Methods, devices, equipment, media, and products for assessing rack server power consumption.

Abstract

This paper provides a method, apparatus, device, medium, and product for evaluating the power consumption of a rack server, relating to the field of computer technology. The method includes: receiving a first operation, the first operation being at least used to instruct the acquisition of the power consumption of the rack server; obtaining a first power consumption of a first component based on a first power consumption evaluation model of a first component, the first component being a component in the rack server, the first power consumption evaluation model including at least a first coefficient used to adjust the power consumption data of the first component; and obtaining a second power consumption of the rack server based on the first power consumption. This method solves the problems of high cost or inaccurate power consumption evaluation results for evaluating the overall power consumption of rack servers. By simulating the actual power consumption of the rack server, it achieves accurate evaluation of the overall power consumption of the rack server while reducing the difficulty of power consumption evaluation.

Description

Technical Field

[0001] This invention relates to the field of computer processing technology, and in particular to a method, apparatus, equipment, medium, and product for evaluating the power consumption of a rack server. Background Technology

[0002] Rack-mount servers integrate a massive number of components, including computing, networking, power supply and distribution, and liquid cooling. They have complex topologies and extremely high integration. Therefore, accurately assessing the power consumption required for rack-mount servers during operation is crucial to ensuring stable system operation. Summary of the Invention

[0003] This invention provides a method, apparatus, equipment, medium, and product for evaluating the power consumption of rack servers, achieving the effect of accurately evaluating the overall power consumption of rack servers.

[0004] In one scenario, this paper provides a method for evaluating the power consumption of a rack server, the method comprising: Receive a first operation, the first operation being at least used to indicate the acquisition of the power consumption of the rack server; Based on the first power consumption evaluation model of the first component, the first power consumption of the first component is obtained. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. Based on the first power consumption, the second power consumption of the rack server is obtained.

[0005] In one scenario, this paper also provides an evaluation device for rack server power consumption, the device comprising: A receiving module is configured to receive a first operation, wherein the first operation is at least used to indicate the power consumption of the rack server; The first module is used to obtain the first power consumption of the first component based on the first power consumption evaluation model of the first component, wherein the first component is a component in the rack server, and the first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. The second module is used to obtain the second power consumption of the rack server based on the first power consumption.

[0006] In one instance, this document also provides an electronic device comprising: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the rack server power consumption assessment method as described in any of these documents.

[0007] In one instance, this document also provides a storage medium containing computer-executable instructions that, when executed by a computer processor, are used to perform a method for evaluating the power consumption of a rack server as described herein.

[0008] In another scenario, this document also provides a computer program product, including a computer program that, when executed by a processor, implements a method for evaluating the power consumption of a rack server as described in any of the embodiments herein.

[0009] The aforementioned method for evaluating the power consumption of a rack server, when evaluating the power consumption of the rack server based on the execution of the first operation, obtains the first power consumption corresponding to the first component through a first power consumption evaluation model. This first power consumption is the power consumption adjusted by a first coefficient in the first power consumption evaluation model to accurately reflect the true power consumption of the first component, including power loss. Based on this, the overall power consumption of the rack server is obtained from the first power consumption of each first component. This achieves accurate evaluation of the true and accurate overall power consumption of the rack server without requiring actual testing of the entire rack server, and reduces the difficulty of power consumption evaluation. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the embodiments described herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0011] Figure 1 This is a schematic diagram of a system for evaluating the power consumption of a rack server under one specific scenario. Figure 2 This is a flowchart illustrating a method for evaluating the power consumption of a rack server under one specific scenario. Figure 3 A flowchart illustrating a method for evaluating the power consumption of rack servers in another scenario; Figure 4 A flowchart illustrating a specific example of a method for evaluating the power consumption of rack servers in another scenario; Figure 5 A flowchart illustrating a method for evaluating the power consumption of rack servers in another scenario; Figure 6 A flowchart illustrating a specific example of a method for evaluating the power consumption of rack servers in another scenario; Figure 7 A flowchart illustrating an alternative example of a method for evaluating the power consumption of rack servers in another scenario; Figure 8 This is a schematic diagram of a device for evaluating the power consumption of a rack server under one specific scenario. Figure 9 This is a schematic diagram of the structure of an electronic device used in one scenario. Detailed Implementation

[0012] The embodiments will now be described in more detail with reference to the accompanying drawings. While some embodiments are shown in the drawings, it should be understood that the technical solutions can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the technical solutions herein. It should be understood that the illustrated drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the technical solutions.

[0013] It should be understood that the steps described in the method implementation may be performed in different orders and / or in parallel. Furthermore, the method implementation may include additional steps and / or omit the steps shown. The scope of this document is not limited in this respect.

[0014] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one situation" means "at least one situation"; the term "another situation" means "at least one additional situation"; the term "some situations" means "at least some situations". Definitions of other terms will be given in the following description.

[0015] It should be noted that the concepts of "first" and "second" mentioned are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions performed by these devices, modules or units or their interdependencies.

[0016] It should be noted that the terms "one" and "more" used in this document are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0017] The names of messages or information exchanged between multiple devices in this document are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0018] It is understood that before using the technical solutions disclosed in the various embodiments of this document, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this document in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0019] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as electronic devices, applications, servers, or storage media, that perform the operations described herein, based on the prompt message.

[0020] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0021] It is understood that the above notification and user authorization process is merely illustrative and does not limit the implementation method described in this article. Other methods that comply with relevant laws and regulations may also be applied to the implementation method described in this article.

[0022] It is understood that the data involved in the technical solutions in this article (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0023] Before introducing the method for evaluating rack server power consumption, we can first illustrate the application scenario. This approach can be applied to any scenario where rack server power consumption calculation and evaluation are required. Optionally, this method can be used to complete the power consumption evaluation of rack servers when initially assessing their overall power consumption. Alternatively, in data center energy management scenarios, when it is necessary to calculate the overall power load of rack servers or match power distribution schemes, this method can be used to complete the rack server power consumption evaluation.

[0024] Before formally explaining the evaluation method for rack server power consumption, a brief description of the rack server architecture will be provided to lay the architectural foundation for subsequent power consumption calculations. The rack server can be abstracted as a multi-level rack system. For example, a rack server can be abstracted as a four-layer hardware hierarchy, which can be divided into device level, board level, node level, and rack level from bottom to top. The device level is the smallest unit of power consumption. For example, devices at the device level can be composed of various electronic components such as a central processing unit (CPU), memory, network interface card (NIC), fan, or chips. Multiple components are integrated and soldered to form a functional board, constituting the board level. Based on this, several functional boards are assembled into independent computing units, which is the node level. For example, nodes in a rack server can be different types of nodes such as computing nodes, storage nodes, switching nodes, or network nodes. Furthermore, multiple nodes, along with auxiliary components such as rack power distribution links, heat dissipation components, or management units, collectively form the rack level. Subsequent power consumption calculations can be based on this hardware hierarchy, accumulating power consumption from the bottom-level devices to calculate the total power consumption of devices, boards, nodes, and the entire cabinet.

[0025] In some cases, the provided solution can be applied to Figure 1 The system shown is a power consumption assessment system for rack servers. This system may include a client 101 and a server 102. The client 101 may include, but is not limited to, web applications such as browsers, applications (Apps), HyperText Markup Language (HTML) applications, lightweight applications (also known as applets), or cloud applications. The client 101 may be deployed on an electronic device and relies on the operation of that device or certain applications within the device to perform its functions. The electronic device may be, for example, a device with a display screen that supports information browsing, such as a smartphone, tablet, personal computer, or other client terminal. For ease of understanding, Figure 1 The client is primarily represented in the form of a device. Other types of applications can also be configured on the electronic device, such as media content publishing applications, session applications, etc. Server 102 can be one or more servers providing various services. That is, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server; furthermore, it can be a server for a distributed system, a server integrating blockchain technology, a cloud server, or an intelligent cloud computing server or intelligent cloud host deployed with machine learning models, etc.

[0026] The rack server power consumption assessment method described in this paper allows interaction between client 101 and server 102, such as receiving or sending messages. For example, server 102 can receive a first operation sent by client 101 based on an information carrier. This first operation at least indicates the acquisition of the rack server's power consumption. Based on a first power consumption assessment model for a first component, server 102 obtains a first power consumption of the first component, where the first power consumption assessment model includes at least a first coefficient used to adjust the power consumption data of the first component. Based on the first power consumption, server 102 obtains a second power consumption of the rack server. Server 102 then sends this first operation (which indicates the acquisition of the rack server's power consumption, the first power consumption of the first component based on the first power consumption assessment model including at least a first coefficient used to adjust the power consumption data of the first component, and the second power consumption of the rack server) to client 101 for display on the client's interface.

[0027] It should be noted that the rack server power consumption assessment method can be executed by client 101, or by client 101 and server 102. The different functional parts of the corresponding rack server power consumption assessment device are deployed on client 101 and server 102 respectively. Specifically, the receiving module of the device is deployed on client 101, and the first and second backend modules are deployed on server 102. Client 101 and server 102 achieve data interaction and functional collaboration through network communication. It should be understood that... Figure 1 The number of clients and servers shown is for illustrative purposes only. Any number of clients and servers can be configured to meet specific implementation requirements.

[0028] Figure 2 This is a flowchart illustrating a method for evaluating the power consumption of a rack server in one scenario. This method is applicable to scenarios requiring power consumption evaluation of any hardware device, particularly for evaluating the power consumption of entire hardware systems, including multi-level structures, such as rack servers. This rack server power consumption evaluation method can be executed by a rack server power consumption evaluation device, which can be implemented in software and / or hardware, optionally through an electronic device such as a mobile terminal, PC, or server. Figure 2 As shown, the specific methods for evaluating the power consumption of the rack server may include: S210, Receive a first operation, the first operation being at least used to indicate the power consumption of the rack server.

[0029] The first operation can be understood as an instruction action used to initiate a power consumption assessment request. The first operation instructs a power consumption assessment to be performed on the rack server to obtain the overall power consumption of the rack server. The first operation represents the intent to perform a power consumption assessment on the rack server. The first operation can be triggered through a first interface. This first interface can be an interface through which the user interacts with the energy management system of the rack server. Optionally, the first interface can include at least one of the following: the main interface of an application (e.g., a session interface), a dialog window on a web page, and a touchscreen of a smart device. The information entered by the user on the first interface includes, but is not limited to, the ambient temperature and / or operating conditions corresponding to the rack server. For example, the operating conditions can be idle conditions, single-component extreme conditions, mixed stress test conditions, and representative business conditions. When the operating conditions and ambient temperature of the rack server differ, the first power consumption model of its corresponding first component will adaptively adjust the power consumption error according to the operating conditions and ambient temperature of the rack server.

[0030] A rack server can be understood as a computing power-bearing system with a multi-layered hardware architecture. A rack server can house various computing, storage, and network hardware units and centrally manage power supply and heat dissipation. A rack server can be divided into multiple hardware layers; for example, its architecture can be divided into a four-layer hardware structure. This four-layer structure serves as the basis for calculating the power consumption of each layer of the rack server. Optionally, the four-layer hardware structure, from bottom to top, can include the device level, board level, node level, and rack level.

[0031] In some scenarios, when it is necessary to evaluate the overall power consumption of the rack server, the user can initiate a power consumption evaluation request for the rack server by entering a first operation on the first interface or triggering the operation control corresponding to the first operation on the first interface. When the energy management center receives the first operation, it will perform an overall power consumption evaluation of the rack server.

[0032] S220. Based on the first power consumption evaluation model of the first component, obtain the first power consumption of the first component. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component.

[0033] It should be noted that in the process of evaluating the power consumption of rack servers, it is difficult to cover all operating conditions and hierarchical scenarios through actual hardware testing. Therefore, this paper uses simulation modeling to quantify and extrapolate the power consumption of hardware units in rack servers.

[0034] In this paper, the rack server can be divided into a four-layer hardware architecture: device level, board level, node level, and rack level. The first component can be understood as the basic hardware statistical unit used for power consumption assessment of the rack server. This basic hardware statistical unit provides the basic hardware parameters and operating conditions required in the power consumption simulation assessment process. The first component can be dynamically defined according to the hierarchical structure of the rack server. For example, based on the statistical accuracy requirements, any one of the four entities—device, board, node, or rack—can be selected as the first component.

[0035] The first power consumption assessment model is used to characterize the digital simulation model constructed for an independent first component. The first power consumption assessment model can be used to accurately reproduce the power consumption variation law of the first component, enabling accurate assessment of the actual power consumption of the first component under conditions without actual measurement. Optionally, the first power consumption assessment model can be at least one of a device-level simulation model, a single-board-level simulation model, a node-level simulation model, and a rack-level simulation model.

[0036] The first power consumption is used to reflect the actual power consumption of the first component under different operating conditions. The first power consumption is the actual power consumption after power consumption correction based on the simulation evaluation power consumption of the first component using the first coefficient.

[0037] The first coefficient can be understood as a power adjustment coefficient configured in the first power consumption assessment simulation model. The first coefficient is used to calibrate the power consumption deviation between the simulated power consumption of the first component and the actual measured power consumption. The first coefficient can be used to improve the reliability of the overall power consumption assessment results for the rack server. The first coefficient can be dynamically adjusted according to the individual differences of the first component, the service load status, and data center environmental parameters (such as ambient temperature). Optionally, the first coefficient is related to factors such as the operating conditions, load conditions, ambient temperature, component topology, and executed services of the first component.

[0038] In practical applications, based on the required statistical accuracy, any one of the four types of entities—devices, boards, nodes, or racks—is selected as the first component. Once an entity is determined as the first component, the power consumption assessment process for that first component does not involve further fine-grained hierarchical breakdown within the entity itself. For example, if the first component is a board, then the board is used as the smallest hardware statistical unit to assess the power consumption of the rack server.

[0039] Based on this, a first power consumption evaluation model is constructed for the first component. In some cases, the first power consumption evaluation model can be obtained in the following way: obtain the third power consumption and the fourth power consumption of the first component in the first state, where the third power consumption is used to characterize the operating power consumption of the first component, and the fourth power consumption is used to characterize the sum of the operating power consumption and the power loss of the first component; based on the third power consumption and the fourth power consumption, obtain the first coefficient of the first component; based on the first coefficient, obtain the first power consumption evaluation model of the first component, where the independent variables in the first power consumption evaluation model are related to the power consumption read by the first software.

[0040] The first state is used to characterize the operating condition of the first component. In this paper, the first state is the operating state of the first component during power consumption simulation evaluation. The first state encompasses comprehensive operating parameters such as load, frequency, power consumption, and environment during the operation of the first component. Optionally, the first state may include at least one of the following: no-load operating state, rated load operating state, peak load operating state, and dynamic load operating state.

[0041] The third power consumption corresponds to the operating power consumption of the first component itself, but does not include various power loss information generated during the operation of the first component. The third power consumption can be understood as the power consumption of the first component in its first state, as read by the first software. The third power consumption is used to characterize the power consumption data of the first component obtained based on the first software. The first software can be understood as a software program that reads the power consumption of the first component. The first software can be deployed in the first power consumption evaluation model and establishes a communication connection with the first component through a software interface to obtain the first power consumption of the first component.

[0042] The fourth power consumption characterizes the actual operating power consumption of the first component in the first state. This actual operating power consumption includes the operating power consumption generated during the operation of the first component, as well as the power loss generated during operation. The fourth power consumption can be understood as the power consumption data obtained by directly acquiring power consumption data of the first component through hardware devices using a direct hardware reading method. The fourth power consumption can be used as a calibration basis for power consumption calibration of the third power consumption obtained by the first software, and is used to correct the power consumption deviation of the third power consumption. In this paper, the hardware-side calibration of the first component is preferentially performed by synchronously measuring the power consumption of the first component using an external power analyzer, oscilloscope, voltage and current sampling module, DC power supply, multimeter, or other precision instruments.

[0043] It is understandable that the power consumption of the first component in a rack server varies under different ambient temperatures and / or operating conditions. Accordingly, the first coefficient corresponding to the first component should be adaptively adjusted according to the changes in the ambient temperature and / or operating conditions of the rack server.

[0044] Optionally, the third power consumption is obtained by the processor from the first component through the first hardware interface, and the processor communicates with the first component through the first hardware interface. The fourth power consumption is obtained by the first power consumption measuring device.

[0045] Preferably, the first software may include low-level drivers, firmware interface programs, current monitoring chips, accelerator management interfaces, other hardware monitoring interfaces, or third-party power consumption analysis tools. For example, the first component communicates with the processor through a first hardware interface. Sensors may be deployed in the processor to acquire signals from the first component. Specifically, during the operation of the first component, analog signals, such as current or voltage signals, are acquired from the sensors and converted into digital signals. These digital signals are then converted into a third power consumption corresponding to the first component, and the third power consumption is displayed through the first interface of the first software via the first hardware interface.

[0046] The first power consumption measurement device may include an external power analyzer, oscilloscope, voltage and current sampling module, DC power supply, multimeter, or any device that can directly perform actual measurements on the first component. The first power consumption measurement device may be hardwired to the first component or deployed around the first component to collect relevant information about the first component. For example, if the first power consumption measurement device is an oscilloscope, it can be physically connected to the first component via a hardwire. If the first power consumption measurement device is a temperature measurement device, it can be either physically connected to the first component via a hardwire or placed around the first component. Specifically, during the power consumption testing phase of the first component, the software-side reading value (i.e., the third power consumption) and hardware-side reading value (i.e., the fourth power consumption) of the first component are simultaneously read under different ambient temperatures and / or operating conditions. The third power consumption and the fourth power consumption are then fitted to obtain the first coefficient corresponding to the first component under the corresponding ambient temperature and / or operating conditions. When adjusting the power consumption of the first component based on the first power consumption evaluation model, the third power consumption of the first component is adjusted by setting the first coefficient under the corresponding ambient temperature and / or operating conditions so that the third power consumption is close to the fourth power consumption, thus obtaining more realistic power consumption data for the first component.

[0047] For example, the actual power consumption (i.e., the measured power consumption in hardware) of each first component can be obtained in different ways. In this paper, for the first component of the device layer, the actual power consumption of each first component is preferably obtained according to the principle of "prioritizing direct measurement and then using the specification value as a fallback for stable reading".

[0048] For example, the actual power consumption of an accelerator can be the power consumption after correcting for the actual operating power consumption of the processor by incorporating voltage conversion losses or voltage conversion efficiency. For coprocessors such as accelerators, the actual power consumption of the first component can be obtained primarily through hardware testing. If the nominal power consumption of the coprocessor such as the accelerator is consistent with the actual power consumption, the nominal power consumption can also be used as the actual power consumption of the first component. For the actual power consumption of memory, the actual power consumption of memory can be calculated based on the actual usage of the memory (for example, when inserting one or more memory modules onto a single board, the actual power consumption of the memory can be calculated using a memory conversion factor when multiple memory modules are inserted). For fan power consumption, the power consumption at different speeds and duty cycles can be obtained through thermal engineering curves or measured fan curves, and the mapping relationship between the corresponding ambient temperature, load, and control strategy can be recorded to query the actual power consumption corresponding to the risk based on the actual operating environment of the rack server.

[0049] In this paper, the power consumption data of the first component directly obtained by the first software is the third power consumption, and the power consumption data of the first component obtained by the hardware power consumption acquisition device is the fourth power consumption.

[0050] In practical applications, there is often a power consumption discrepancy between the third and fourth power consumption values. The third power consumption value often fails to directly reflect the true power consumption of the first component, while the fourth power consumption value is generally more accurate and reliable. This is because hardware power consumption acquisition devices can perform physical measurements directly on the power supply circuit, covering not only the component's own operating power consumption but also capturing hidden losses such as board line losses, power conversion losses, and leakage losses, thus accurately reflecting the total power consumption obtained by the first component from the power grid or power source. However, software has limitations in collecting power consumption data from the first component. On the one hand, software typically relies on underlying drivers or sensor interfaces to obtain power consumption data. Therefore, the power consumption data obtained by the software is often an estimate of the first component's power consumption, frequently only counting the core operating power consumption and ignoring losses in peripheral circuits and conversion stages. On the other hand, due to sampling frequency limitations, sensor calibration differences, and system errors within the software itself, the power consumption data obtained by the software is prone to deviation.

[0051] In this paper, when collecting the first power consumption based on the first power consumption evaluation model of the first component, the first power consumption of the first component is directly read by the first software deployed in the first power consumption evaluation model. Therefore, in order to ensure that the first power consumption obtained based on the first software is closer to the actual power consumption of the first component, a first coefficient for power consumption adjustment for the first software is first set in the first power consumption evaluation model to adjust the power consumption data of the first component directly obtained by the first software, thus obtaining the first power consumption. The first coefficient can be determined by obtaining the third power consumption of the first component based on the first software and the fourth power consumption of the first component based on the hardware power consumption acquisition device under the same operating conditions, and then obtaining the first coefficient by the ratio of the third power consumption to the fourth power consumption.

[0052] S230: Based on the first power consumption, obtain the second power consumption of the rack server.

[0053] The second power consumption characterizes the overall power consumption of the rack server. It indicates the summarization of the first power consumption of each first component, completing the rack-level power consumption summary for the rack server. The second power consumption reflects the actual power consumption level of the rack server. It can be understood as the power consumption generated by progressively adding the sums of the first power consumption of each first component after introducing a first coefficient correction to the first power consumption of a single first component. Optionally, the second power consumption can be the actual power consumption of the rack server under any operating condition. For example, this operating condition can be at least one of the following: idle operating condition, single-component extreme operating condition, mixed stress test operating condition, representative business operating condition, or peak full load operating condition. The representative business operating condition can be an operating condition where the rack server is performing model inference, a operating condition where the rack server is performing large model training, or an operating condition where the rack server is performing agent training, etc.

[0054] In practical applications, because the hardware inside a rack server does not have a strict nesting relationship, there are components deployed independently of a single board and independent boards that do not belong to any node. Therefore, when calculating the second power consumption of a rack server based on the first power consumption of the first component, simply adding the first power consumption of the first component at a single level to calculate the second power consumption can easily lead to omissions or double calculations of power consumption.

[0055] To ensure accurate calculation of the second power consumption of rack servers, this paper supports a multi-level hybrid calculation mode for first components. This means that different levels of hardware can be selected as independent first components simultaneously. The relationships between different hardware components are clarified, and first components of appropriate granularity are used to summarize the data within nested structures. Hardware that is independent across levels is defined as an independent first component and counted separately. While ensuring that each physical entity participates in the statistics only once and avoiding duplicate summaries, the first power consumption of all first components at different levels is summarized to finally obtain the second power consumption of the entire rack server.

[0056] The aforementioned method for evaluating the power consumption of a rack server, when evaluating the power consumption of the rack server based on the execution of the first operation, obtains the first power consumption corresponding to the first component through a first power consumption evaluation model. This first power consumption is the power consumption adjusted by a first coefficient in the first power consumption evaluation model to accurately reflect the true power consumption of the first component, including power loss. Based on this, the overall power consumption of the rack server is obtained from the first power consumption of each first component. This achieves accurate evaluation of the true and accurate overall power consumption of the rack server without requiring actual testing of the entire rack server, and reduces the difficulty of power consumption evaluation. Figure 3 This is a flowchart illustrating a method for evaluating the power consumption of a rack server under one scenario. The technical solution in this scenario can be combined with implementation methods in other scenarios. For identical or related parts, descriptions of other scenarios can be used, and will not be repeated here. Figure 3 As shown, the method in this case may specifically include: S310, Receive a first operation, the first operation being used at least to indicate the power consumption of the rack server.

[0057] S320. Based on the first power consumption evaluation model of the first component, obtain the first power consumption of the first component. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component.

[0058] S330, in response to the first component being associated with a second power consumption evaluation model, obtain the first power consumption of the first component associated with the second power consumption evaluation model, wherein the second power consumption evaluation model includes a second coefficient, which is used to adjust the power consumption of the second component.

[0059] The second component characterizes hardware entities that have cross-level physical or logical connections with the first component. The second component identifies hardware objects requiring power correction during power consumption assessment, encompassing related elements in the multi-level architecture of the server, excluding the first component. The second component's hierarchical position includes at least one of device level, board level, node level, and rack level, and is different from the level of the first component. The second component includes, but is not limited to, the first component's upstream power supply or cooling equipment, downstream load hardware, peer-to-peer collaborative components, and related hardware that generates thermal or electrical coupling across levels. Optionally, the second component can reflect the board substrate used by the first component in the rack server. The board substrate power consumption is used to characterize the power consumption of the motherboard and onboard auxiliary circuits, excluding the split components. For example, board substrate power consumption = board input power consumption - split processor power consumption - split accelerator power consumption - memory power consumption - storage power consumption - network component power consumption - fan power consumption - other individually identifiable device power consumption. It is understood that the power consumption of the single-board substrate here is only for illustrative purposes and is not intended to limit the power consumption of the single-board substrate in reality.

[0060] The second power consumption assessment model can be used to characterize a dedicated simulation model built for the second component. The second power consumption assessment model can be understood as a simulation power consumption calculation model bound to the level of the first component, used to perform cross-level power consumption calculations. The second power consumption assessment model contains all the algorithmic requirements for dynamically adjusting the power consumption of the second component using a second coefficient to achieve accurate power consumption assessment of the first component.

[0061] The second coefficient can be understood as a power adjustment coefficient configured in the second power consumption evaluation simulation model, serving as a correction factor for the fifth power consumption. The second coefficient is used to compensate for the power consumption deviation between the fifth power consumption and the actual power consumption of the second component. The second coefficient can also be understood as a power adjustment coefficient used to adjust the overall power consumption of the second component. The second coefficient is used to compensate for the power consumption deviation between the software power consumption reading and the actual power consumption of the second component. The second coefficient is also used to eliminate the difference between the simple summation of power consumption based on the next-level device and the actual power consumption of the second component. The second coefficient can correspond to the single-board level conversion efficiency between the second components. Optionally, the second coefficient is used to adjust for power consumption calculation errors caused by the assembly loss, line loss, and cooperative operation deviation of the second component itself.

[0062] In practical applications, the first component can correspond to different levels of the measured hardware hierarchy of a rack server. For example, the first component can be a device at the device level, a board at the board level, a node at the node level, or the entire rack server. In this paper, when the first component is associated with a second component at a higher level, it can be considered that the first component is associated with a second power consumption evaluation model.

[0063] Taking the first component as an example, it can be deployed on a single-board substrate (i.e., the second component) to form a single board. However, during operation, the devices deployed on the single board experience interaction losses between hardware layers, such as coupling losses or voltage conversion losses, leading to a discrepancy between the software power consumption reading of the first component and its actual power consumption. Therefore, this paper, upon detecting that the first component is associated with a second power consumption evaluation model, uses the second coefficient in the second power consumption evaluation model to adjust the power consumption of the second component. Simultaneously, by accurately calculating the interaction losses between the first and second components, the first power consumption of the first component is indirectly readjusted, resulting in a more accurate and reliable first power consumption corresponding to the first component.

[0064] S340, based on the first power consumption, the fifth power consumption of the second component, and the second coefficient, obtains the sixth power consumption.

[0065] The fifth power consumption is used to characterize the power consumption data corresponding to the second component, obtained based on the software program. The fifth power consumption is uncalibrated and deviates from the actual power consumption of the second component. It only corresponds to the operating power consumption of the second component itself, but does not cover various power losses generated during the operation of the second component.

[0066] The sixth power consumption can be understood as the total power consumption, including the power consumption of the second component and the first power consumption of all the first components deployed in the second component. The sixth power consumption is used to characterize the overall power consumption of the second component that covers all the first components.

[0067] In one situation, such as Figure 4 As shown, taking the first component as an example, the first component can be deployed on a single-board substrate (i.e., the second component). By detecting the current passing through the first component using first software, the software power consumption reading (i.e., the first power consumption) of the first component can be obtained. Similarly, the power consumption of the second component can be read using the first software to obtain the fifth power consumption of the second component, which is displayed through the first interface. By adjusting the fifth power consumption using the second coefficient in the second power consumption evaluation model, the actual power consumption of the second component is obtained. Based on this, the first power consumption of each first component deployed on the second component is superimposed with the fifth power consumption adjusted by the second coefficient to obtain the overall power consumption of the second component, which covers all first components, i.e., the sixth power consumption.

[0068] S350, based on the sixth power consumption, obtains the second power consumption.

[0069] Building upon the example above, a rack server includes multiple second components, each of which can correspond to a sixth power consumption. In some cases, the second power consumption is obtained based on the sixth power consumption, including obtaining the second power consumption based on the sum of all sixth power consumptions.

[0070] In some cases, a second power consumption is obtained based on a sixth power consumption, including: obtaining a second power consumption based on a sixth power consumption and a first power consumption of at least a portion of the first component, wherein at least a portion of the first component is used to reflect a first component that is not associated with a second power consumption model.

[0071] In practical applications, please refer to [link / reference]. Figure 4 In a rack server, the first component can be associated with a second power consumption assessment model or it can exist independently. Therefore, when calculating the second power consumption of an actual rack server, directly adding the sixth power consumption corresponding to all the second components might miss the power consumption of some independent first components, leading to an error between the final second power consumption and the actual power consumption of the rack server. Therefore, if an independent first component exists in the rack server, the first power consumption corresponding to the independent first component should be added to the sum of all the sixth power consumption values ​​to obtain a more accurate second power consumption for the rack server.

[0072] The aforementioned method for evaluating the power consumption of rack servers, when the first component is associated with a second power consumption evaluation model, comprehensively considers the power loss between the first and second components during operation, and obtains the sixth power consumption corresponding to the second component based on the second power consumption evaluation model. Furthermore, by superimposing the power consumption data of all second components in the rack server and independent first components not associated with the second power consumption evaluation model, a more accurate true power consumption of the rack server can be obtained.

[0073] Figure 5 This is a flowchart illustrating a method for evaluating the power consumption of a rack server under one scenario. The technical solution in this scenario can be combined with implementation methods in other scenarios. For identical or related parts, descriptions of other scenarios can be used, and will not be repeated here. Figure 5 As shown, the method in this case may specifically include: S410, Receive a first operation, the first operation being used at least to indicate the power consumption of the rack server.

[0074] S420. Based on the first power consumption evaluation model of the first component, obtain the first power consumption of the first component. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component.

[0075] S430, in response to the second component being associated with a third power consumption evaluation model, obtain the sixth power consumption of the second component associated with the third power consumption evaluation model, the third power consumption evaluation model including a third coefficient, the third coefficient being used to adjust the power consumption of the third component.

[0076] The third component characterizes hardware entities that have cross-level physical or logical connections with the first component. The third component clarifies the hardware objects requiring power correction during power consumption assessment, encompassing related elements in the multi-level architecture of the server, excluding the first and second components. The third component's hierarchical position includes at least one of device level, board level, node level, and rack level, and differs from the levels of the first and second components. The third component includes, but is not limited to, the parent network node, compute node, or switching node of the second component. Optionally, the third component reflects the service nodes and switching nodes within the rack server that support the operation of both the second and first components.

[0077] The third power consumption assessment model can be used to characterize a dedicated simulation model built for the third component. The third power consumption assessment model can be understood as a simulation power consumption calculation model bound to the level of the second component, used to perform cross-level power consumption calculations. The third power consumption assessment model contains all the algorithmic requirements for dynamically adjusting the power consumption of the third component using a third coefficient to achieve accurate power consumption assessment of the second component.

[0078] The third coefficient can be understood as a power adjustment coefficient configured in the second power consumption evaluation simulation model, serving as a correction factor for the sixth power consumption. The third coefficient is used to compensate for the power consumption deviation between the sixth power consumption and the actual power consumption of the third component. The third coefficient can also be understood as a power adjustment coefficient used to adjust the overall power consumption of the third component. The third coefficient is used to compensate for the power consumption deviation between the software power consumption reading and the actual power consumption of the third component. The third coefficient is also used to eliminate the difference between the simple summation of power consumption based on the next-level device and the actual power consumption of the third component. The second coefficient can correspond to the node-level conversion efficiency between the third components. Optionally, the third coefficient is used to adjust for power consumption calculation errors caused by the third component's own assembly losses, line losses, and cooperative operation deviations.

[0079] In practical applications, the third component can correspond to different levels of the measured hardware hierarchy of a rack server. For example, the second component can be an accelerator board at the board level, a node at the node level, or the entire rack server. In this paper, when the second component is associated with a third component at a higher level, it can be considered that the second component is associated with a third power consumption evaluation model.

[0080] Taking the second component as a single board as an example, the second component can be deployed in a computer system (i.e., the third component) to form a node, such as a computing node. In this case, due to the interaction losses between cross-level hardware during operation, such as coupling losses or voltage conversion losses, there is a certain deviation between the software power consumption reading of the second component and the actual power consumption. Based on this, this paper, when the second component is detected to be associated with a third power consumption evaluation model, uses the third coefficient in the third power consumption evaluation model to adjust the power consumption of the third component. Simultaneously, by accurately calculating the interaction losses between the second and third components, the sixth power consumption of the second component is indirectly readjusted to obtain a more realistic and reliable sixth power consumption.

[0081] S440, based on the sixth power consumption, the seventh power consumption of the third component, and the third coefficient, obtains the eighth power consumption output by the third power consumption evaluation model. The seventh power consumption is used to reflect the node power consumption of the auxiliary node associated with the third component.

[0082] The seventh power consumption is used to characterize the power consumption data corresponding to the third component, obtained based on the software program. The seventh power consumption is uncalibrated and deviates from the actual power consumption of the third component. It only corresponds to the operating power consumption of the third component itself, but does not cover various power losses generated during the operation of the third component.

[0083] In one situation, such as Figure 6 As shown, taking the second component as an example, the second component can be deployed on nodes (such as compute nodes, network nodes, or switching nodes). The first software detects the current passing through the second component to obtain its corresponding software power consumption value (i.e., the sixth power consumption). Similarly, the first software can read the power consumption of the third component to obtain the seventh power consumption. By adjusting the seventh power consumption using the third coefficient in the third power consumption evaluation model, the actual power consumption of the third component is obtained. Based on this, the sixth power consumption of each second component deployed on the third component is superimposed with the seventh power consumption adjusted by the third coefficient to obtain the overall power consumption of the third component, encompassing all second components, i.e., the eighth power consumption.

[0084] For example, taking the power consumption assessment of a certain node (i.e., the third component) as an example, the node power consumption can be determined by the sum of the power consumption of each board within the node, the power consumption of the node fan, the power consumption of the node auxiliary load, and the third coefficient. For example, node power consumption = (sum of power consumption of each board + power consumption of the node fan + power consumption of the node auxiliary load) / third coefficient.

[0085] The power consumption of node auxiliary loads includes, but is not limited to, management controllers, standby power supplies, bridge chips, and other secondary components that are not separately metered. Auxiliary loads that cannot be separated individually can be included in the node auxiliary load power consumption. It is understood that the node auxiliary loads mentioned here are only illustrative examples and do not constitute a limitation on the actual power consumption of node auxiliary loads.

[0086] S450, based on the eighth power consumption, obtains the second power consumption.

[0087] Building upon the example above, a rack server includes multiple third components, each corresponding to an eighth power consumption. In some cases, a second power consumption is obtained based on the eighth power consumption, including obtaining the second power consumption based on the sum of all eighth power consumptions.

[0088] In some cases, a second power consumption is obtained based on an eighth power consumption, including: obtaining a second power consumption based on the eighth power consumption, the sixth power consumption of at least a portion of the second components, and the first power consumption of at least a portion of the first components, wherein at least a portion of the second components are used to reflect components not associated with a third power consumption evaluation model, and at least a portion of the first components are used to characterize components not associated with a second power consumption evaluation model.

[0089] It should be noted that, please continue to refer to Figure 6 In a rack server, the first component can be associated with a second power consumption assessment model or be an independent component. Similarly, the second component can be associated with a third power consumption assessment model or be an independent component. Therefore, when calculating the second power consumption of an actual rack server, simply adding the eighth power consumption corresponding to all the third components may miss the power consumption of some independent first components (such as device 1) and second components (such as board 6), leading to an error between the final second power consumption and the actual power consumption of the rack server. Therefore, if there are independent first and second components in the rack server, the first power consumption corresponding to the independent first component and the sixth power consumption corresponding to the second component should be added to the sum of all the eighth power consumptions to obtain a more accurate second power consumption for the rack server.

[0090] The aforementioned method for evaluating the power consumption of rack servers, when the second component is associated with a third power consumption evaluation model, comprehensively considers the power loss between the second and third components during operation, and obtains the eighth power consumption corresponding to the third component based on the third power consumption evaluation model. On this basis, by superimposing the power consumption data of all third components in the rack server, independent second components not associated with the third power consumption evaluation model, and independent first components not associated with the second power consumption evaluation model, a more accurate true power consumption of the rack server can be obtained.

[0091] Figure 7This is a flowchart illustrating a method for evaluating the power consumption of a rack server under one scenario. The technical solution in this scenario can be combined with implementation methods in other scenarios. For identical or related parts, descriptions of other scenarios can be used, and will not be repeated here. Figure 7 As shown, the method in this case may specifically include: S510: Receive a first operation, the first operation being used at least to indicate the power consumption of the rack server.

[0092] S520. Based on the first power consumption evaluation model of the first component, the first power consumption of the first component is obtained. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component.

[0093] S530: Based on the first power consumption, obtain the second power consumption of the rack server.

[0094] The operating data of the S540 rack server is used to obtain the fourth coefficient, which is used to reflect the power consumption adjustment coefficient of the rack server.

[0095] The fourth coefficient can be understood as a power consumption adjustment factor characterizing the rack server. It quantifies the efficiency loss of the rack server during rack-level power conversion. The fourth coefficient can be obtained by taking the reciprocal of the rack-level conversion efficiency of the rack server.

[0096] S550, based on the second power consumption and the fourth coefficient, obtains the ninth power consumption of the rack server.

[0097] Among them, the ninth power consumption is used to characterize the actual power consumption required by the rack server during actual operation.

[0098] During the operation of a rack server, there may be some power loss during the rack-level power conversion process. Based on this, after obtaining the second power consumption of the rack server, the power consumption is adjusted by the fourth coefficient to obtain the ninth power consumption of the rack server in actual operation.

[0099] In some cases, the method for evaluating the power consumption of a rack server also includes: obtaining a fifth coefficient of the first component in response to adjusting the first component; and adjusting the first power consumption evaluation model of the first component based on the fifth coefficient of the first component.

[0100] The fifth coefficient characterizes the power consumption adjustment coefficient corresponding to the updated first component after its parameters are updated or the entire first component is replaced in the rack server. The fifth coefficient is configured in the first power consumption evaluation model and replaces the first coefficient in the first power consumption evaluation model.

[0101] In practical applications, the components in a rack server are not static due to actual business needs or natural aging of components. If the parameters of the components in the rack server are updated or the components are replaced, it is necessary to re-evaluate the power consumption of the entire rack server.

[0102] Based on this, after standard testing, calibration, and hierarchical modeling of the rack server, this paper obtains a first power consumption assessment model corresponding to the first component in the rack server, and this first power consumption assessment model is a reusable power consumption calculation model. In subsequent power consumption assessment processes, if the component parameters of the first component (including the component's own parameters and the environmental parameters associated with the first component, such as temperature parameters) change or the first component is replaced as a whole, only the applicability adjustment of the first power consumption assessment model corresponding to the first component needs to be performed to quickly obtain a new first power consumption assessment model.

[0103] In a preferred implementation, when evaluating the power consumption of the updated rack server, the overall power consumption of the updated rack server can be quickly evaluated without rebuilding the rack server by updating at least one input item of the first power consumption evaluation model of the first component. Furthermore, when the input item corresponds to AI inference, AI training, or large model algorithm loads, the calculation results can also be used to analyze the chip power consumption distribution, node energy efficiency, and hardware utilization under unit algorithm loads.

[0104] The advantage of this setup is that it is applicable not only to the initial evaluation of the overall power consumption of rack servers, but also to continuous power management scenarios after rack servers are installed in data centers. This is particularly relevant when components are replaced, AVLs are changed, the number of nodes is adjusted, switching modules are reconfigured, or cooling strategies are adjusted. By dynamically adjusting the initial power consumption evaluation model of the first component in the rack server, the workload of repetitive rack-wide testing can be significantly reduced, while improving power resource utilization and delivery efficiency while ensuring accuracy.

[0105] The above-mentioned method for evaluating the power consumption of rack servers allows for the rapid acquisition of the first power consumption evaluation model corresponding to each component in the updated rack server without rebuilding the rack server hardware entity, by adjusting the first power consumption evaluation model corresponding to the first component when the component parameters of the first component are updated or the first component is replaced. This enables the rapid evaluation of the overall power consumption of the rack server.

[0106] Figure 8 This is a schematic diagram of a device for evaluating the power consumption of a rack server under one specific scenario, such as... Figure 8 As shown, the above-mentioned device includes: a receiving module 610, used to receive a first operation, the first operation being at least used to indicate the power consumption of the rack server; The first module 620 is used to obtain the first power consumption of the first component based on the first power consumption evaluation model of the first component. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. The second module 630 is used to obtain the second power consumption of the rack server based on the first power consumption.

[0107] The aforementioned rack server power consumption assessment device, when performing a power consumption assessment on the rack server based on the first operation, obtains the first power consumption corresponding to the first component through a first power consumption assessment model. This first power consumption is the power consumption adjusted by a first coefficient in the first power consumption assessment model to accurately reflect the true power consumption of the first component, including power loss. Based on this, the total power consumption of the rack server is obtained from the first power consumption of each first component. This achieves accurate assessment of the true and accurate total power consumption of the rack server without requiring actual testing of the entire rack server, and reduces the difficulty of power consumption assessment. In one scenario, the first power consumption evaluation model is obtained as follows: The third power consumption and the fourth power consumption of the first component in the first state are obtained. The third power consumption is used to characterize the operating power consumption of the first component, and the fourth power consumption is used to characterize the sum of the operating power consumption and the power loss of the first component. Based on the third power consumption and the fourth power consumption, obtain the first coefficient of the first component; Based on the first coefficient, a first power consumption evaluation model for the first component is obtained, and the independent variables in the first power consumption evaluation model are related to the power consumption read by the first software.

[0108] In one scenario, the third power consumption is obtained by the processor reading from the first component through the first hardware interface, the processor communicating with the first component through the first hardware interface, and the fourth power consumption is obtained by measuring the first power consumption device.

[0109] In one scenario, the second module includes: a first response subunit, configured to, in response to a first component being associated with a second power consumption evaluation model, obtain a first power consumption of the first component associated with the second power consumption evaluation model, wherein the second power consumption evaluation model includes a second coefficient, the second coefficient being used to adjust the power consumption of the second component; The sixth power consumption obtaining unit is used to obtain the sixth power consumption based on the first power consumption, the fifth power consumption of the second component, and the second coefficient; The second power consumption acquisition unit is used to obtain the second power consumption based on the sixth power consumption.

[0110] In one scenario, the second power consumption obtaining unit is used to obtain the second power consumption based on the sum of all the sixth power consumptions.

[0111] In one scenario, a second power consumption obtaining unit is used to obtain a second power consumption based on a sixth power consumption and a first power consumption of at least a portion of the first component, wherein at least a portion of the first component is used to reflect a first component that is not associated with a second power consumption model.

[0112] In one scenario, the second unit includes: a second response subunit, configured to obtain a sixth power consumption of the second component associated with a third power consumption evaluation model in response to the second component being associated with the third power consumption evaluation model, wherein the third power consumption evaluation model includes a third coefficient, the third coefficient being used to adjust the power consumption of the third component; The eighth power consumption acquisition sub-unit is used to obtain the eighth power consumption output by the third power consumption evaluation model based on the sixth power consumption, the seventh power consumption of the third component and the third coefficient. The seventh power consumption is used to reflect the node power consumption of the auxiliary node associated with the third component. The second power consumption acquisition subunit is used to obtain the second power consumption based on the eighth power consumption.

[0113] In one scenario, a second power consumption acquisition sub-unit is used to obtain the second power consumption based on the sum of all the eighth power consumptions.

[0114] In one scenario, a second power consumption obtaining subunit is used to obtain a second power consumption based on an eighth power consumption, a sixth power consumption of at least a portion of the second component, and a first power consumption of at least a portion of the first component, wherein at least a portion of the second component is used to reflect a component not associated with a third power consumption evaluation model, and at least a portion of the first component is used to characterize a component not associated with a second power consumption evaluation model.

[0115] In one scenario, the first component indicates the devices in the rack server, the second component reflects the single-board substrate in the rack server that carries the first component, and the third component reflects the service nodes and switching nodes that support the operation of the second component and the first component.

[0116] In one scenario, the rack server power consumption assessment device further includes: a coefficient acquisition module, used to obtain a fourth coefficient based on the rack server's operating data after obtaining the second power consumption, the fourth coefficient being used to reflect the rack server's power consumption adjustment coefficient. The ninth power consumption acquisition module is used to obtain the ninth power consumption of the rack server based on the second power consumption and the fourth coefficient.

[0117] In one scenario, the rack server power consumption assessment device further includes: a fifth coefficient acquisition module, used to acquire a fifth coefficient of the first component in response to adjusting the first component; An adjustment module is used to adjust the first power consumption evaluation model of the first component based on the fifth coefficient of the first component.

[0118] The aforementioned rack server power consumption assessment device can execute the rack server power consumption assessment method provided in any of the embodiments described herein, and has the corresponding functional modules and beneficial effects for executing the rack server power consumption assessment method.

[0119] It is worth noting that the various units and modules included in the above-mentioned rack server power consumption evaluation device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments described herein.

[0120] The following is for reference. Figure 9 This document illustrates a schematic diagram of an electronic device (e.g., a terminal device or server) 700 suitable for implementing the aforementioned method for evaluating the power consumption of rack servers. The terminal device referred to herein may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, personal digital assistants (PDAs), tablet computers (PADs), portable multimedia players (PMPs), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital televisions and desktop computers. Figure 9 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments described herein.

[0121] like Figure 9 As shown, the electronic device 700 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 into a random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device 700. The processing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0122] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic device 700 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 9 An electronic device 700 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0123] In particular, according to embodiments of this document, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, the technical solutions of this document include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 709, or installed from storage device 708, or installed from ROM 702. When the computer program is executed by processing device 701, it performs the functions defined in the methods of the embodiments of this document.

[0124] The names of messages or information exchanged between multiple devices in this document are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0125] The electronic device provided in this embodiment and the rack server power consumption evaluation method provided in the above technical solutions belong to the same inventive concept. Technical details not described in detail in this document can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0126] This article provides a computer storage medium storing a computer program that, when executed by a processor, implements the rack server power consumption evaluation method provided in the above embodiments.

[0127] It should be noted that the computer-readable medium mentioned above can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM, also known as flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this document, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0128] According to one or more scenarios in this article, Example 1 provides a method for evaluating the power consumption of a rack server, including: receiving a first operation, the first operation being used at least to indicate the acquisition of the power consumption of the rack server; Based on the first power consumption evaluation model of the first component, the first power consumption of the first component is obtained. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. Based on the first power consumption, the second power consumption of the rack server is obtained.

[0129] Based on one or more scenarios described herein, Example 2 provides the method of Example 1, and further includes: optionally, obtaining the first power consumption evaluation model in the following manner: The third power consumption and the fourth power consumption of the first component in the first state are obtained. The third power consumption is used to characterize the operating power consumption of the first component, and the fourth power consumption is used to characterize the sum of the operating power consumption and the power loss of the first component. Based on the third power consumption and the fourth power consumption, obtain the first coefficient of the first component; Based on the first coefficient, a first power consumption evaluation model for the first component is obtained, and the independent variables in the first power consumption evaluation model are related to the power consumption read by the first software.

[0130] According to one or more scenarios in this article, Example 3 provides the method of Example 1, which further includes: Optionally, the third power consumption is obtained by the processor from the first component through the first hardware interface, the processor is communicatively connected to the first component through the first hardware interface, and the fourth power consumption is obtained by the first power consumption measuring device.

[0131] According to one or more scenarios in this article, Example 4 provides the method of Example 1, which further includes: optionally, in response to the first component being associated with a second power consumption evaluation model, obtaining the first power consumption of the first component associated with the second power consumption evaluation model, wherein the second power consumption evaluation model includes a second coefficient, the second coefficient being used to adjust the power consumption of the second component; The sixth power consumption is obtained based on the first power consumption, the fifth power consumption of the second component, and the second coefficient; Based on the sixth power consumption, the second power consumption is obtained.

[0132] According to one or more scenarios in this paper, Example 5 provides the method of Example 4, which also includes: optionally, obtaining a second power consumption based on a sixth power consumption and a first power consumption of at least a portion of the first component, wherein at least a portion of the first component is used to reflect a first component that is not associated with a second power consumption model.

[0133] According to one or more scenarios in this paper, Example 6 provides the method of Example 4, which further includes: optionally, in response to the second component being associated with a third power consumption evaluation model, obtaining a sixth power consumption of the second component associated with the third power consumption evaluation model, the third power consumption evaluation model including a third coefficient, the third coefficient being used to adjust the power consumption of the third component; Based on the sixth power consumption, the seventh power consumption of the third component, and the third coefficient, the eighth power consumption output by the third power consumption evaluation model is obtained. The seventh power consumption is used to reflect the node power consumption of the auxiliary node associated with the third component. Based on the eighth power consumption, the second power consumption is obtained.

[0134] According to one or more of the present invention, Example 7 provides the method of Example 6, which further includes: optionally, obtaining a second power consumption based on an eighth power consumption, a sixth power consumption of at least a portion of the second component, and a first power consumption of at least a portion of the first component, wherein at least a portion of the second component is used to reflect a component not associated with a third power consumption evaluation model, and at least a portion of the first component is used to characterize a component not associated with a second power consumption evaluation model.

[0135] According to one or more scenarios in this document, Example 8 provides the method of Example 1, which further includes: optionally, a first component is used to indicate devices in a rack server, a second component is used to reflect the single-board substrate in the rack server used to carry the first component, and a third component is used to reflect the service node and switching node that support the operation of the second component and the first component.

[0136] According to one or more scenarios in this article, Example 9 provides any of the evaluation methods in Examples 1 to 6, and also includes: optionally, obtaining a fourth coefficient based on the operating data of the rack server, the fourth coefficient being used to reflect the power consumption adjustment coefficient of the rack server; Based on the second power consumption and the fourth coefficient, the ninth power consumption of the rack server is obtained.

[0137] According to one or more scenarios in this article, Example 10 provides the method of Example 1, which also includes: optionally, in response to adjusting the first component, obtaining the fifth coefficient of the first component; The first power consumption evaluation model of the first component is adjusted based on the fifth coefficient of the first component.

[0138] According to one or more scenarios in this article, Example 11 provides the method of Example 1, and also includes: optionally, obtaining the second power consumption based on the sum of all sixth power consumptions.

[0139] According to one or more scenarios in this article, Example Twelve provides the method of Example One, and also includes: optionally, obtaining the second power consumption based on the sum of all eighth power consumptions.

[0140] According to one or more scenarios described herein, Example Thirteen provides an apparatus for evaluating the power consumption of a rack server, comprising: a receiving module for receiving a first operation, the first operation being at least used to indicate the acquisition of the power consumption of the rack server; The first module is used to obtain the first power consumption of the first component based on the first power consumption evaluation model of the first component. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. The second module is used to obtain the second power consumption of the rack server based on the first power consumption.

[0141] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol, such as HTTP (Hypertext Transfer Protocol), and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include Local Area Networks (LANs), Wide Area Networks (WANs), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0142] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0143] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: receive a first operation, the first operation being at least used to instruct the acquisition of the power consumption of the rack server; obtain a first power consumption of the first component based on a first power consumption evaluation model of the first component, the first power consumption evaluation model including at least a first coefficient, the first coefficient being used to adjust the power consumption data of the first component; and obtain a second power consumption of the rack server based on the first power consumption.

[0144] Computer program code for performing the operations described herein can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" programming language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0145] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this document. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0146] The modules or units described herein may be implemented in software or hardware. The names of modules or units do not necessarily constitute a limitation on the module or unit itself; for example, the first unit may also be described as "a unit that acquires at least two Internet Protocol addresses".

[0147] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include at least one of the following: Field-Programmable Gate Array (FPGA), Application-Specific Integrated Circuit (ASIC), Application-Specific Standard Product (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), etc.

[0148] In the context of this document, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (flash memory), optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0149] The above description is merely a preferred embodiment and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure herein is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed herein that have similar functions.

[0150] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of this document. Certain features described in the context of individual implementations may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0151] Although the subject matter has been described using a programming language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative forms of implementing the claims.

Claims

1. A method for evaluating the power consumption of a rack server, comprising: Receive a first operation, the first operation being at least used to indicate the acquisition of the power consumption of the rack server; Based on the first power consumption evaluation model of the first component, the first power consumption of the first component is obtained. The first component is a component in the rack server. The first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. Based on the first power consumption, the second power consumption of the rack server is obtained.

2. The evaluation method according to claim 1, wherein the first power consumption evaluation model is obtained in the following manner: The third power consumption and the fourth power consumption of the first component in the first state are obtained. The third power consumption is used to characterize the operating power consumption of the first component, and the fourth power consumption is used to characterize the sum of the operating power consumption and the power loss of the first component. Based on the third power consumption and the fourth power consumption, obtain the first coefficient of the first component; Based on the first coefficient, a first power consumption evaluation model for the first component is obtained, wherein the independent variables in the first power consumption evaluation model are related to the power consumption read by the first software.

3. The method according to claim 2, comprising: The third power consumption is obtained by the processor from the first component through the first hardware interface. The processor communicates with the first component through the first hardware interface. The fourth power consumption is obtained by the first power consumption measuring device.

4. The evaluation method according to claim 1, wherein obtaining the second power consumption of the rack server based on the first power consumption includes: In response to the first component being associated with a second power consumption evaluation model, the first power consumption of the first component associated with the second power consumption evaluation model is obtained. The second power consumption evaluation model includes a second coefficient, which is used to adjust the power consumption of the second component. The sixth power consumption is obtained based on the first power consumption, the fifth power consumption of the second component, and the second coefficient; The second power consumption is obtained based on the sixth power consumption.

5. The evaluation method according to claim 4, wherein obtaining the second power consumption based on the sixth power consumption includes: The second power consumption is obtained based on the sixth power consumption and at least a portion of the first power consumption of the first component, wherein at least a portion of the first component is used to reflect the first component that is not associated with the second power consumption model.

6. The method according to claim 4, wherein obtaining the second power consumption based on the sixth power consumption comprises: In response to the second component being associated with a third power consumption evaluation model, a sixth power consumption of the second component associated with the third power consumption evaluation model is obtained, wherein the third power consumption evaluation model includes a third coefficient, which is used to adjust the power consumption of the third component; Based on the sixth power consumption, the seventh power consumption of the third component, and the third coefficient, the eighth power consumption output by the third power consumption evaluation model is obtained. The seventh power consumption is used to reflect the node power consumption of the auxiliary node associated with the third component. The second power consumption is obtained based on the eighth power consumption.

7. The evaluation method according to claim 6, wherein obtaining the second power consumption based on the eighth power consumption comprises: The second power consumption is obtained based on the eighth power consumption, the sixth power consumption of at least a portion of the second component, and the first power consumption of at least a portion of the first component. The second component is used to reflect the component that is not associated with the third power consumption evaluation model, and the first component is used to characterize the component that is not associated with the second power consumption evaluation model.

8. The evaluation method according to any one of claims 1-7, wherein the first component is used to indicate the devices in the rack server, the second component is used to reflect the single-board substrate in the rack server for carrying the first component, and the third component is used to reflect the service node and switching node that support the operation of the second component and the first component.

9. The evaluation method according to any one of claims 1-7, wherein after obtaining the second power consumption, the method further comprises: Based on the operating data of the rack server, a fourth coefficient is obtained, which is used to reflect the power consumption adjustment coefficient of the rack server. Based on the second power consumption and the fourth coefficient, the ninth power consumption of the rack server is obtained.

10. The evaluation method according to claim 1, further comprising: In response to adjusting the first component, the fifth coefficient of the first component is obtained; The first power consumption evaluation model of the first component is adjusted based on the fifth coefficient of the first component.

11. A device for evaluating the power consumption of a rack server, characterized in that, include: A receiving module is configured to receive a first operation, wherein the first operation is at least used to indicate the power consumption of the rack server; The first module is used to obtain the first power consumption of the first component based on the first power consumption evaluation model of the first component, wherein the first component is a component in the rack server, and the first power consumption evaluation model includes at least a first coefficient, which is used to adjust the power consumption data of the first component. The second module is used to obtain the second power consumption of the rack server based on the first power consumption.

12. An electronic device, the electronic device comprising: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the rack server power consumption evaluation method as described in any one of claims 1-10.

13. A storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the method for evaluating the power consumption of a rack server as described in any one of claims 1-10.

14. A computer program product comprising a computer program that, when executed by a processor, implements a method for evaluating the power consumption of a rack server as described in any one of claims 1-10.

Images