate the target service in the first server to the second server based on a preset scheduling strategy when the performance characteristics of any one of the first servers exceed the corresponding scheduling threshold.
[0089] Optionally, the device further includes: The third acquisition module is used to acquire historical business load data; The processing module is used to process the historical business load data to obtain the first business load data. The data processing includes at least one of the following: data cleaning and data aggregation. The first analysis module is used to perform business analysis on the first business load data to obtain at least one business scenario.
[0090] Optionally, the device further includes: The second analysis module is used to analyze and process the server performance characteristics and actual power consumption of the target server using a pre-trained fault prediction model, and obtain the fault prediction result of the target server.
[0091] Optionally, the device further includes: The evaluation module is used to evaluate the actual power consumption of all servers in the data center to which the target server belongs using a pre-trained evaluation model, so as to obtain the power usage efficiency of the data center.
[0092] This application also provides an electronic device for performing the above-described server management method. Please refer to... Figure 5 It illustrates a schematic diagram of an electronic device provided by some embodiments of this application. For example... Figure 5 As shown, the electronic device 50 includes: a processor 500, a memory 501, a bus 502, and a communication interface 503. The processor 500, the communication interface 503, and the memory 501 are connected via the bus 502. The memory 501 stores a computer program that can run on the processor 500. When the processor 500 runs the computer program, it executes the server management method provided in any of the foregoing embodiments of this application.
[0093] The memory 501 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this device network element and at least one other network element is achieved through at least one communication interface 503 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.
[0094] Bus 502 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 501 is used to store programs. After receiving an execution instruction, the processor 500 executes the program. The server management method disclosed in any of the foregoing embodiments of this application can be applied to the processor 500, or implemented by the processor 500.
[0095] The processor 500 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 500 or by instructions in software form. The processor 500 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 501. The processor 500 reads the information in memory 501 and, in conjunction with its hardware, completes the steps of the above method.
[0096] The electronic device provided in this application embodiment and the server management method provided in this application embodiment are based on the same inventive concept and have the same beneficial effects as the methods they adopt, operate or implement.
[0097] This application also provides a computer-readable storage medium corresponding to the server management method provided in the foregoing embodiments. The computer-readable storage medium shown may be an optical disc, on which a computer program is stored. When the computer program is run by a processor, it executes the server management method provided in any of the foregoing embodiments.
[0098] It should be noted that examples of the computer-readable storage medium may also 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 optical and magnetic storage media, which will not be elaborated here.
[0099] The computer-readable storage medium provided in the above embodiments of this application and the server management method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.
[0100] This application also provides a computer program product 600, such as... Figure 6 As shown. This computer program product carries a computer program 601. The instructions included in the program code can be used to execute the steps of the server management method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here.
[0101] The aforementioned computer program product can be implemented through hardware, software, or a combination thereof. In one optional embodiment, the computer program product is specifically embodied in a computer storage medium; in another optional embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.
[0102] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.
[0103] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.
[0104] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.
[0105] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.
[0106] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.
[0107] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.
[0108] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.
Claims
1. A server management method, characterized in that, include: Obtain target service load data of the target server; wherein, the target service load data represents the association between the service layer and the business layer corresponding to the target server; Based on a pre-established power consumption model, the target actual power consumption of the target server is determined according to the target service load data; wherein, the pre-established power consumption model represents the mapping relationship between the service layer, server layer and power consumption layer of at least one server, and the pre-established power consumption model is used to quantify the actual power consumption of each server under different service scenarios, and the target server is a server among at least one of the servers. The target server is dynamically managed based on its actual power consumption.
2. The method according to claim 1, characterized in that, The dynamic management of the target server based on the actual power consumption of the target server includes: Obtain the target nominal power of the target server, whereby the target nominal power represents the rated output power of the target server; The number of units installed in the target server is determined based on the target's actual power consumption and the target's nominal power.
3. The method according to claim 1, characterized in that, The method further includes: Obtain historical business load data and historical actual power consumption for at least one business scenario; Feature extraction is performed on the historical business load data to obtain the historical server performance characteristics corresponding to each historical business load data, and the historical server performance characteristics represent the resource usage relationship between the business layer and the server layer. Based on the performance characteristics of each historical server and the corresponding historical actual power consumption, the power consumption coefficient corresponding to each historical server performance characteristic is determined. The power consumption model is constructed based on at least one of the historical server performance characteristics and the corresponding power consumption coefficient.
4. The method according to claim 3, characterized in that, Server performance characteristics include at least one of the following: CPU utilization, memory utilization, storage performance consumption, and network consumption.
5. The method according to claim 3, characterized in that, The determination of the target actual power consumption of the target server based on the pre-established power consumption model and the target service load data includes: The target service load data is subjected to feature extraction processing to obtain the target server performance characteristics; Based on the pre-established power consumption model, the target power consumption coefficient is determined according to the performance characteristics of the target server; The actual power consumption of the target server is determined based on the target server performance characteristics and the target power consumption coefficient.
6. The method according to claim 1, characterized in that, The dynamic management of the target server based on the actual power consumption of the target server includes: For any one of the multiple servers, if the performance characteristics of any one of the first servers exceed the corresponding scheduling threshold, the target service in the first server is reallocated to the second server based on a preset scheduling strategy.
7. The method according to claim 6, characterized in that, The scheduling strategy is used to determine the target service to be scheduled and the second server by combining the performance characteristics of each server and the actual power consumption of the target.
8. The method according to claim 1, characterized in that, The method further includes: Obtain historical business load data; The historical business load data is processed to obtain first business load data. The data processing includes at least one of the following: data cleaning and data aggregation. Perform business analysis on the first business load data to obtain at least one business scenario.
9. The method according to claim 1, characterized in that, The method further includes: Using a pre-trained fault prediction model, the server performance characteristics and actual power consumption of the target server are analyzed and processed to obtain the fault prediction results of the target server.
10. The method according to claim 1, characterized in that, The method further includes: Using a pre-trained evaluation model, the actual power consumption of all servers in the data center to which the target server belongs is evaluated to obtain the power utilization efficiency of the data center.