Energy efficiency measurement method, apparatus, device, storage medium and computer program product

Abstract

The application discloses an energy efficiency measurement method, device, equipment, storage medium and computer program product. A control command carrying an infrastructure identifier is received to obtain power information and task load information of an infrastructure corresponding to the infrastructure identifier when the infrastructure runs a task, wherein the power information and the task load information are provided with a timestamp. Energy efficiency of the infrastructure is calculated according to the power information and the task load information of the same timestamp, and the accuracy of the energy efficiency measurement of the computing power of the infrastructure is improved.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and more specifically, to an energy efficiency measurement method, apparatus, device, storage medium, and computer program product. Background Technology

[0002] In existing technologies, the computing power of infrastructure is typically measured from the dimensions of computing, network, memory, and storage. However, when infrastructure deploys software with the same business capabilities, the efficiency of the same version of software varies greatly on different platforms. This is not entirely dependent on the computing power of the infrastructure's hardware resources, but also on the differences in the infrastructure's system architecture. This can lead to a situation where the actual energy efficiency is disproportionate to the nominal computing power of the hardware resources (such as the fixed-point / floating-point operation capability of the CPU). Existing technologies cannot accurately measure the energy efficiency level of infrastructure operation by measuring general resources. Summary of the Invention

[0003] Based on this, the present invention provides an energy efficiency measurement method, apparatus, device, storage medium, and computer program product, which can calculate energy efficiency by acquiring task load information and power information of infrastructure when it is running tasks, thereby improving the accuracy of computing power energy efficiency measurement of infrastructure.

[0004] To achieve the above objectives, embodiments of the present invention provide an energy efficiency measurement method, comprising:

[0005] Receive control commands carrying infrastructure identifiers;

[0006] Obtain the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped;

[0007] The energy efficiency of the infrastructure is calculated based on the power information and the task load information at the same timestamp.

[0008] As an improvement to the above scheme, the task load information includes at least one cell key parameter defined by 3GPP that is related to the processing overhead of the cell handled by the infrastructure.

[0009] As an improvement to the above scheme, the step of calculating the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp includes:

[0010] Calculate the normalized load based on the cell key parameters and their corresponding set weights in the task load information;

[0011] The energy efficiency of the infrastructure is obtained by dividing the normalized load at the same timestamp by the power information.

[0012] As an improvement to the above scheme, the task load information includes at least one of RB utilization, number of users, modulation order, and cell bandwidth.

[0013] As an improvement to the above scheme, when the deviation between the timestamp of the task load information and the timestamp of the power information is less than a set deviation threshold, the two timestamps are considered to be the same timestamp.

[0014] As an improvement to the above solution, the power information is obtained by a power detection circuit designed inside the infrastructure, or by the power supply facilities of the computer room where the infrastructure is located.

[0015] As an improvement to the above scheme, it also includes: performing a weighted average calculation of the energy efficiency of the infrastructure at multiple timestamps to obtain the average energy efficiency of the infrastructure.

[0016] As an improvement to the above solution, it also includes:

[0017] From among multiple infrastructures with the same infrastructure model and data center conditions, the infrastructure with higher average energy efficiency is selected as the preferred choice for new deployments of the same type of task.

[0018] As an improvement to the above solution, the method is applied to the first unit; the control command is sent by the second unit to the first unit after the deployment and operation of the task on the infrastructure is initiated; the power information is obtained by the first unit from the infrastructure or the computer room where the infrastructure is located.

[0019] To achieve the above objectives, embodiments of the present invention also provide an energy efficiency measurement device, comprising:

[0020] The command receiving module is used to receive control commands carrying infrastructure identifiers.

[0021] The information acquisition module is used to acquire the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped.

[0022] An energy efficiency measurement module is used to calculate the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp.

[0023] To achieve the above objectives, embodiments of the present invention also provide an energy efficiency measurement device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the energy efficiency measurement method as described in any of the above embodiments.

[0024] To achieve the above objectives, embodiments of the present invention also provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the energy efficiency measurement method as described in any of the above embodiments.

[0025] To achieve the above objectives, embodiments of the present invention also provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the energy efficiency measurement method as described in any of the above embodiments.

[0026] Compared with existing technologies, the energy efficiency measurement method, apparatus, device, storage medium, and computer program product disclosed in this invention obtain the power information and task load information of the infrastructure corresponding to the infrastructure identifier when it is running a task by receiving a control command carrying an infrastructure identifier, wherein the power information and the task load information are timestamped; by calculating the energy efficiency of the infrastructure based on the power information and the task load information with the same timestamp, the accuracy of the computing power energy efficiency measurement of the infrastructure is improved. Attached Figure Description

[0027] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart illustrating an energy efficiency measurement method according to an embodiment of the present invention;

[0029] Figure 2 This is an architecture diagram of an energy efficiency measurement system provided in an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram of the structure of an energy efficiency measurement device provided in an embodiment of the present invention;

[0031] Figure 4 This is a schematic diagram of the structure of an energy efficiency measurement device provided in an embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] See Figure 1 This is a flowchart illustrating an energy efficiency measurement method according to an embodiment of the present invention. Specifically, the energy efficiency measurement method includes steps S11 to S13:

[0034] S11. Receive control commands carrying infrastructure identifiers;

[0035] S12. Obtain the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped;

[0036] S13. Calculate the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp.

[0037] It is worth noting that computing networks have become a crucial driving force for technological progress and digital transformation. In computing network applications, the computational power of infrastructure is one of the fundamental technologies. Current technologies typically measure infrastructure resources from the dimensions of computing, network, memory, and storage; however, this approach relies on runtime resource usage such as CPU utilization and is not entirely suitable for all high real-time services. Furthermore, in computing networks, such as baseband or protocol processing in wireless base stations, there are numerous binding relationships to cope with sudden changes in service. Even if the current resource utilization is low, it does not mean that the bound processing units can be shared. The actual service load cannot directly correspond to the resource usage accurately obtained through traditional measurement methods. Since service load is directly related to wireless air interface characteristics and user services, such as weather changes, user location, number of users, and service changes, all of these will affect the cell's service processing overhead (service load). These factors exhibit time-varying characteristics, resulting in time-varying wireless communication service scenarios. The load on the existing network changes constantly, and even if user service types and throughput are identical, differences in air interface characteristics across different coverage areas (such as the influence of buildings and weather) lead to varying processing overhead. Therefore, traditional methods of collecting user service information are insufficient to accurately calculate the infrastructure's service load in wireless service scenarios, given the multi-dimensional changes in time-varying air interface characteristics and user services. Furthermore, regarding power consumption, when deploying software with the same wireless service capabilities on wireless infrastructure, the efficiency of the same software version varies significantly across different platforms. This is not solely dependent on the computing power of the hardware resources; differences in heterogeneous system architectures can also cause disproportion between actual energy efficiency and the nominal computing power of hardware resources (such as the fixed-point / floating-point arithmetic capabilities of the CPU). For example, a heterogeneous platform using a higher-performance CPU and accelerator may consume significantly more system energy than a platform using a lower-specification CPU with the same accelerator due to insufficient software optimization. Therefore, its operational energy efficiency cannot be simply measured using common resource metrics in traditional solutions. Furthermore, existing approaches, such as conducting computational measurements through simulation tests, use estimated models of all business loads instead of real business loads. This necessitates the provision of dedicated infrastructure for offline measurement processing, resulting in the occupation of infrastructure resources.

[0038] In this embodiment of the invention, the infrastructure running a task is measured by receiving a control command carrying an infrastructure identifier. Specifically, the measurement method involves calculating the energy efficiency of the infrastructure by obtaining its timestamped power information and task load information. This method does not measure the infrastructure's resources to achieve energy efficiency measurement, and it is independent of the infrastructure's resource usage during task execution. It is suitable for high real-time tasks. The binding relationship between the infrastructure and the task, heterogeneous system architecture, etc., will not affect the accuracy of the load and power information obtained in this embodiment. It can accurately measure the energy efficiency of the infrastructure, and this embodiment does not require offline measurement processing through mode testing, thus not consuming a large amount of infrastructure resources.

[0039] It is worth noting that the method is applicable to the measurement of computing power efficiency of wireless infrastructure, and also applicable to the computing power of other non-wireless infrastructure.

[0040] Compared with existing technologies, the method provided by the embodiments of the present invention can ensure accurate energy efficiency measurement for infrastructures with high real-time requirements, relatively small single-node computing power in distributed deployments, and limitations imposed by software optimization on task load and resource utilization. Furthermore, by achieving accurate energy efficiency measurement at the task level, the method can more efficiently schedule such computing resources, achieving energy conservation and consumption reduction, and has broad application prospects.

[0041] In one implementation, the task load information includes at least one cell key parameter defined by 3GPP that relates to the processing overhead of the cell being processed by the infrastructure.

[0042] Further, calculating the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp includes:

[0043] Calculate the normalized load based on the cell key parameters and their corresponding set weights in the task load information;

[0044] The energy efficiency of the infrastructure is obtained by dividing the normalized load at the same timestamp by the power information.

[0045] Specifically, the energy efficiency calculation process for infrastructure is as follows:

[0046] (1) Normalize the load based on the task load information. The following methods can be used, but are not limited to:

[0047] ① Use individual cell key parameters to directly correspond to the normalized load; among them, the cell key parameters have the greatest impact on processing overhead.

[0048] ② Use multiple key parameters of the cell, set weights for each parameter, and statistically normalize the load.

[0049] ③ Simulate the load on the infrastructure under test using external testing equipment to reach the specified load value.

[0050] (2) Calculate the energy efficiency statistics based on the actual specified real-time accuracy (the deviation between task load information and power information at the statistical time does not exceed a certain range). The specific calculation formula is: Normalized load / power = Normalized energy efficiency, where power information is power.

[0051] Furthermore, the key cell parameters are parameters defined by 3GPP that are related to the processing overhead of the cell handled by the infrastructure, such as RB utilization, number of users, modulation order, cell bandwidth and frequency spacing, etc. The specific selection of key cell parameters is set according to the actual situation and is not limited here.

[0052] For example, taking a communication base station as an example, a New Radio (NR) 100M bandwidth cell can have key parameters selected as needed, such as 1200 active users, specific RB utilization conditions (RB stands for Resource Block), or specific cell air interface rate, etc., which can affect processing overhead. The actual workload is calculated with the cell's maximum capacity set at 100%, and the weights of each parameter are adjusted and summed according to the actual situation. For example: RB utilization 50%, 64QAM modulation (maximum 256QAM, 40% in this example), 1200 active users (100% of the maximum number of users). RB utilization weight 60%, modulation method weight 35%, and number of users weight 5%. Then the normalized load corresponding to this cell is: 50%*60% + 40%*35% + 100%*5% = 29.5%. If two cells are deployed, the normalized loads of the two cells are summed to obtain the final normalized load.

[0053] It is worth noting that the specific types of the key parameters of the community are not limited to the specific examples mentioned above, and can be set according to the actual situation.

[0054] In one implementation, when the deviation between the timestamp of the task load information and the timestamp of the power information is less than a set deviation threshold, the two timestamps are considered to be the same timestamp.

[0055] It is understandable that in practical applications, the statistical time of task load information and the statistical time of power information may not be completely consistent. Therefore, as long as the timestamps of the two types of information are close enough, they are considered to correspond to each other and belong to the same timestamp.

[0056] In one implementation, the power information is detected by a power detection circuit designed into the infrastructure, or by the power supply facilities of the computer room where the infrastructure is located.

[0057] In one embodiment, the method further includes: performing a weighted average calculation of the energy efficiency of the infrastructure at multiple timestamps to obtain the average energy efficiency of the infrastructure.

[0058] Furthermore, the method also includes: selecting the infrastructure with higher average energy efficiency from among multiple infrastructures with the same infrastructure model and data center conditions as the preferred option for new deployments of similar tasks.

[0059] Specifically, by performing multiple calculations and weighted averages on the energy efficiency of the infrastructure for the tasks in operation, an average energy efficiency can be obtained, which can reduce errors to a certain extent and obtain a more accurate energy efficiency assessment result. In subsequent applications, if similar tasks need to be deployed, infrastructure with higher average energy efficiency will be selected for deployment and operation.

[0060] It is worth noting that in the weighted average calculation of multiple energy efficiency levels mentioned above, the weights of each energy efficiency level can be the same or different, depending on the actual application, and are not limited here.

[0061] In one implementation, the method is applied to a first unit; the control command is sent by a second unit to the first unit after the second unit deploys and starts the operation task on the infrastructure; the power information is obtained by the first unit from the infrastructure or the data center where the infrastructure is located.

[0062] Specifically, the method provided in any of the above embodiments is applicable to energy efficiency measurement systems, see [link to relevant documentation]. Figure 2 The system architecture diagram shown is as follows:

[0063] 1. Measurable infrastructure design:

[0064] (1) Infrastructure with shared computing power, such as wireless base station infrastructure with shared computing power, of which the wireless infrastructure is mainly BBU (Building Baseband Unit), or it can be an integrated form or other different forms.

[0065] (2) The infrastructure must have time synchronization capabilities. GPS / BeiDou satellite synchronization and / or IEEE 1588V2 synchronization can be used. In practical applications, Network Time Protocol (NTP) synchronization is generally only used for services with low synchronization accuracy requirements. However, in this implementation, it can also be applied to base station energy consumption measurement, such as power data acquisition in the equipment room. If the base station can only provide NTP time synchronization technology for real-time load feedback, then NTP time synchronization technology can also be used for energy efficiency measurement.

[0066] (3) Optionally, the infrastructure needs to be internally designed with power detection circuitry capable of detecting the power of the entire unit. This could include integrated voltage, current detection, or other power detection circuitry or hardware modules.

[0067] (4) Optionally, the infrastructure can detect ambient temperature.

[0068] 2. Design of the data center where the infrastructure is located:

[0069] (1) Optionally, the power supply facilities in the computer room have load power detection capabilities and can acquire load data while recording the time when the data is generated. It is worth noting that the power supply facilities in the computer room must have either load power detection capabilities or the power detection circuit is designed into the infrastructure.

[0070] (2)Optionally, the computer room has an ambient temperature monitoring function.

[0071] 3. Unit 1:

[0072] The first unit issues control instructions to the infrastructure being measured, specifying the requirements for reporting timestamped data content, as well as information such as detection and reporting time and frequency.

[0073] The first unit issues control commands to the computer room facilities, specifying the data content requirements with time information to be reported, as well as the detection and reporting time, number of times, and other information.

[0074] The first unit can be any functional unit or device that has deployed a functional control software module that supports the method described in any of the above embodiments, such as in the core network, the infrastructure being measured, or other equipment in the computer room where the infrastructure is located.

[0075] 4. Unit Two:

[0076] The second unit is responsible for deploying tasks to the infrastructure and managing the task lifecycle, while also collecting energy efficiency measurement data calculated by the first unit. The second unit may also have other functions, which are not limited here.

[0077] Based on the above system architecture, the specific energy efficiency measurement process is as follows:

[0078] 1. The second unit learns the status of the task where the infrastructure required to perform the computing task is already running normally;

[0079] 2. After the second unit learns that the remaining computing resources of the infrastructure can meet the deployment of the specified task, it deploys the task to the infrastructure and starts running it, and sends a control command carrying the infrastructure identifier to the first unit.

[0080] 3. The first unit issues control commands to the infrastructure, requesting the reporting of information such as the infrastructure's task load and power.

[0081] (1) In this control instruction, the task load information to be reported at a specific time and / or at a specific time interval is specified, and the number of times to report periodically is specified.

[0082] (2) Optionally, the control command may specify the reporting time and / or the power information obtained from internal measurements at a specific time interval, and specify the number of times to report periodically.

[0083] (3) Optionally, the control instruction may specify the reporting of ambient temperature information detected by the infrastructure at specific times and / or periods, and specify the number of times to report periodically.

[0084] 4. The first unit sends control commands to the management node of the data center where the infrastructure is located, requesting the data center to report information such as the output power and ambient temperature of the power supply to the infrastructure being measured.

[0085] (1) Optionally, the control instruction specifies the reporting of output power information for powering the infrastructure at specific times and / or periods, and specifies the number of reporting periods (it is understood that the power information can be reported by the computer room or by the infrastructure).

[0086] (2) Optionally, the control instruction specifies the reporting of ambient temperature information at the location of the computer room where the infrastructure is located at a specific time and / or period, and specifies the number of reporting cycles (it can be understood that the ambient temperature information can be reported by the computer room or by the infrastructure).

[0087] 5. The first unit calculates the energy efficiency of the task on the specified infrastructure based on the task load information and power information at the same time.

[0088] To make the specific process of the above embodiments clearer, a specific example is given below:

[0089] Taking 5G wireless base stations as an example:

[0090] (1) The second unit issues the deployment computing power task to the base station infrastructure (which has no other task processing) that can deploy the computing power task to be measured and starts running it. If it is already running, proceed directly to the second step.

[0091] (2) The second unit notifies the first unit that the infrastructure is ready for measurement.

[0092] (3) The first unit issues an energy efficiency measurement request according to the format in Table 1, and controls the base station to periodically report measurement data parameters, including real-time task load, real-time power, and ambient temperature. (These information can be reported by the infrastructure after self-inspection and / or by the equipment room after inspection.) Among them, real-time power can be reported in the format in Table 2.

[0093] (4) Task load information is determined by key cell parameters. For example, the RB utilization rate of the cell supported by the infrastructure is used for normalized statistics. 100% RB utilization rate is normalized to 100 (this data is only an example, and the specific value is determined according to the actual situation, and is not limited here). Assuming that the infrastructure supports 4 cells, and the RB utilization rates of each cell are a%, b%, c%, d%, and the power of the infrastructure is A kilowatts (kW), then the normalized energy efficiency is (a%+b%+c%+d%) / (A). Optionally, the key cell parameters in the task load information include, but are not limited to: RB utilization rate, number of users, modulation order, cell bandwidth, etc.

[0094] (5) When multiple infrastructures are used for deployment, the energy efficiency of different infrastructures is statistically analyzed and stored during task execution. The number of tests for statistical analysis can be set to n.

[0095] (6) The average energy efficiency of each infrastructure is obtained by averaging or weighted averaging the energy efficiency of n data. The infrastructure is then classified according to its energy efficiency or the average energy efficiency is directly used for horizontal energy efficiency measurement.

[0096] (7) Based on the infrastructure model and data center conditions, select the infrastructure with higher energy efficiency for subsequent deployment of similar new tasks. That is, under the same conditions, select the infrastructure with higher energy efficiency for task deployment.

[0097] The data format is as follows:

[0098] Table 1

[0099]

[0100] In Table 1, “M” indicates mandatory, “O” indicates optional, and “0..15” indicates 0 to 15.

[0101] Table 2

[0102]

[0103] In Table 2, “M” indicates mandatory, “O” indicates optional, “0..8” indicates 0 to 8, “0..15” indicates 0 to 15, “0..32” indicates 0 to 32, and “0..400” indicates 0 to 400.

[0104] Table 3

[0105]

[0106] In Table 3, "M" indicates mandatory, "O" indicates optional, "1..<maximum number of measurement result records>" represents 1 to the maximum number of measurement result records (this value is set according to the actual situation), "1..<maximum measurement value>" represents 1 to the maximum measurement value (this value is set according to the actual situation), "0..<maximum number of measurement information>" represents 0 to the maximum number of measurement information (this value is set according to the actual situation), and "1..<maximum label length>" represents 0 to the maximum label length (this value is set according to the actual situation).

[0107] Table 4

[0108]

[0109] In Table 4, "M" indicates mandatory.

[0110] Compared with existing technologies, the method provided in this embodiment, after the infrastructure is deployed with specific tasks (including but not limited to wireless communication services, computing power service tasks, or other decomposed sub-tasks), controls and acquires the real-time load status of the infrastructure that can share computing power, and obtains power information at the same time for accurate online energy efficiency measurement. This method, by conducting task-level measurements online and adding timestamps to report key cell parameters, accurately assesses energy efficiency by associating power and load in real time. This effectively solves the problem of decreased measurement accuracy caused by the asynchronous acquisition time of power measurement results when the task load changes dynamically in real time during the operation of products such as wireless base stations. The binding relationship between infrastructure and tasks, and heterogeneous system architecture, do not affect the accuracy of the load and power information acquired in this embodiment. It can accurately measure the energy efficiency of the infrastructure, and this embodiment does not require offline measurement processing through simulation testing, thus not consuming a large amount of infrastructure resources and has broad application prospects.

[0111] See Figure 3 , Figure 3 This is a schematic diagram of the structure of an energy efficiency measuring device provided in an embodiment of the present invention. The energy efficiency measuring device 20 includes:

[0112] Command receiving module 21 is used to receive control commands carrying infrastructure identifiers;

[0113] The information acquisition module 22 is used to acquire the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped.

[0114] The energy efficiency measurement module 23 is used to calculate the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp.

[0115] In one implementation, the task load information includes at least one cell key parameter defined by 3GPP that relates to the processing overhead of the cell being processed by the infrastructure.

[0116] In one embodiment, the energy efficiency measurement module 23 is specifically used for:

[0117] Calculate the normalized load based on the cell key parameters and their corresponding set weights in the task load information;

[0118] The energy efficiency of the infrastructure is obtained by dividing the normalized load at the same timestamp by the power information.

[0119] In one implementation, the task load information includes at least one of RB utilization, number of users, modulation order, and cell bandwidth.

[0120] In one implementation, when the deviation between the timestamp of the task load information and the timestamp of the power information is less than a set deviation threshold, the two timestamps are considered to be the same timestamp.

[0121] In one implementation, the power information is detected by a power detection circuit designed into the infrastructure, or by the power supply facilities of the computer room where the infrastructure is located.

[0122] In one embodiment, the device further includes an average calculation module for: performing a weighted average calculation of the energy efficiency of the infrastructure at multiple timestamps to obtain the average energy efficiency of the infrastructure.

[0123] In one embodiment, the device further includes a device selection module for:

[0124] From among multiple infrastructures with the same infrastructure model and data center conditions, the infrastructure with higher average energy efficiency is selected as the preferred choice for new deployments of the same type of task.

[0125] In one implementation, the power information is obtained from the infrastructure or the data center where the infrastructure is located.

[0126] It is worth noting that the specific working process of the energy efficiency measurement device can refer to the working process of the energy efficiency measurement method in any of the above embodiments, and will not be repeated here.

[0127] Compared with the prior art, the energy efficiency measurement device disclosed in this embodiment of the invention obtains the power information and task load information of the infrastructure corresponding to the infrastructure identifier when it is running a task by receiving a control command carrying an infrastructure identifier, wherein the power information and the task load information are timestamped; by calculating the energy efficiency of the infrastructure based on the power information and the task load information with the same timestamp, the accuracy of the computing power energy efficiency measurement of the infrastructure is improved.

[0128] See Figure 4 , Figure 4 This is a schematic diagram of the structure of an energy efficiency measurement device provided in an embodiment of the present invention. The energy efficiency measurement device 30 includes a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, it implements the steps as described in the above-described energy efficiency measurement method embodiment, for example... Figure 1 The steps S11 to S13 described above; or, when the processor 31 executes the computer program, it implements the functions of each module in the above-described device embodiments.

[0129] For example, the computer program can be divided into one or more modules, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the energy efficiency measurement device. For example, the computer program can be divided into multiple modules, each with the following specific functions:

[0130] Command receiving module 21 is used to receive control commands carrying infrastructure identifiers;

[0131] The information acquisition module 22 is used to acquire the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped.

[0132] The energy efficiency measurement module 23 is used to calculate the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp.

[0133] The specific working process of each module can be referred to the working process of the energy efficiency measurement device described in the above embodiments, and will not be repeated here.

[0134] The energy efficiency measurement device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The energy efficiency measurement device may include, but is not limited to, a processor 31 and a memory 32. Those skilled in the art will understand that the energy efficiency measurement device may also include input / output devices, network access devices, buses, etc.

[0135] The processor 31 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 31 is the control center of the energy efficiency measurement device, connecting all parts of the device via various interfaces and lines.

[0136] The memory 32 can be used to store the computer program and / or modules. The processor 31 implements various functions of the energy efficiency measurement device by running or executing the computer program and / or modules stored in the memory 32 and calling the data stored in the memory 32. The memory 32 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory 32 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (SMC), secure digital (SD) card, flash memory card, at least one magnetic energy efficiency measurement storage device, flash memory device, or other volatile solid-state storage device.

[0137] If the module integrated into the energy efficiency measurement device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by the processor 31, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0138] This invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the energy efficiency measurement method as described in any of the above embodiments, such as steps S11 to S13.

[0139] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An energy efficiency measurement method, characterized in that, include: Receive control commands carrying infrastructure identifiers; Obtain the power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein the power information and the task load information are timestamped, and the power information is power; The energy efficiency of the infrastructure is calculated based on the power information and the task load information at the same timestamp; the infrastructure is a wireless infrastructure. The task load information includes at least one key cell parameter defined by 3GPP that is related to the processing overhead of the cell being processed by the infrastructure; the task load information includes RB utilization, number of users, modulation order, and cell bandwidth; The step of calculating the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp includes: Calculate the normalized load based on the cell key parameters and their corresponding set weights in the task load information; The energy efficiency of the infrastructure is obtained by dividing the normalized load at the same timestamp by the power information. When the deviation between the timestamp of the task load information and the timestamp of the power information is less than a set deviation threshold, the two timestamps are considered to be the same timestamp.

2. The energy efficiency measurement method as described in claim 1, characterized in that, The power information is obtained by a power detection circuit designed inside the infrastructure, or by the power supply facilities in the computer room where the infrastructure is located.

3. The energy efficiency measurement method as described in claim 1 or 2, characterized in that, Also includes: The average energy efficiency of the infrastructure is obtained by weighted averaging the energy efficiency of the infrastructure at multiple timestamps.

4. The energy efficiency measurement method as described in claim 3, characterized in that, Also includes: From among multiple infrastructures with the same infrastructure model and data center conditions, the infrastructure with higher average energy efficiency is selected as the preferred choice for new deployments of the same type of task.

5. The energy efficiency measurement method as described in claim 1, characterized in that, The method is applied to the first unit; the control command is sent by the second unit to the first unit after the deployment and operation of the task on the infrastructure is initiated; the power information is obtained by the first unit from the infrastructure or the computer room where the infrastructure is located.

6. An energy efficiency measuring device, characterized in that, include: The command receiving module is used to receive control commands carrying infrastructure identifiers. An information acquisition module is used to acquire power information and task load information of the infrastructure corresponding to the infrastructure identifier; wherein, the power information and the task load information are timestamped; the task load information includes at least one cell key parameter defined by 3GPP related to the processing overhead of the cell processed by the infrastructure; the infrastructure is a wireless infrastructure; the power information is power, and the task load information includes RB utilization, number of users, modulation order, and cell bandwidth; An energy efficiency measurement module is used to calculate the energy efficiency of the infrastructure based on the power information and the task load information at the same timestamp; wherein, when the deviation between the timestamp of the task load information and the timestamp of the power information is less than a set deviation threshold, the timestamps of the two are considered to be the same timestamp. The energy efficiency measurement module is specifically used for: Calculate the normalized load based on the cell key parameters and their corresponding set weights in the task load information; The energy efficiency of the infrastructure is obtained by dividing the normalized load at the same timestamp by the power information.

7. An energy efficiency measurement device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the energy efficiency measurement method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the energy efficiency measurement method as described in any one of claims 1 to 5.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the energy efficiency measurement method as described in any one of claims 1 to 5.

Images