CN110308782B

CN110308782B - Power consumption prediction and control method and device and computer readable storage medium

Info

Publication number: CN110308782B
Application number: CN201810241512.5A
Authority: CN
Inventors: 杨波; 陈立波; 刘毅
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2018-03-22
Filing date: 2018-03-22
Publication date: 2023-09-08
Anticipated expiration: 2038-03-22
Also published as: CN110308782A

Abstract

The embodiment of the application provides a power consumption prediction and control method, equipment and a computer readable storage medium. In this embodiment, a device parameter that changes with a change in the power consumption adjustment parameter and has a change sensitivity greater than a corresponding threshold value is adopted as the power consumption prediction parameter; and carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters to obtain a predicted power consumption value of the equipment to be processed, so that the predicted power consumption value has higher accuracy, and further carrying out power consumption control according to the predicted power consumption value with higher accuracy, thereby improving the power consumption control effect.

Description

Power consumption prediction and control method and device and computer readable storage medium

Technical Field

The present application relates to the field of electronic technologies, and in particular, to a power consumption prediction and control method, apparatus, and computer readable storage medium.

Background

With the progress of technology, the computing capability of electronic devices is continuously improved, and functions are continuously expanded, which increases the power consumption of the electronic devices to some extent. Too high power consumption will reduce the service performance of the electronic device, so it is necessary to control the power consumption of the electronic device, to reduce the energy consumption, prolong the service life of the battery, improve the stability of the electronic device, and so on.

In the prior art, power consumption of an electronic device is generally predicted by using a central processing unit (Central Processing Unit, CPU) utilization of the electronic device, and then power consumption control is performed on the electronic device based on the predicted power consumption. However, the existing power consumption prediction results are not accurate enough, so that the power consumption control effect is not ideal.

Disclosure of Invention

Aspects of the embodiments of the present application provide a power consumption prediction method, a power consumption control device, and a computer readable storage medium, which are used for improving accuracy of power consumption prediction and providing a foundation for improving a power consumption control effect.

The embodiment of the application provides a power consumption prediction method, which comprises the following steps:

acquiring power consumption prediction parameters from equipment parameters of equipment to be processed; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has a change sensitivity larger than a corresponding threshold value;

and carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the equipment to be processed.

The embodiment of the application also provides electronic equipment, which comprises: a memory and a processor;

the memory is used for storing a computer program;

the processor, coupled to the memory, is configured to execute the computer program for:

Acquiring power consumption prediction parameters from equipment parameters of the electronic equipment; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has a change sensitivity larger than a corresponding threshold value;

The embodiment of the application also provides a computer readable storage medium storing a computer program, and the computer program can realize the steps in the power consumption prediction method provided by the embodiment of the application when being executed.

The embodiment of the application also provides a power consumption control method, which comprises the following steps:

carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the equipment to be processed;

and adjusting the power consumption adjusting parameter according to the predicted power consumption value so as to control the power consumption of the equipment to be processed.

The memory is used for storing a computer program;

The embodiment of the application also provides a computer readable storage medium storing a computer program, and the computer program can realize the steps in the power consumption control method provided by the embodiment of the application when being executed.

In the embodiment of the application, the predicted power consumption value of the equipment to be processed is predicted by adopting the equipment parameter which can change along with the change of the power consumption adjustment parameter and has higher change sensitivity, so that the predicted power consumption value has higher accuracy, and further, when the power consumption control is performed based on the predicted power consumption value with higher accuracy, the power consumption control effect is favorably improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:

FIG. 1a is a flowchart illustrating a power consumption prediction method according to an exemplary embodiment of the present application;

FIG. 1b is a flowchart illustrating a power consumption control method according to an exemplary embodiment of the present application;

FIG. 2 is a flow chart of another power consumption control method according to another exemplary embodiment of the present application;

FIG. 3 is a flow chart of a method for controlling power consumption according to another exemplary embodiment of the present application;

FIG. 4a is an internal block diagram of a device to be treated provided by a further exemplary embodiment of the present application;

FIG. 4b is a schematic diagram illustrating an interaction flow of the modules in the block diagram of FIG. 4a according to another exemplary embodiment of the present application;

FIG. 4c is a schematic diagram of a power consumption control system according to another exemplary embodiment of the present application;

fig. 5 is a schematic structural diagram of an electronic device according to another exemplary embodiment of the present application;

fig. 6 is a schematic structural diagram of another electronic device according to another exemplary embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Aiming at the technical problem that the existing power consumption prediction result is inaccurate, in some exemplary embodiments of the application, the device parameter which can change along with the change of the power consumption adjustment parameter and has higher change sensitivity is used as the power consumption prediction parameter to predict the predicted power consumption value of the device to be processed, so that the predicted power consumption value has higher accuracy. Further, when power consumption control is performed based on a predicted power consumption value with higher accuracy, it is advantageous to improve the power consumption control effect.

The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

Fig. 1a is a flowchart illustrating a power consumption prediction method according to an exemplary embodiment of the present application. As shown in fig. 1a, the method comprises:

101. And acquiring a power consumption prediction parameter from the device parameters of the device to be processed, wherein the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has the change sensitivity larger than a corresponding threshold value.

102. And carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the equipment to be processed.

In the present embodiment, an electronic device capable of generating power consumption during operation and for which power consumption prediction is necessary is referred to as a device to be processed, and may be any electronic device capable of generating power consumption during operation.

In some alternative embodiments, the device to be processed may be a terminal device such as a smart phone, tablet, personal computer, wearable device, etc. The device to be processed may comprise at least one processing unit and at least one memory. The number of processing units and memories depends on the configuration and type of the terminal device. The Memory may include volatile such as RAM, nonvolatile such as Read-Only Memory (ROM), flash Memory, or the like, or both. The memory typically stores an Operating System (OS), one or more application programs (apps), program data, and the like. In addition to the processing unit and the memory, the device to be processed also includes basic configuration such as a network card chip, an IO bus, an audio/video component, and the like. Optionally, depending on the implementation of the device to be processed, the device to be processed may also include some peripheral devices, such as a keyboard, mouse, stylus, printer, etc. Such peripheral devices are well known in the art and are not described in detail herein.

In other alternative embodiments, the device to be processed may be any server device with a certain computing power, for example, a conventional server, a cloud host, a virtual center, etc. The server mainly comprises a processor, a hard disk, a memory, a system bus and the like, and is similar to a general computer architecture.

The device to be processed has some device parameters, which may include software related parameters, hardware related parameters, and some performance parameters, etc. Among these device parameters, some are related to the power consumption of the device to be processed. Embodiments of the present application focus mainly on parameters related to power consumption of a device to be processed, including, for example, parameters that cause a change in power consumption and/or parameters that are affected by a change in power consumption.

In some application scenarios, power consumption control of a device to be processed is required. The power consumption control process can be divided into two phases: a power consumption prediction phase and a power consumption adjustment phase. The power consumption prediction stage is mainly responsible for predicting the power consumption of the equipment to be processed and providing data support for the power consumption adjustment stage. The power consumption adjusting stage is mainly responsible for changing the power consumption of the equipment to be processed to an expected direction by adjusting related parameters according to the power consumption value predicted in the power consumption predicting stage, so as to achieve the purpose of power consumption control.

The prediction result of the power consumption prediction stage is the basis of power consumption control, and the accuracy of the power consumption prediction result seriously influences the power consumption control effect. The embodiment provides a power consumption prediction method, which is used for improving the accuracy of a power consumption prediction result. In the power consumption prediction process of the present embodiment, power consumption prediction may be performed on a device to be processed based on some device parameters of the device to be processed. In order to improve accuracy of the prediction result, the embodiment combines parameters (simply referred to as power consumption adjustment parameters) to be adjusted in the power consumption adjustment stage, and selects some device parameters which are sensitive to changes of the power consumption adjustment parameters from the device parameters of the device to be processed as power consumption prediction parameters for predicting power consumption of the device to be processed.

The power consumption prediction parameters in the embodiment belong to equipment parameters of equipment to be processed, and can reflect the power consumption characteristics of the equipment to be processed to a certain extent. In addition, the power consumption prediction parameters and the power consumption adjustment parameters in the embodiment can be mutually influenced and are relatively sensitive to the change of each other, namely, once the power consumption adjustment parameters are changed, the power consumption prediction parameters can quickly generate more obvious change, accordingly, the change of the power consumption prediction parameters can be timely reflected to the power consumption adjustment parameters, and the accuracy of a power consumption prediction result can be improved by adopting the power consumption prediction parameters with the sensitive characteristic.

In the present embodiment, in order to show the sensitivity of the power consumption prediction parameter that changes with the power consumption adjustment parameter, the threshold value corresponding to each parameter is set in advance. Based on the above, for a certain device parameter, whether the sensitivity of the device parameter along with the change of the power consumption adjustment parameter is larger than a corresponding threshold value can be judged; when the variation sensitivity is greater than the corresponding threshold, the determination of the device parameter may be used as the power consumption prediction parameter in the present embodiment to predict the power consumption of the device to be processed. Alternatively, the sensitivity of each device parameter to the power consumption adjustment parameter may be obtained in advance from an empirical value.

After the power consumption prediction parameter is acquired, a power consumption value of the device to be processed may be predicted according to the power consumption prediction parameter. For convenience of distinction and description, the predicted power consumption value of the device to be processed is referred to herein as a predicted power consumption value.

In some application scenarios, after obtaining the predicted power consumption value, power consumption control may be performed on the device to be processed according to the predicted power consumption value. The flow of the power consumption control method is shown in fig. 1b, and after step 102, the method further includes:

103. and adjusting the power consumption adjusting parameter according to the predicted power consumption value to control the power consumption of the equipment to be processed.

Optionally, if the predicted power consumption value is larger, which indicates that the current power consumption of the device to be processed is larger, the power consumption adjustment parameter can be adjusted to reduce the power consumption of the device to be processed; if the predicted power consumption value is smaller, the current power consumption of the equipment to be processed is smaller, and the power consumption of the equipment to be processed can be increased by adjusting the power consumption adjustment parameter. It should be noted that, increasing the power consumption of the device to be processed generally improves the processing performance of the device to be processed.

In the above embodiment, the parameter which changes along with the change of the power consumption adjustment parameter and has higher sensitivity of change is adopted as the power consumption prediction parameter, and then the power consumption value of the device to be processed is predicted according to the power consumption prediction parameter, so that the predicted power consumption value has higher accuracy. Further, when power consumption control is performed based on a predicted power consumption value with higher accuracy, it is advantageous to improve the power consumption control effect.

It should be noted that, in many application scenarios related to power consumption, power consumption prediction is required, and the power consumption prediction method provided by the embodiment of the application is applicable to various application scenarios with power consumption prediction requirements. The following is illustrative:

for example, in one type of scene, a projector projects in an environment with variable brightness. In the scene, the intensity of the projection light of the projector can be flexibly adjusted by combining the ambient brightness, so that the purpose of saving power consumption is achieved. For example, when the ambient brightness is high, the intensity of the projection light can be reduced, and the power consumption is reduced as much as possible on the basis of ensuring the definition of the projection picture; when the ambient brightness is lower, the intensity of projection light can be improved, and the definition of the projection picture is preferentially ensured. In this scenario, the projector may be regarded as a device to be processed in the embodiments of the present application; accordingly, the brightness of the projected light is used as a power consumption adjustment parameter. Based on this, in the process of performing power consumption control on the projector, projector parameters that may vary greatly with the variation in brightness of the projected light, such as the bulb irradiation time period, the current magnitude, and the like, may be selected as power consumption prediction parameters; obtaining a predicted power consumption value of the projector based on the power consumption prediction parameters; further, a theoretical power consumption threshold value can be determined by combining the current environment brightness, and the brightness of the projection light is adjusted according to the difference value between the predicted power consumption value and the corresponding theoretical power consumption threshold value, so that the power consumption of the projector is adjusted to the theoretical power consumption threshold value, and the power consumption is reduced as much as possible under the condition that the definition of a projection picture is ensured.

For another example, in another application scenario, multiple servers are deployed in a machine room, rack, or cabinet, and the deployment density of the servers may need to be adjusted to meet the service requirements, such as capacity expansion. When the deployment density of the server is adjusted, in order to ensure that the overall power consumption of the machine room, the rack or the cabinet is still within a reasonable power consumption range, the power consumption of the server may need to be adjusted. For example, in the case of increasing the deployment density of servers, the number of servers in the entire machine room, rack, or cabinet may increase, and in the case of ensuring that the overall power consumption of the machine room, rack, or cabinet is unchanged, it is necessary to reduce the power consumption of some or all of the existing servers. At this time, the existing server needing to reduce the power consumption is the equipment to be processed; accordingly, parameters such as the CPU frequency of the existing server can be used as the power consumption adjustment parameters. Based on the above, in the power consumption control process, device parameters with larger changes along with the changes of the power consumption adjustment parameters can be selected, for example, memory temperature, CPU temperature, memory utilization rate, read-write operation number of the disk per second and the like can be used as power consumption prediction parameters; predicting the power consumption value of the corresponding existing server by using the power consumption prediction parameters; furthermore, the CPU frequency of the corresponding existing server can be respectively adjusted according to the maximum allowable power consumption value of each server after the deployment density is improved and the predicted power consumption value of the corresponding existing server, so that the purpose of reducing the power consumption of the corresponding existing server to the maximum allowable power consumption value is achieved. Correspondingly, for the situation that the deployment density of the servers needs to be reduced, the number of the servers in the whole machine room, the rack or the cabinet can be reduced, namely, part of the servers can be removed, and under the condition that the whole power consumption of the machine room, the rack or the cabinet is unchanged, the rest or all of the servers are allowed to have more power consumption, and the power consumption control is also needed, wherein the power consumption prediction process is similar and is not repeated.

For another example, in other scenes, for the servers in the same cabinet, rack or machine room, the method provided by the embodiment of the application can also be adopted to predict the current power consumption value of the servers in real time, and further power consumption control can be performed when the predicted power consumption value reaches a certain threshold value, so as to ensure electricity safety.

In some example embodiments, the power consumption adjustment parameter may be determined from device parameters of the device to be processed before the device to be processed is power consumption predicted. In these exemplary embodiments, the power consumption adjustment parameter, like the power consumption prediction parameter, also belongs to the device parameter of the device to be processed.

In different application scenarios, the power consumption adjustment parameters may be different. For example, in some application scenarios, the load quantity of the device to be processed may be used as a power consumption adjustment parameter, so that the purpose of controlling the power consumption of the device to be processed within a reasonable range may be achieved by adjusting the load quantity of the device to be processed. For example, in other application scenarios, the fan speed of the device to be processed may be used as the power consumption adjustment parameter, so that the purpose of controlling the power consumption of the device to be processed within a reasonable range may be achieved by adjusting the fan speed of the device to be processed. The power consumption adjusting parameters are different, and the power consumption prediction parameters which are changed along with the change of the power consumption adjusting parameters and have the change sensitivity larger than the corresponding threshold value are also different.

In some exemplary embodiments of the present application, the CPU frequency may be taken as a power consumption adjustment parameter in consideration of an association relationship between the CPU frequency and power consumption. The CPU frequency is the frequency of the CPU's system clock, i.e., the number of synchronous pulses that occur per second when the CPU is operating, in Hz. The CPU frequency can determine the operation speed of the CPU, and in general, the higher the CPU frequency, the faster the operation speed of the CPU.

The CPU is the main core component in the device to be processed. The CPU is a digital integrated circuit, and may be manufactured by using Metal-Oxide-Semiconductor (MOS) technology, such as complementary Metal-Oxide-Semiconductor (Complementary Metal Oxide Semiconductor, CMOS), N-channel-Metal-Oxide-Semiconductor (NMOS), and P-channel-Metal-Oxide-Semiconductor (Positive channel Metal Oxide Semiconductor, PMOS) technology. Such digital integrated circuits operate under control of a system clock, with internal gates changing state every clock cycle. A large current flows each time the gate changes state. The shorter the clock period, the higher the frequency with which the gate changes state, and correspondingly the higher the frequency with which large currents flow, resulting in increased power consumption. The clock period is the inverse of the frequency, that is, the higher the frequency of the CPU, the shorter the clock period, the greater the average current it consumes and the higher the power consumption.

In the exemplary embodiment having the CPU frequency as the power consumption adjustment parameter, in the power consumption prediction process, a device parameter other than the CPU frequency that can be changed with a change in the CPU frequency and whose change sensitivity is greater than the corresponding threshold value may be selected as the power consumption prediction parameter in the power consumption prediction stage.

Among the device parameters of the device to be processed, there are a number of device parameters whose change sensitivity is larger than the corresponding threshold (simply referred to as device parameters that are relatively sensitive to the change in the CPU frequency) that can be changed with the change in the CPU frequency. Some exemplary parameters are given below:

register parameters:

in the special module register (Model Specific Registers, MSR) of the device to be processed, some device parameters related to the key components of CPU, memory, etc. of the device to be processed are stored. These device parameters are sensitive to changes in CPU frequency, and when the CPU frequency changes, these parameters also change in time. For ease of description, the parameters stored in the MSR are referred to as register parameters. Based on this, at least one register parameter may be read from the MSR of the device to be processed as a power consumption prediction parameter. The following list several register parameters:

(1) Temperature type parameters in MSR:

in the MSR of the device to be treated, temperature parameters of some components are included, such as temperature values per second. In general, the temperature of each component on the device to be processed can reflect the power consumption of the device to be processed to a certain extent, and in addition, the temperature parameters of the components in the MSR are sensitive to the change of the CPU frequency, so that the temperature parameters can be used as the power consumption prediction parameters in various embodiments of the application.

For example, temperature parameters of components such as a CPU, a memory, a solid state disk (Solid State Drives, SSD) and the like of the device to be processed may be read from the MSR of the device to be processed. For components having a housing, such as an SSD, the temperature parameter may include a housing temperature and/or an internal temperature.

In addition, the fan of the device to be processed is also a component of the device to be processed, and the rotating speed of the fan is taken as an indirect parameter related to the temperature parameter, so that the temperature of the fan can be reflected, and the fan can be regarded as a special temperature type parameter and can also be taken as a power consumption prediction parameter to carry out power consumption prediction.

(2) Power consumption type parameters in MSR:

in the MSR of the device to be processed, a power consumption parameter of some components, such as a power consumption value per second, is also included. In general, the power consumption of each component on the device to be processed may directly reflect the power consumption of the device to be processed, and in addition, the power consumption parameters of these components in the MSR are sensitive to the change of the CPU frequency, so they may be used as the power consumption prediction parameters in the embodiments of the present application.

For example, power consumption parameters of components such as a CPU, a memory, a dynamic random access memory (Dynamic Random Access Memory, DRAM) and the like of the device to be processed may be read from the MSR of the device to be processed. Wherein, the components are different, and the implementation modes of the power consumption parameters are also different. For example, in an MSR, the power consumption value of the CPU may be read, or the running average power limit (Running Average Power Limit, RAPL) value of the CPU may be read, which RAPL value may represent the CPU's power consumption. In addition, in the MSR, the dc power consumption value of some other components may be read, and power consumption type parameters such as power supply efficiency and power supply time of some other components may be read, where the power consumption value may be indirectly reflected.

In the above-described embodiment, at least one register parameter may be read from the MSR of the device to be processed as a power consumption prediction parameter that varies with CPU frequency variation and has a variation sensitivity greater than a set threshold. This way of reading register parameters from the MSR has high stability and high accuracy. In addition, the method for reading the parameters has low resource consumption and low business influence.

System parameters:

the device to be processed has some system parameters, such as CPU utilization, memory utilization, disk Input/Output (I/O) performance parameters, and network card I/O performance parameters, etc., which are typically stored in a system file of the device to be processed. Among these system parameters, there are some system parameters that are relatively sensitive to the change of the CPU frequency, and when the CPU frequency changes, these system parameters that are relatively sensitive to the change of the CPU frequency also change in time. Based on this, at least one system parameter may be read from a system file of the device to be processed as a power consumption prediction parameter.

In one embodiment, in order to improve accuracy of the power consumption prediction result, in consideration of relatively low sensitivity of the CPU utilization as a function of the CPU frequency, other system parameters, such as memory utilization, disk Input/Output (I/O) performance parameters, and network card I/O performance parameters, may be selected as the power consumption prediction parameters.

In another embodiment, the sensitivity of CPU utilization as a function of CPU frequency is considered to be relatively low, but the adverse effect of CPU utilization on the accuracy of the power consumption prediction result is reduced by reducing the CPU utilization's importance ratio in the power consumption prediction. Based on this, the CPU utilization and other system parameters may be selected as the power consumption prediction parameters, but considering that the sensitivity of the CPU utilization with the CPU frequency is relatively low, it is required that the importance duty ratio of the CPU utilization in the power consumption prediction is not the maximum and is smaller than the preset duty ratio threshold. The importance of CPU utilization in power consumption prediction and the corresponding duty cycle threshold may depend on the application requirements, and it is generally desirable to reduce the importance of CPU utilization to a level that has little effect on the accuracy of the power consumption prediction results.

The disk I/O performance parameters described above include, but are not limited to: the number of I/O operations per second (Input/Output Operations Per Second, iops), the average latency per I/O operation (await), the number of bits per second (bps) transferred per second, the number of read I/O devices per second completed (r/s), the number of write I/O devices per second completed (w/s), the time duty (% util) the I/O queue is in a non-empty state in one second, and the like.

The network card I/O performance parameters include, but are not limited to: the network card has parameters such as the number of transmission bits (tx_bytes), the number of reception bits (rx_bytes), the transmission packet loss rate (tx_drop), the reception packet loss rate (rx_drop), the frequency of receiving data (rx_fifo) in the reception first-in-first-out queue, the frequency of transmitting data (tx_fifo) in the transmission first-in-first-out queue, the reception packet quantity (rx_packets), and the transmission packet quantity (tx_packets).

In the above embodiment, at least one system parameter may be read from a system file of the device to be processed as the power consumption prediction parameter. The method for reading the system parameters from the system file is fast in speed, and the accuracy of the read parameter values is high.

It should be noted that, when the device to be processed is in the cluster environment, in addition to some parameters related to the device to be processed, some parameters of other devices related to the power consumption of the device to be processed may be obtained as power consumption prediction parameters in the power consumption prediction stage. Alternatively, the other devices associated with the power consumption of the device to be processed may be devices that have a mutual restriction in a set range with the total power consumption of the device to be processed, for example, assuming that the device to be processed is one server in a machine room or a cabinet, the power consumption of other servers located in the same machine room or cabinet may affect the power consumption of the server, and belong to other devices having a power consumption association relationship with the server.

It should be noted that, depending on the application requirements, one or several parameters shown above may be selected to be used as the power consumption prediction parameters. In addition, the same power consumption prediction parameters can be selected and used in different power consumption prediction processes, and the power consumption prediction parameters which are not completely the same can also be selected and used, and the power consumption prediction parameters are specific to application requirements.

In order to facilitate a person skilled in the art to more clearly understand the implementation process of the embodiment of the present application, in the following power consumption control embodiment, the technical solution of the embodiment of the present application is described in detail by taking the power consumption adjustment parameter as an example of the CPU frequency of the device to be processed.

Fig. 2 is a flowchart of another power consumption control method according to another exemplary embodiment of the present application. As shown in fig. 2, the method includes:

200. from the device parameters of the device to be processed, the CPU frequency is determined as a power consumption adjustment parameter.

201. From other device parameters than the CPU frequency, a device parameter which changes with the change of the CPU frequency and whose change sensitivity is larger than a corresponding threshold value is acquired as a power consumption prediction parameter.

202. And carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value.

203. Judging whether the predicted power consumption value is larger than a set power consumption threshold value or not; if yes, go to step 204, otherwise, go to step 205.

204. In the power consumption adjustment stage, the CPU frequency of the device to be processed is reduced to reduce the power consumption of the device to be processed below the set power consumption threshold, and step 205 is entered.

205. Wait to enter the next power consumption prediction phase and return to step 201 when the next power consumption prediction phase arrives.

In the present embodiment, the CPU frequency is determined as the power consumption adjustment parameter, and then the power consumption of the device to be processed is predicted using the power consumption prediction parameter which varies with the variation of the CPU frequency and has a variation sensitivity larger than the corresponding threshold value, to obtain the predicted power consumption value. And then, adaptively adjusting the CPU frequency according to the predicted power consumption value so as to achieve the purpose of controlling the power consumption of the equipment to be processed.

In the present embodiment, the power consumption of the device to be processed is required to be smaller than the set power consumption threshold. Based on this, after obtaining the predicted power consumption value, the predicted power consumption value may be compared with a preset power consumption threshold; and automatically triggering whether to enter a power consumption adjustment stage according to the comparison result. When the predicted power consumption value is larger than the set power consumption threshold value, determining to enter a power consumption adjustment stage, and in the power consumption adjustment stage, reducing the CPU frequency of the equipment to be processed so as to reduce the power consumption of the equipment to be processed below the set power consumption threshold value, thereby achieving the purpose of power consumption control. Otherwise, when the predicted power consumption value is smaller than or equal to the set power consumption threshold value, the power consumption of the equipment to be processed can be not adjusted, and the next power consumption prediction stage is waited to be entered, so that the purpose of monitoring the power consumption of the equipment to be processed according to the requirement is achieved.

The set power consumption threshold herein refers to an expected power consumption value or an upper limit of an allowed power consumption value of the device to be processed, and is not limited to a value, and can be adaptively set according to application scenarios.

It should be noted that, in addition to whether to enter the power consumption adjustment phase or not triggered automatically by the comparison result of the predicted power consumption value and the preset power consumption threshold, whether to enter the power consumption adjustment phase may be triggered by an external instruction. For example, when the predicted power consumption value is greater than the set power consumption threshold value, the external may issue a power consumption adjustment instruction; after receiving the power consumption adjustment instruction from the outside, determining to enter a power consumption adjustment stage, and reducing the CPU frequency of the equipment to be processed in the power consumption adjustment stage so as to reduce the power consumption of the equipment to be processed below a set power consumption threshold value, thereby achieving the purpose of power consumption control.

In this embodiment, the CPU frequency is used as the power consumption adjustment parameter in the power consumption adjustment stage, so that the power consumption adjustment is more direct and effective. In addition, the device with the sensitivity larger than the corresponding threshold value is used for predicting the parameter to obtain the predicted power consumption value of the device to be processed, and the predicted power consumption value is used as the basis for power consumption adjustment, so that the accuracy of power consumption adjustment is higher, and the effect of power consumption adjustment is better.

In embodiments of the present application, power consumption prediction may be performed on a device to be processed according to a power consumption prediction parameter. Alternatively, a power consumption prediction model may be trained in advance, so that after the power consumption prediction parameter is obtained, the power consumption prediction parameter may be used as a model to run the power consumption prediction model, so as to obtain a predicted power consumption value of the device to be processed.

In some exemplary embodiments, considering that the types of the power consumption prediction parameters may be different, for example, some are power consumption type parameters and some are non-power consumption type parameters, in order to facilitate the power consumption prediction, the parameter type conversion may be performed in advance. Based on the above, an embodiment for predicting power consumption of a device to be processed based on a power consumption prediction model includes: converting non-power consumption type parameters in the power consumption prediction parameters into power consumption type parameters according to a preset parameter conversion relation, for example, converting temperature type parameters and non-power consumption type system parameters into power consumption type parameters; and then taking the original power consumption type parameter and the target power consumption type parameter in the power consumption prediction parameters as a model-based operation power consumption prediction model to obtain a predicted power consumption value of the equipment to be processed. In order to facilitate distinguishing between the power consumption type parameter originally contained in the power consumption prediction parameter and the power consumption type parameter converted from the non-power consumption type parameter, the power consumption type parameter converted from the non-power consumption type parameter is referred to as a target power consumption type parameter, and the power consumption type parameter originally contained in the power consumption prediction parameter is referred to as an original power consumption type parameter.

Alternatively, for the energy consumption components in the device to be processed, the correspondence relationship between the temperature and the power consumption thereof may be established in advance. Taking a certain type of server as an example, the rated power consumption range of a CPU is 65-155W, the rated power consumption range of a memory is 4-7W, the rated power consumption range of a hard disk is 15-35W, the rated power consumption range of a network card is 2-25W, and the rated power consumption range of a fan is 5-35W. In this embodiment, the power consumption of the above-described components at different temperatures may be measured in advance, for example, when the temperature of the CPU is about 23 degrees, the power consumption thereof is 65W; at 30 degrees, the power consumption is 70W; at 50 degrees, the power consumption is 155W. For another example, when the temperature of the hard disk is 30 degrees, the power consumption thereof is 15W; at 55 degrees, the power consumption was 35W. The embodiment can perform multiple measurements, and based on the measurement results, establish the conversion relation between the temperature and the power consumption of different energy consumption components. The conversion relation can be expressed by a conversion function, and after the temperature parameter is obtained, the target power consumption type parameter corresponding to the temperature parameter can be obtained according to the conversion function.

Similarly, the conversion functions of the performance parameters and the power consumption parameters of other types can be established in the above manner, and the corresponding conversion is performed after the performance parameters are obtained, which is not repeated.

The power consumption prediction model adopted in the above embodiment may be obtained by training in advance according to the power consumption prediction parameter sample and the actual power consumption value of the device to be processed acquired by the BMC. In the power consumption prediction model, corresponding model parameters (including original power consumption type parameters and converted target power consumption type parameters) are allocated in advance for each power consumption prediction parameter, for example, weight values, wherein the model parameters can represent importance ratios of different power consumption prediction parameters in power consumption prediction, and the importance ratios represent contribution amounts of the power consumption prediction parameters to predicted power consumption values of equipment to be processed. The higher the sensitivity of a power consumption prediction parameter to changes in a power consumption adjustment parameter (e.g., CPU frequency), the greater the importance of the power consumption prediction parameter in power consumption prediction. Model parameters corresponding to different power consumption prediction parameters can be preset according to experience values, and can also be obtained by model self-adaptive learning.

It should be noted that, whether or not the above power consumption prediction model is adopted, the power consumption prediction of the device to be processed by using the power consumption prediction parameter may be: and combining the power consumption prediction parameters and the weight values corresponding to the power consumption prediction parameters to perform model prediction on the equipment to be processed. The magnitude of the weight value corresponding to the power consumption prediction parameter is in direct proportion to the change sensitivity of the power consumption prediction parameter along with the power consumption adjustment parameter, namely, the higher the change sensitivity of the power consumption prediction parameter along with the power consumption adjustment parameter is, the larger the corresponding weight value is; conversely, the lower the sensitivity of the power consumption prediction parameter to changes in the power consumption adjustment parameter, the smaller the corresponding weight value.

Further, on the basis of the power consumption prediction model, the power consumption prediction model can be continuously corrected along with the power consumption prediction process, so that the prediction precision of the power consumption prediction model is improved. In an alternative embodiment, the power consumption prediction model may be continuously modified in connection with the actual power consumption value of the device to be processed during the power consumption prediction process. The baseboard management controller (Baseboard Management Controller, BMC) can be installed on the device to be processed, and the BMC is generally installed on a main board of the device to be processed and is mainly used for collecting actual power consumption values of the main board. Based on the above, in the model prediction process, the actual power consumption value of the device to be processed can be acquired through the BMC on the device to be processed according to the set period, and the model parameters in the power consumption prediction model are continuously adjusted according to the actual power consumption value acquired in each period and the model prediction result of the power consumption prediction model corresponding to each period until the difference value between the model prediction result of the power consumption prediction model and the actual power consumption value acquired through the MBC is smaller than the set difference value threshold. The model prediction result of the power consumption prediction model corresponding to each period may be a result of performing power consumption prediction on the device to be processed by the power consumption prediction model according to the power consumption prediction parameters in the corresponding period.

Based on the above, still another exemplary embodiment of the present application provides a power consumption control method, as shown in fig. 3, including the steps of:

301. and determining the power consumption adjustment parameters of the power consumption adjustment stage from the device parameters of the device to be processed.

302. From other device parameters than the power consumption adjustment parameter, a device parameter that changes with a change in the power consumption adjustment parameter and has a change sensitivity greater than a corresponding threshold value is determined as the power consumption prediction parameter.

303. And converting the non-power consumption type parameter in the power consumption prediction parameters into a target power consumption type parameter according to a preset parameter conversion relation.

304. Taking the original power consumption type parameter in the target power consumption type parameter and the power consumption prediction parameter as a model reference operation power consumption prediction model to obtain a predicted power consumption value of the equipment to be processed.

305. And if the actual power consumption value of the equipment to be processed is acquired through the BMC when the predicted power consumption value is obtained, acquiring a difference value between the predicted power consumption value and the actual power consumption value.

306. Judging whether the difference value between the predicted power consumption value and the actual power consumption value is larger than a set difference value threshold value or not; if yes, go to step 307, if no, go to step 308.

307. Model parameters in the power consumption prediction model are adjusted to improve the accuracy of the power consumption prediction model, and step 302 is returned.

308. Judging whether an external power consumption adjustment instruction is received or not; if yes, go to step 309; if the determination is negative, the process returns to step 302.

309. And adjusting the power consumption adjusting parameter according to the predicted power consumption value to control the power consumption of the equipment to be processed.

In this embodiment, the device to be processed includes a BMC, and the BMC may continuously collect the actual power consumption value of the device to be processed at a lower frequency. In the power consumption prediction stage described in steps 302-307, if the BMC acquires the actual power consumption value of the device to be processed at the corresponding time when the predicted power consumption value of the device to be processed is obtained, the two power consumption values may be compared, and the difference value of the two power consumption values may be compared with the set difference threshold. Wherein the difference threshold is capable of characterizing how close an actual power consumption value of the device to be processed is to the predicted power consumption value. The difference threshold is an empirical value, and embodiments of the present application are not limited.

When the difference between the predicted power consumption value and the actual power consumption value is larger than the set difference threshold, the predicted power consumption value can be considered to have larger deviation and is inconsistent with the actual power consumption value. And then, model parameters in the power consumption prediction model can be adjusted, the power consumption prediction model is continuously optimized until the difference value between the prediction result of the optimized power consumption prediction model and the actual power consumption value acquired by the BMC is reduced within a set difference value threshold, and a more accurate predicted power consumption value is provided for a power consumption adjustment stage so as to further improve the power consumption control effect.

Alternatively, variable control methods may be employed when adjusting model parameters in the power consumption prediction model. The model parameters corresponding to some power consumption prediction parameters are controlled to be unchanged, and the mode of changing the model parameters corresponding to other power consumption prediction parameters is adopted, so that the purpose of optimizing the power consumption prediction model is achieved. The model parameters of which power consumption prediction parameters are changed, and the model parameters of which power consumption prediction parameters are kept unchanged, which are determined randomly or according to the specific application requirements, and the method is not limited. Of course, in practical application, other parameter adjustment methods may be adopted, and the embodiment is not limited.

When an external power consumption adjustment instruction is received, a power consumption adjustment stage can be entered, and the power consumption adjustment parameter is adjusted according to a predicted power consumption value of the power consumption prediction stage, so as to control the power consumption of the device to be processed.

In some of the above embodiments and the described flows in the drawings, a plurality of operations appearing in a particular order are included, but it should be clearly understood that the operations may be performed out of the order in which they appear herein or in parallel, the sequence numbers of the operations such as 101, 102, etc. are merely used to distinguish between the various operations, and the sequence numbers themselves do not represent any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.

The above embodiment describes the flow of the power consumption control method. When the method is deployed and implemented, the method can be applied to the internal implementation of the equipment to be processed; or may be implemented in a power consumption control system.

In some alternative embodiments, the power consumption prediction or power consumption control method provided in the above embodiments is applied to an internal implementation of a device to be processed. The internal block diagram of the device to be processed is shown in fig. 4a, and the power consumption control process is described in detail below with reference to the internal block diagram shown in fig. 4 a.

As shown in fig. 4a, the apparatus to be processed includes: a power consumption prediction module 41 and a power consumption adjustment module 42. Power consumption prediction module 41 may include, among other things, a parameter collection sub-module 411, a calculation sub-module 412, and a power consumption prediction model. In addition, as shown in fig. 4a, the device to be processed further includes an MSR, a system file, and the like. The MSR stores the temperature of the CPU, the RAPL value of the CPU, the memory temperature, the memory power consumption and other register parameters; the system file stores system parameters such as memory utilization, network card I/O performance parameters and the like.

Alternatively, the CPU frequency may be used as the power consumption adjustment parameter of the power consumption adjustment stage, but is not limited thereto.

The parameter collecting sub-module 411 is mainly used for reading parameters such as CPU temperature, memory temperature, CPU power consumption, memory power consumption, etc. from the MSR, and/or reading performance parameters such as memory utilization, disk I/O performance parameters, network card I/O performance parameters, etc. from the system file, as power consumption prediction parameters in the power consumption prediction stage.

Wherein, the parameter collection sub-module 411 may provide the power consumption prediction parameters to the calculation sub-module 412 after obtaining the power consumption prediction parameters. The calculating sub-module 412 is mainly configured to calculate a predicted power consumption value of the device to be processed according to the power consumption prediction parameter provided by the parameter collecting sub-module 411, and provide the predicted power consumption value to the power consumption adjusting module 42. Optionally, as shown in fig. 4a, a first interface 43 is provided between the power consumption prediction module 41 and the power consumption control module 42. Based on this, the power consumption control module 42 may acquire the predicted power consumption value calculated by the calculation sub-module 412 through the first interface 43. The first interface 43 may be a hardware interface or a software interface, which is not limited in the embodiment of the present application.

The power consumption adjustment module 42 is mainly configured to adjust a power consumption adjustment parameter, such as a CPU frequency, according to the predicted power consumption value provided by the calculation sub-module 412, so as to control the power consumption of the device to be processed. Alternatively, the power consumption adjustment module 42 may reduce the CPU frequency of the device to be processed to reduce the power consumption of the device to be processed when the predicted power consumption value is greater than the set power consumption threshold.

In the above procedure, the calculation submodule 412 is specifically configured to: and taking the power consumption prediction parameters as a model reference operation power consumption prediction model to obtain a predicted power consumption value of the equipment to be processed.

Further, the calculation submodule 412 may convert the non-power consumption type parameter in the power consumption prediction parameters into the target power consumption type parameter according to a preset parameter conversion relation; and then taking the original power consumption type parameter and the target power consumption type parameter in the power consumption prediction parameters as a model-based operation power consumption prediction model to obtain a predicted power consumption value of the equipment to be processed.

Further optionally, as shown in fig. 4a, the device to be processed further includes a BMC 40, where the BMC 40 is mainly configured to collect an actual power consumption value of the device to be processed. As shown in fig. 4a, the power consumption adjustment module 42 may be connected to the BMC 40 on the device to be processed, and may collect the actual power consumption value of the device to be processed at a lower frequency (with a set period) through the BMC 40. Accordingly, the power consumption adjustment module 42 may also provide the actual power consumption value collected by the BMC 40 to the calculation sub-module 412, so that the calculation sub-module 412 trains and corrects the power consumption prediction model according to the actual power consumption value. As shown in fig. 4a, a second interface 44 is further provided between the power consumption prediction module 41 and the power consumption control module 42. Wherein the computing submodule 412 may read the actual power consumption value obtained by the BMC 40 from the power consumption control module 42 through the second interface 44. Alternatively, the second interface 44 may be a hardware interface or a software interface, which is not limited by the embodiment of the present application.

In the model prediction stage, for the calculation submodule 412, when the predicted power consumption value of the device to be processed is calculated, the actual power consumption value of the device to be processed, acquired by the BMC 40 at the corresponding moment, is read from the power consumption control module 42 through the second interface 44, and then a difference value between the predicted power consumption value and the actual power consumption value is acquired; when the difference is larger than the set difference threshold, model parameters in the power consumption prediction model are adjusted to optimize the power consumption prediction model, and the prediction precision of the power consumption prediction model is improved.

The interaction flow between each module in the internal block diagram shown in fig. 4a is as shown in fig. 4 b:

in the power consumption prediction stage, the parameter collection submodule 411 inputs the collected power consumption prediction parameters to the calculation submodule 412; the calculation sub-module 412 outputs the power consumption prediction parameters to the power consumption prediction model; the power consumption prediction model obtains a power consumption prediction result according to the power consumption prediction parameters, and inputs the result to the calculation sub-module 412, so that the calculation sub-module 412 determines a predicted power consumption value.

Alternatively, after the calculation submodule 412 obtains the predicted power consumption value, the predicted power consumption value may be directly written to the first interface 43, so that the power consumption adjustment module 42 may read the predicted power consumption value from the first interface 43. Or alternatively

As shown in fig. 4b, in the power consumption prediction stage of the above process, the power consumption adjustment module 42 may write the actual power consumption value of the device to be processed acquired by the BMC 40 to the second interface 44, and the parameter collection sub-module 411 reads the actual power consumption value of the device to be processed from the second interface 44 and inputs the actual power consumption value to the calculation sub-module 412. The calculation sub-module 412 may optimize the power consumption prediction model according to the actual power consumption value, and retrieve the predicted power consumption value using the optimized power consumption prediction model, and write the power prediction power consumption value to the first interface 43, so that the power consumption adjustment module 42 may read the predicted power consumption value from the first interface 43.

In the power consumption adjustment phase, the power consumption adjustment module 42 may receive a power consumption adjustment instruction input by a user, and when receiving the power consumption adjustment instruction, read a predicted power consumption value from the first interface 43 to perform power consumption adjustment according to the predicted power consumption value. At this stage, calculation submodule 412 may continually write the calculated predicted power consumption value to first interface 43 for reading by power consumption adjustment module 42.

In other alternative embodiments, the power consumption control method provided in the above embodiments is applied to a power consumption control system shown in fig. 4 c. As shown in fig. 4c, the system comprises: a first electronic device 401 and a second electronic device 402. The second electronic device 402 is a device to be processed, and the first electronic device 401 may perform power consumption control on the second electronic device 402.

In each power consumption control process, the first electronic device 401 first enters a power consumption prediction stage, and in the power consumption prediction stage, in combination with parameters to be adjusted in the power consumption adjustment stage, namely power consumption adjustment parameters, some device parameters which are sensitive to the change of the power consumption adjustment parameters are obtained from the device parameters of the second electronic device 402 as power consumption prediction parameters; then, the second electronic device 402 is subjected to power consumption prediction according to the power consumption prediction parameters, and a predicted power consumption value is obtained. Further, when power consumption adjustment needs to be performed on the first electronic device 401, for example, a set power consumption adjustment period arrives, or an external power consumption adjustment instruction is received, a power consumption adjustment stage is entered; in the power consumption adjustment stage, the first electronic device 401 adaptively adjusts the power consumption adjustment parameter according to the predicted power consumption value, so as to achieve the purpose of controlling the power consumption of the second electronic device 402.

The detailed implementation of the operations of selecting the power consumption prediction parameter, calculating the predicted power consumption value, and adjusting the power consumption adjustment parameter by the first electronic device 401 may be referred to the description in the foregoing embodiment, and will not be repeated here.

Fig. 5 is a schematic structural diagram of an electronic device according to another exemplary embodiment of the present application. As shown in fig. 5, the electronic device includes: a memory 51 and a processor 52.

Memory 51 is used to store one or more computer instructions and may be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on an electronic device.

The memory 51 may be implemented by any type of volatile or non-volatile memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

In some exemplary embodiments, memory 51 may optionally include memory located remotely from processor 52, which may be communicatively coupled to processor 52 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

A processor 52 coupled to the memory 51 for executing one or more computer instructions for: acquiring power consumption prediction parameters from equipment parameters of electronic equipment; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has the change sensitivity larger than a corresponding threshold value; and carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the equipment to be processed.

In some exemplary embodiments, the processor 52 is further configured to determine a power consumption adjustment parameter from the device parameters of the device to be processed prior to obtaining the power consumption prediction parameter.

Further alternatively, the processor 52 is specifically configured to: determining CPU frequency from equipment parameters of the electronic equipment as a power consumption adjustment parameter; and acquiring, as the power consumption prediction parameter, a device parameter which changes with a change in the CPU frequency and whose change sensitivity is greater than a corresponding threshold value, from other device parameters other than the CPU frequency.

Further alternatively, as shown in FIG. 5, the processor 52 internally contains an MSR 57. Based on this, the processor 52 is specifically configured to: reading at least one register parameter from an MSR 57 contained within the processor 52; and/or reading at least one system parameter from a system file of the electronic device, wherein the at least one system parameter comprises other system parameters besides CPU utilization, or the at least one system parameter comprises CPU utilization, and the importance ratio of the CPU utilization in the power consumption prediction is not the maximum and is smaller than a preset duty ratio threshold value.

In some exemplary embodiments, the processor 52, when obtaining the predicted power consumption value, is specifically configured to: taking the power consumption prediction parameters as a model to enter a running power consumption prediction model so as to obtain a predicted power consumption value.

Further, the processor 52 is specifically configured to: converting non-power consumption type parameters in the power consumption prediction parameters into target power consumption type parameters according to a preset parameter conversion relation; taking the original power consumption type parameter and the target power consumption type parameter in the power consumption prediction parameters as a model to enter a running power consumption prediction model so as to obtain a predicted power consumption value.

Further alternatively, as shown in fig. 5, the electronic device further includes: BMC 58. Based thereon, the processor 52 is further configured to: the method comprises the steps that actual power consumption values of equipment to be processed are collected through a baseboard management controller BMC in a set period; and continuously adjusting model parameters in the power consumption prediction model according to the actual power consumption value acquired in each period and the model prediction result of the power consumption prediction model corresponding to each period until the difference value between the model prediction result of the power consumption prediction model and the actual power consumption value acquired through the MBC is smaller than a set difference value threshold.

In some exemplary embodiments, the processor 52 is further configured to: and adjusting the power consumption adjusting parameter according to the predicted power consumption value to control the power consumption of the equipment to be processed.

Further alternatively, the processor 52 is specifically configured to, when adjusting the power consumption adjustment parameter: and if the predicted power consumption value is larger than the set power consumption threshold value, reducing the CPU frequency of the electronic equipment.

Further, as shown in fig. 5, the electronic device further includes: communication component 53, display 54, power component 55, audio component 56, and other components. Only some of the components are schematically shown in fig. 5, which does not mean that the electronic device only comprises the components shown in fig. 5.

The electronic device provided in this embodiment may execute the power consumption prediction method or the power consumption control method provided in the embodiment of the present application, and the working principle and the beneficial effects thereof may be referred to the description in the above method embodiment, which is not repeated herein.

Fig. 6 is a schematic structural diagram of an electronic device according to another exemplary embodiment of the present application. As shown in fig. 6, the electronic device includes: a memory 61 and a processor 62.

Memory 61 is used to store one or more computer instructions and may be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on an electronic device.

The memory 61 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

In some exemplary embodiments, memory 61 may optionally include memory located remotely from processor 62, which may be communicatively coupled to processor 62 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

A processor 62 coupled to the memory 61 for executing one or more computer instructions for:

acquiring power consumption prediction parameters from equipment parameters of electronic equipment; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has the change sensitivity larger than a corresponding threshold value; carrying out power consumption prediction on the equipment to be processed according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the equipment to be processed; and adjusting the power consumption adjusting parameter according to the predicted power consumption value to control the power consumption of the equipment to be processed.

In some exemplary embodiments, the processor 62 is further configured to determine a power consumption adjustment parameter from the device parameters of the device to be processed prior to obtaining the power consumption prediction parameter.

Further optionally, the processor 62 is specifically configured to: determining CPU frequency from equipment parameters of the electronic equipment as a power consumption adjustment parameter; and acquiring, as the power consumption prediction parameter, a device parameter which changes with a change in the CPU frequency and whose change sensitivity is greater than a corresponding threshold value, from other device parameters other than the CPU frequency.

Further alternatively, as shown in FIG. 6, the processor 62 internally contains an MSR 67. Based on this, the processor 62 is specifically configured to: reading at least one register parameter from an MSR 67 contained within the processor 62; and/or reading at least one system parameter from a system file of the electronic device, wherein the at least one system parameter comprises other system parameters besides CPU utilization, or the at least one system parameter comprises CPU utilization, and the importance ratio of the CPU utilization in the power consumption prediction is not the maximum and is smaller than a preset duty ratio threshold value.

In some exemplary embodiments, the processor 62, when obtaining the predicted power consumption value, is specifically configured to: taking the power consumption prediction parameters as a model to enter a running power consumption prediction model so as to obtain a predicted power consumption value.

Further, the processor 62 is specifically configured to: converting non-power consumption type parameters in the power consumption prediction parameters into target power consumption type parameters according to a preset parameter conversion relation; taking the original power consumption type parameter and the target power consumption type parameter in the power consumption prediction parameters as a model to enter a running power consumption prediction model so as to obtain a predicted power consumption value.

Further alternatively, as shown in fig. 6, the electronic device further includes: BMC 68. Based thereon, the processor 62 is also configured to: the method comprises the steps that actual power consumption values of equipment to be processed are collected through a baseboard management controller BMC in a set period; and continuously adjusting model parameters in the power consumption prediction model according to the actual power consumption value acquired in each period and the model prediction result of the power consumption prediction model corresponding to each period until the difference value between the model prediction result of the power consumption prediction model and the actual power consumption value acquired through the MBC is smaller than a set difference value threshold.

Further alternatively, the processor 62 is specifically configured to, when adjusting the power consumption adjustment parameter: and if the predicted power consumption value is larger than the set power consumption threshold value, reducing the CPU frequency of the electronic equipment.

Further, as shown in fig. 6, the electronic device further includes: communication component 63, display 64, power component 66, audio component 66, and other components. Only some of the components are schematically shown in fig. 6, which does not mean that the electronic device only comprises the components shown in fig. 6.

The electronic device provided in this embodiment may execute the power consumption control method provided in the embodiment of the present application, and the working principle and the beneficial effects thereof may be referred to the description in the above method embodiment, which is not repeated herein.

The communication component of fig. 5 or 6 may be configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device in which the communication component is located may access a wireless network based on a communication standard, such as WiFi,2G or 3G, or a combination thereof. In one exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.

The display in fig. 5 or 6 includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or sliding action, but also the duration and pressure associated with the touch or sliding operation.

The power supply assembly of fig. 5 or 6 provides power to the various components of the device in which the power supply assembly is located. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the devices in which the power components are located.

The audio component of fig. 5 or 6 may be configured to output and/or input audio signals. For example, the audio component includes a Microphone (MIC) configured to receive external audio signals when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a speech recognition mode. The received audio signal may be further stored in a memory or transmitted via a communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

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

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims

1. A power consumption prediction method, comprising:

acquiring power consumption prediction parameters from equipment parameters of equipment to be processed; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has a change sensitivity larger than a corresponding threshold value; the power consumption adjustment parameter is CPU frequency, the power consumption prediction parameter at least comprises other system parameters except CPU utilization rate and/or at least one register parameter in a special module register of the equipment to be processed, or the power consumption prediction parameter comprises CPU utilization rate and other system parameters, and the importance ratio of the CPU utilization rate in the power consumption prediction is not the maximum and is smaller than a preset duty ratio threshold;

2. The method of claim 1, wherein prior to obtaining the power consumption prediction parameter from the device parameters of the device to be processed, the method further comprises:

and determining the power consumption adjustment parameter from the device parameters of the device to be processed.

3. The method of claim 2, wherein determining the power consumption adjustment parameter from device parameters of the device to be processed comprises:

determining CPU frequency from the equipment parameters of the equipment to be processed as the power consumption adjustment parameter;

the obtaining the power consumption prediction parameter from the device parameters of the device to be processed comprises the following steps:

from other device parameters than the CPU frequency, a device parameter which changes with the change of the CPU frequency and has a change sensitivity larger than a corresponding threshold value is acquired as the power consumption prediction parameter.

4. A method according to claim 3, wherein said obtaining, as said power consumption prediction parameter, a device parameter which changes with a change in the CPU frequency and whose change sensitivity is greater than a corresponding threshold value, from other device parameters than the CPU frequency, comprises:

reading the at least one register parameter from a special module register of the device to be processed;

And/or

At least one system parameter is read from a system file of the device to be processed.

5. The method according to claim 1, wherein the predicting the power consumption of the device to be processed according to the power consumption prediction parameter to obtain the predicted power consumption value of the device to be processed comprises:

and taking the power consumption prediction parameters as a model reference operation power consumption prediction model to obtain the predicted power consumption value.

6. The method of claim 5, wherein said referencing the power consumption prediction parameters as a model into a power consumption prediction model to obtain the predicted power consumption value comprises:

according to a preset parameter conversion relation, converting the non-power consumption type parameter in the power consumption prediction parameters into a target power consumption type parameter;

and taking the original power consumption type parameter and the target power consumption type parameter in the power consumption prediction parameters as model parameters to operate the power consumption prediction model so as to obtain the predicted power consumption value.

7. The method of claim 5, wherein prior to referencing the power consumption prediction parameters as a model to a power consumption prediction model to obtain the predicted power consumption value, the method further comprises:

Collecting an actual power consumption value of the equipment to be processed through a baseboard management controller BMC in a set period;

and continuously adjusting model parameters in the power consumption prediction model according to the actual power consumption value acquired in each period and the model prediction result of the power consumption prediction model corresponding to each period until the difference value between the model prediction result of the power consumption prediction model and the actual power consumption value acquired through the BMC is smaller than a set difference value threshold.

8. The method according to any of claims 1-7, wherein after obtaining the predicted power consumption value of the device to be processed, the method further comprises:

9. The method of claim 8, wherein adjusting the power consumption adjustment parameter to control the power consumption of the device to be processed according to the predicted power consumption value comprises:

responding to a power consumption adjustment instruction, and comparing the predicted power consumption value with a set power consumption threshold value;

and if the predicted power consumption value is larger than the set power consumption threshold value, adjusting the power consumption adjustment parameter to reduce the power consumption of the equipment to be processed.

10. An electronic device, comprising: a memory and a processor;

the memory is used for storing a computer program;

acquiring power consumption prediction parameters from equipment parameters of the electronic equipment; the power consumption prediction parameter is a device parameter which changes along with the change of the power consumption adjustment parameter and has a change sensitivity larger than a corresponding threshold value; the power consumption adjustment parameter is CPU frequency, the power consumption prediction parameter at least comprises other system parameters except CPU utilization rate and/or at least one register parameter in a special module register of the electronic equipment, or the power consumption prediction parameter comprises CPU utilization rate and other system parameters, and the importance ratio of the CPU utilization rate in the power consumption prediction is not the maximum and is smaller than a preset duty ratio threshold;

11. The electronic device of claim 10, wherein the processor is further configured to:

and before the power consumption prediction parameters are acquired, determining the power consumption adjustment parameters from the device parameters of the device to be processed.

12. The electronic device of claim 11, wherein the processor is specifically configured to:

determining CPU frequency from the device parameters of the electronic device as the power consumption adjustment parameter; and

13. The electronic device of claim 12, wherein the processor is specifically configured to:

reading at least one register parameter from a special module register contained within the processor;

and/or

At least one system parameter is read from a system file of the electronic device.

14. The electronic device according to any of the claims 10-13, wherein the processor is specifically configured to:

15. The electronic device of claim 14, wherein the processor is specifically configured to:

16. The electronic device of claim 14, wherein the processor is further configured to:

17. A computer readable storage medium storing a computer program, which when executed is capable of carrying out the steps of the method of any one of claims 1-9.

18. A power consumption control method, characterized by comprising:

19. An electronic device, comprising: a memory and a processor;

the memory is used for storing a computer program;

Carrying out power consumption prediction on the electronic equipment according to the power consumption prediction parameters so as to obtain a predicted power consumption value of the electronic equipment;

and adjusting the power consumption adjusting parameter according to the predicted power consumption value so as to control the power consumption of the electronic equipment.

20. A computer readable storage medium storing a computer program, which when executed is capable of carrying out the steps of the method of claim 18.