CN110308782A - Power consumption prediction, control method, equipment and computer readable storage medium - Google Patents
Power consumption prediction, control method, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN110308782A CN110308782A CN201810241512.5A CN201810241512A CN110308782A CN 110308782 A CN110308782 A CN 110308782A CN 201810241512 A CN201810241512 A CN 201810241512A CN 110308782 A CN110308782 A CN 110308782A
- Authority
- CN
- China
- Prior art keywords
- power consumption
- equipment
- prediction
- parameter
- processed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/324—Power saving characterised by the action undertaken by lowering clock frequency
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The embodiment of the present application provides a kind of power consumption prediction, control method, equipment and computer readable storage medium.In the present embodiment, be changed using meeting with the variation of power consumption adjusting parameter, and changing sensitivity is greater than the device parameter of respective threshold as power consumption Prediction Parameters;Processing equipment, which is treated, according to power consumption Prediction Parameters carries out power consumption prediction, to obtain the prediction power consumption number of equipment to be processed, so that prediction power consumption number accuracy with higher, and then power consumption control is carried out according to the prediction power consumption number with high accuracy, power consumption control effect can be improved.
Description
Technical field
This application involves electronic technology field more particularly to a kind of prediction of power consumption, control method, equipment and computer-readable
Storage medium.
Background technique
With advances in technology, the operational capability of electronic equipment is continuously improved, and function constantly extends, this is to a certain extent
It will increase the power consumption of electronic equipment.Excessively high power consumption will reduce the service performance of electronic equipment, it is therefore necessary to set to electronics
It is standby to carry out power consumption control, to reduce energy consumption, extends battery, improve the stability etc. of electronic equipment.
In the prior art, the central processing unit (Central Processing Unit, CPU) of electronic equipment is usually utilized
The power consumption of usage forecast electronic equipment is then based on the power consumption predicted and carries out power consumption control to electronic equipment.But it is existing
Power consumption prediction result is not accurate enough, causes power consumption control effect undesirable.
Summary of the invention
The many aspects of the embodiment of the present application provide a kind of power consumption prediction, control method, equipment and computer-readable storage
Medium provides basis to improve the accuracy of power consumption prediction to promote power consumption control effect.
The embodiment of the present application provides a kind of power consumption prediction technique, comprising:
Power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed;The power consumption Prediction Parameters refer to power consumption tune
The variation of whole parameter and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the equipment to be processed
Prediction power consumption number.
The embodiment of the present application also provides a kind of electronic equipment, comprising: memory and processor;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
Power consumption Prediction Parameters are obtained from the device parameter of the electronic equipment;The power consumption Prediction Parameters refer to power consumption
The variation of adjusting parameter and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the equipment to be processed
Prediction power consumption number.
The embodiment of the present application also provides a kind of computer readable storage medium for being stored with computer program, the computer
Program is performed the step that can be realized in power consumption prediction technique provided by the embodiments of the present application.
The embodiment of the present application also provides a kind of power consumption control method, comprising:
Power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed;The power consumption Prediction Parameters refer to power consumption tune
The variation of whole parameter and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the equipment to be processed
Prediction power consumption number;
The power consumption adjusting parameter is adjusted according to the prediction power consumption number, to control the function of the equipment to be processed
Consumption.
The embodiment of the present application also provides a kind of electronic equipment, comprising: memory and processor;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
From the device parameter of the electronic equipment, power consumption Prediction Parameters are obtained;The power consumption Prediction Parameters refer to function
It consumes the variation of adjusting parameter and changes, and changing sensitivity is greater than the device parameter of respective threshold;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the equipment to be processed
Prediction power consumption number;
The power consumption adjusting parameter is adjusted according to the prediction power consumption number, to control the function of the equipment to be processed
Consumption.
The embodiment of the present application also provides a kind of computer readable storage medium for being stored with computer program, the computer
Program is performed the step that can be realized in power consumption control method provided by the embodiments of the present application.
In the embodiment of the present application, be changed using meeting with the variation of power consumption adjusting parameter, and changing sensitivity is higher
Device parameter predicts the prediction power consumption number of equipment to be processed, so that prediction power consumption number accuracy with higher works as base in turn
When the prediction power consumption number with high accuracy carries out power consumption control, be conducive to improve power consumption control effect.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 a is the flow diagram for the power consumption prediction technique that one exemplary embodiment of the application provides;
Fig. 1 b is a kind of flow diagram for power consumption control method that one exemplary embodiment of the application provides;
Fig. 2 is the flow diagram for another power consumption control method that the application another exemplary embodiment provides;
Fig. 3 is the flow diagram for another power consumption control method that the application another exemplary embodiment provides;
Fig. 4 a is the internal frame diagram for the equipment to be processed that the application another exemplary embodiment provides;
The interaction flow signal of each module in block diagram shown in Fig. 4 a that Fig. 4 b provides for the application another exemplary embodiment
Figure;
Fig. 4 c is the structural schematic diagram for the power consumption control system that the application another exemplary embodiment provides;
Fig. 5 is the structural schematic diagram for a kind of electronic equipment that the application another exemplary embodiment provides;
Fig. 6 is the structural schematic diagram for another electronic equipment that the application another exemplary embodiment provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
For the technical problem of existing power consumption prediction result inaccuracy, in some exemplary embodiments of the application, use
It can change with the variation of power consumption adjusting parameter, and the higher device parameter of changing sensitivity is predicted as power consumption Prediction Parameters
The prediction power consumption number of equipment to be processed, so that prediction power consumption number accuracy with higher.In turn, when based on compared with high precision
When the prediction power consumption number of degree carries out power consumption control, be conducive to improve power consumption control effect.
Below in conjunction with attached drawing, the technical scheme provided by various embodiments of the present application will be described in detail.
Fig. 1 a is the flow diagram for the power consumption prediction technique that one exemplary embodiment of the application provides.As shown in Figure 1a,
This method comprises:
101, power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed, which refers to power consumption
The variation of adjusting parameter and change, and changing sensitivity be greater than respective threshold device parameter.
102, processing equipment is treated according to power consumption Prediction Parameters and carry out power consumption prediction, to obtain the pre- measurement of power of equipment to be processed
Consumption value.
In the present embodiment, power consumption will be generated in the process of running and it is necessary to the electronics of power consumption prediction is carried out to it
Equipment is known as equipment to be processed, can be any electronic equipment that can generate power consumption in the process of running.
In some optional embodiments, equipment to be processed can be smart phone, tablet computer, PC, wearing and set
It is standby to wait terminal devices.Equipment to be processed may include at least one processing unit and at least one processor.It processing unit and deposits
The quantity of reservoir depends on the configuration and type of terminal device.Memory may include volatibility, such as RAM, also can wrap
Non-volatile, such as read-only memory (Read-Only Memory, ROM), flash memory etc. are included, or can also simultaneously include two
Seed type.Operating system (Operating System, OS), one or more application program are typically stored in memory
(application, app) also can store program data etc..Other than processing unit and memory, equipment to be processed
Also the basic configuration such as network card chip, IO bus, audio-video component be will include.Optionally, according to the way of realization of equipment to be processed,
Equipment to be processed also may include some peripheral equipments, such as keyboard, mouse, input pen, printer etc..These peripheral equipments exist
It is well-known in this field, this will not be repeated here.
In other alternative embodiments, equipment to be processed can be any server with certain computing capability and set
It is standby, such as can be General Server, Cloud Server, cloud host, virtual center etc..The composition of server mainly includes processing
Device, hard disk, memory, system bus etc. are similar with general computer architecture.
Wherein, equipment to be processed has some device parameters, these device parameters may include the relevant parameter of software, hard
The relevant parameter of part and some performance parameters etc..In these device parameters, the power consumption phase of some parameters and equipment to be processed
It closes.Each embodiment of the application is primarily upon parameter relevant to the power consumption of equipment to be processed, for example including can cause power consumption
The parameter of variation, and/or, the parameter that can be influenced by change of power consumption.
In application scenes, needs to treat processing equipment and carry out power consumption control.The power consumption control process can be divided into two
A stage: power consumption forecast period and power consumption adjusting stage.Power consumption forecast period is mainly responsible for the power consumption for predicting equipment to be processed, is
The power consumption adjusting stage provides data and supports.The power consumption adjusting stage is mainly responsible for the power consumption number predicted according to power consumption forecast period,
Make the power consumption of equipment to be processed to expected direction change by adjusting relevant parameter, and then achievees the purpose that power consumption control.
The prediction result of power consumption forecast period is the basis of power consumption control, and the accuracy of power consumption prediction result drastically influences
Power consumption control effect.The present embodiment provides a kind of power consumption prediction techniques, to improve the accuracy of power consumption prediction result.In this reality
During the power consumption prediction for applying example, processing equipment can be treated based on some device parameters of equipment to be processed and carries out power consumption prediction.
In order to improve the accuracy of prediction result, the present embodiment combination power consumption adjusting stage needs parameter (the referred to as power consumption tune adjusted
Whole parameter), select the variation of some pairs of power consumption adjusting parameters to join from the device parameter of equipment to be processed than more sensitive equipment
Number is used as power consumption Prediction Parameters, for predicting the power consumption of equipment to be processed.
Power consumption Prediction Parameters in the present embodiment belong to the device parameter of equipment to be processed, can reflect to a certain extent to
The power consumption features of processing equipment.In addition, can mutual shadow between the power consumption Prediction Parameters in the present embodiment and power consumption adjusting parameter
It rings and more sensitive to variation each other, once i.e. power consumption adjusting parameter changes, power consumption Prediction Parameters can be rapidly
Significantly variation is generated, correspondingly, the variation of power consumption Prediction Parameters can reflect in time in power consumption adjusting parameter again, using tool
There are the power consumption Prediction Parameters of this sensitivity characteristic that the accuracy of power consumption prediction result can be improved.
In addition, in the present embodiment, in order to embody, power consumption Prediction Parameters change with power consumption adjusting parameter and change sensitive
Degree, presets the corresponding threshold value of each parameter.Based on this, to some device parameter, it can be determined that the device parameter is with power consumption tune
Whether the changing sensitivity of whole parameter is greater than respective threshold;When the changing sensitivity is greater than respective threshold, determine that the equipment is joined
Number can be used as the power consumption Prediction Parameters in the present embodiment, for predicting the power consumption of equipment to be processed.It is alternatively possible to according to warp
It tests value and is previously obtained each device parameter with the changing sensitivity of power consumption adjusting parameter.
After obtaining power consumption Prediction Parameters, the power consumption number of equipment to be processed can be predicted according to power consumption Prediction Parameters.For
Convenient for distinguishing and description, the power consumption number of the equipment to be processed predicted here is known as to predict power consumption number.
In application scenes, after obtaining prediction power consumption number, processing equipment can be treated according to the prediction power consumption number
Carry out power consumption control.Then a kind of process of power consumption control method is as shown in Figure 1 b, after step 102, further includes:
103, power consumption adjusting parameter is adjusted according to prediction power consumption number, to control the power consumption of equipment to be processed.
Optionally, if prediction power consumption number is larger, illustrate that equipment current power consumption to be processed is larger, adjustable power consumption adjustment ginseng
It counts to reduce the power consumption of equipment to be processed;If predicting, power consumption number is smaller, illustrates that equipment current power consumption to be processed is smaller, then can lead to
Cross the power consumption that adjustment power consumption adjusting parameter increases equipment to be processed.It is worth noting that increasing the power consumption of equipment to be processed, wait locate
The process performance of reason equipment typically results in improvement.
In the above-described embodiments, be changed using meeting with the variation of power consumption adjusting parameter, and changing sensitivity is higher
Parameter predicts according to power consumption Prediction Parameters the power consumption number of equipment to be processed as power consumption Prediction Parameters, so that prediction power consumption
It is worth accuracy with higher.Further, when carrying out power consumption control based on the prediction power consumption number with high accuracy, favorably
In raising power consumption control effect.
It is worth noting that require to carry out power consumption prediction in much application scenarios relevant to power consumption, and the application is real
The power consumption prediction technique for applying example offer is applicable to the various application scenarios with power consumption forecast demand.It is exemplified below:
For example, projector is projected in the environment of Dimmable in a kind of scene.In this scenario, it can combine
The intensity of the projection ray of projector is adjusted flexibly in ambient brightness, to achieve the purpose that save power consumption.For example, when ambient brightness compared with
Gao Shi can reduce the intensity of projection ray, on the basis of guaranteeing projected picture clarity, reduce power consumption as far as possible;Work as environment
When brightness is lower, the intensity of projection ray can be improved, the preferential clarity for guaranteeing projected picture.In this scenario, projector
It can be considered the equipment to be processed in each embodiment of the application;Correspondingly, the brightness of projection ray is as power consumption adjusting parameter.It is based on
This, to projector carry out power consumption control during, can choose can be generated with the variation of the brightness of projection ray it is larger
Proj ector parameters of variation, such as bulb irradiation duration, size of current etc. are used as power consumption Prediction Parameters;It is predicted based on these power consumptions
The prediction power consumption number of gain of parameter projector;Further, theoretical power consumption threshold value is determined in combination with current environment brightness, and according to this
It predicts the difference between power consumption number and corresponding theoretical power consumption threshold value, the brightness of projection ray is adjusted, thus by the function of projector
Consumption is adjusted to theoretical power consumption threshold value, in the case where guaranteeing projected picture clarity, reduces power consumption as far as possible.
In another example multiple servers are deployed in computer room, rack or cabinet in another application scenarios, in order to meet
Business demand, such as dilatation, it may be necessary to adjust the deployment density of server.When adjusting the deployment density of server, to protect
The overall power of computer room, rack or cabinet is demonstrate,proved still in more reasonable power consumption range, it may be necessary to the power consumption of server
It is adjusted.For example, the number of servers in the case where improving server disposition density, in entire computer room, rack or cabinet
It will increase, in the case where guaranteeing computer room, rack or constant cabinet overall power, need to reduce partly or entirely existing server
Power consumption.At this time, it may be necessary to which the existing server for reducing power consumption is equipment to be processed;It correspondingly, can be by existing server
The parameters such as cpu frequency are as power consumption adjusting parameter.Based on this, during power consumption control, it can choose to adjust with power consumption and join
Several variations have the device parameter of large change, for example, can be memory temperature, cpu temperature, memory usage, disk it is every
Second read-write operation number etc. is used as power consumption Prediction Parameters;The power consumption of accordingly existing server is predicted using these power consumption Prediction Parameters
Value;In turn, can accordingly have clothes with what is predicted according to the maximum power dissipation value of every server after raising deployment density
The power consumption number of business device adjusts separately the cpu frequency of corresponding existing server, reaches and arrives the lower power consumption of corresponding existing server
The purpose of maximum power dissipation value.The case where accordingly for needing to reduce server disposition density, entire computer room, rack or
Number of servers in cabinet can be reduced, i.e., part server can be removed, and guarantee computer room, rack or cabinet overall power not
In the case where change, allow remainder or whole servers that there are more power consumptions, it is also desirable to carry out power consumption control, wherein power consumption is pre-
Survey process is similar to be repeated no more.
For another example in other scenes, for the server in same cabinet, rack or computer room, this Shen can also be used
Please the method that provides of embodiment predict its current power consumption value in real time, and then can be in the when progress that reaches certain threshold value of prediction power consumption number
Power consumption control, to ensure Electrical Safety.
It in some exemplary embodiments, can be from equipment to be processed before treating processing equipment and carrying out power consumption prediction
Device parameter in, determine power consumption adjusting parameter.In these exemplary embodiments, power consumption adjusting parameter and power consumption Prediction Parameters
Equally, the device parameter of equipment to be processed is also belonged to.
In different application scene, power consumption adjusting parameter be will be different.For example, can be incited somebody to action in application scenes
The load number of equipment to be processed as power consumption adjusting parameter, and then can reach by adjusting the load number of equipment to be processed by
The purpose of the power consumption control of equipment to be processed in the reasonable scope.It, can will be to be processed in another example in other application scenarios
The fan speed of equipment is as power consumption adjusting parameter, and then can reach by adjusting the fan speed of equipment to be processed will be to be processed
The purpose of the power consumption control of equipment in the reasonable scope.Wherein, power consumption adjusting parameter is different, with the variation of power consumption adjusting parameter
The power consumption Prediction Parameters that variation and changing sensitivity are greater than respective threshold would also vary from.
In some exemplary embodiments of the application, it is contemplated that the incidence relation between cpu frequency and power consumption, it can be by CPU
Frequency is as power consumption adjusting parameter.Cpu frequency is the frequency of the system clock of CPU, i.e. the synchronization of CPU interior generation per second when working
Umber of pulse, unit are Hz.Cpu frequency can determine the arithmetic speed of CPU, in general, cpu frequency is higher, the arithmetic speed of CPU is got over
Fastly.
CPU is main core component in equipment to be processed.CPU is a kind of digital integrated electronic circuit, can use metal oxygen
Compound semiconductor (Metal-Oxide-Semiconductor, MOS) technology, such as complementary metal oxide semiconductor
(Complementary Metal Oxide Semiconductor, CMOS), N-channel type metal oxide semiconductor
(Negative channel--Metal-Oxide-Semiconductor, NMOS), P-channel type metal oxide semiconductor
(Positive channel Metal Oxide Semiconductor, PMOS) technology is made.This digital integration electricity
Road works under the control of system clock, and in each clock cycle, internal gate circuit can change state.It is every in gate circuit
When primary change state, has very big electric current and flow through.Clock cycle is shorter, and the frequency that gate circuit changes state is higher,
The frequency that correspondingly high current flows through is also higher, and then power consumption is caused to increase.Clock cycle is the inverse of frequency, that is to say, that
The frequency of CPU is higher, and the clock cycle is shorter, and the average current of consumption is bigger, and power consumption is higher.
Using cpu frequency as in the exemplary embodiment of power consumption adjusting parameter, during power consumption prediction, can to
In the device parameter of processing equipment, selection in addition to cpu frequency, can change with the variation of cpu frequency, and change spirit
Sensitivity is greater than power consumption Prediction Parameters of the device parameter of respective threshold as power consumption forecast period.
In the device parameter of equipment to be processed, it can change with the variation of cpu frequency, and changing sensitivity is greater than
The device parameter (refer to change cpu frequency than more sensitive device parameter in simple terms) of respective threshold has very much.It gives below
Some exemplary parameters out:
Register parameters:
One is stored in the special module register (Model Specific Registers, MSR) of equipment to be processed
Device parameter relevant to the critical components such as the CPU of equipment to be processed, memory a bit.Variation of these device parameters to cpu frequency
Compare it is sensitive, cpu frequency variation when, these parameters can also change in time.For ease of description, the ginseng that will be stored in MSR
Number is known as register parameters.Based on this, at least one register parameters can be read from the MSR of equipment to be processed, as function
Consume Prediction Parameters.Several register parameters are set forth below:
(1) the temperature shape parameter in MSR:
In the MSR of equipment to be processed, the temperature parameter comprising some components, such as temperature value per second.In general, to be processed
The temperature of each component can reflect out the power consumption of equipment to be processed to a certain extent in equipment, in addition, in MSR these components temperature
It is more sensitive to the variation of cpu frequency to spend parameter, therefore can be used as the power consumption Prediction Parameters in each embodiment of the application.
For example, the CPU of equipment to be processed, memory, solid state hard disk (Solid can be read from the MSR of equipment to be processed
State Drives, SSD) etc. components temperature parameter.For the component with shell, such as SSD, temperature parameter be can wrap
Include shell temperature and/or internal temperature.
In addition, the fan of equipment to be processed is also the component of equipment to be processed, the revolving speed of fan as with temperature parameter phase
Associated indirect parameter can embody the temperature of fan, can be regarded as a kind of special temperature shape parameter, can also be used as function
Consumption Prediction Parameters are used to carry out power consumption prediction.
(2) the power consumption shape parameter in MSR:
It also include the power consumption parameter of some components, such as power consumption number per second in the MSR of equipment to be processed.In general, wait locate
The power consumption of each component can directly reflect the power consumption of equipment to be processed in reason equipment, in addition, the power consumption of these components is joined in MSR
The variation of several pairs of cpu frequencies is more sensitive, therefore can be used as the power consumption Prediction Parameters in each embodiment of the application.
For example, the CPU, memory, dynamic random access memory of equipment to be processed can be read from the MSR of equipment to be processed
The power consumption parameter of the components such as device (Dynamic Random Access Memory, DRAM).Wherein, component is different, power consumption ginseng
Several embodiment modes would also vary from.For example, the power consumption number of CPU can be read in MSR, or the operation of reading CPU is put down
Equal power limit (Running Average Power Limit, RAPL) value, the RAPL value can embody the power consumption of CPU.Except this
Except, in MSR, the DC power value of some other component can also be read, the power supply of other component can also be read
Efficiency, power-on time etc. can embody the power consumption shape parameter of power consumption number indirectly.
In the above-described embodiments, at least one register parameters can be read from the MSR of equipment to be processed, as can be with
Cpu frequency variation and change and changing sensitivity be greater than given threshold power consumption Prediction Parameters.It is this that register is read from MSR
The mode of parameter has high stability and high accuracy.In addition to this, this mode for reading parameter is few to resource consumption,
It is low to service impact.
System parameter:
Equipment to be processed has some system parameters, such as cpu busy percentage, memory usage, disk input/output
(Input/Output, I/O) performance parameter and network interface card I/O performance parameter etc., these system parameters are typically stored to be processed
In the system file of equipment.In these system parameters, there are the variations of some pairs of cpu frequencies than more sensitive system parameter,
In cpu frequency variation, these can also change to the variation of cpu frequency than more sensitive system parameter in time.Based on this,
At least one system parameter can be read from the system file of equipment to be processed, as power consumption Prediction Parameters.
In one embodiment, it is contemplated that cpu busy percentage is relatively low with the sensitivity that cpu frequency changes, in order to mention
The accuracy of high power consumption prediction result can choose other system parameters in addition to cpu busy percentage, such as memory usage,
Disk input/output (Input/Output, I/O) performance parameter and network interface card I/O performance parameter etc. are used as power consumption Prediction Parameters.
In another embodiment, it is contemplated that cpu busy percentage is relatively low with the sensitivity that cpu frequency changes, but logical
Crossing reduces different degree accounting of the cpu busy percentage in power consumption prediction to reduce accuracy of the cpu busy percentage to power consumption prediction result
Adverse effect.Based on this, cpu busy percentage and other system parameters can choose as power consumption Prediction Parameters, but consider
Cpu busy percentage is relatively low with the sensitivity that cpu frequency changes, it is desirable that different degree accounting of the cpu busy percentage in power consumption prediction
It is not the largest and is less than default accounting threshold value.Different degree accounting and phase answer seizure ratio threshold of the cpu busy percentage in power consumption prediction
In general value can need to reduce to the different degree accounting of cpu busy percentage to power consumption prediction result depending on application demand
The degree that accuracy has little effect.
Above-mentioned magnetic disc i/o performance parameter includes but is not limited to: the number (Input/Output per second for carrying out I/O operation
Operations Per Second, iops), average each I/O operation waiting time (await), bit per second (bits
Per second, bps), I/O equipment number (w/s) is write in the reading I/O equipment number (r/s) of completion per second, completion per second, and one
I/O queue, IOQ is in the parameters such as the time accounting (%util) of non-null states in second.
Above-mentioned network interface card I/O performance parameter includes but is not limited to: the transmission number of bits (tx_bytes) of network interface card, receives bit
Digit (rx_bytes) is sent packet loss (tx_dropped), is received packet loss (rx_dropped), and First Input First Output is received
The frequency (rx_fifo) of data is received, the frequency (tx_fifo) that First Input First Output issues data, received data packet amount are sent
(rx_packages), parameters such as data packet amount (tx_packages) are sent.
In the above-described embodiments, at least one system parameter can be read from the system file of equipment to be processed, as
Power consumption Prediction Parameters.This mode that system parameter is read from system file, speed is fast, and the parameter value read is accurate
Degree is high.
It is worth noting that when equipment to be processed is in cluster environment, it is related in addition to obtaining equipment itself to be processed
Some parameters except, some parameters for having the associated other equipment of power consumption with equipment to be processed can also be obtained, as power consumption
The power consumption Prediction Parameters of forecast period.Optionally, with equipment to be processed have the associated other equipment of power consumption can be with wait locate
There is the equipment mutually restricted in the power consumption total amount for managing equipment, for example, it is assumed that equipment to be processed is computer room or machine within the set range
A server in cabinet, then the server may be will affect by being located at same computer room or the power consumption of other servers in cabinet
Power consumption belongs to the other equipment for having power consumption incidence relation with the server.
It is worth noting that can choose according to the difference of application demand using the above-mentioned one or several kinds of ginsengs shown
Number, as power consumption Prediction Parameters.In addition, can choose during different power consumption prediction using identical power consumption Prediction Parameters,
Also it can choose using not exactly the same power consumption Prediction Parameters, specifically depending on application demand.
For the implementation process that can be more apparent from the embodiment of the present application convenient for those skilled in the art, function below
In consumption control embodiment, by taking power consumption adjusting parameter is the cpu frequency of equipment to be processed as an example, to the technical side of the embodiment of the present application
Case is described in detail.
Fig. 2 is the flow diagram for another power consumption control method that the application another exemplary embodiment provides.Such as Fig. 2
It is shown, this method comprises:
200, from the device parameter of equipment to be processed, determine cpu frequency as power consumption adjusting parameter.
201, from other device parameters in addition to cpu frequency, acquisition changes with the variation of cpu frequency and changes spirit
Sensitivity is greater than the device parameter of respective threshold as power consumption Prediction Parameters.
202, processing equipment is treated according to power consumption Prediction Parameters and carry out power consumption prediction, to obtain prediction power consumption number.
203, judge to predict whether power consumption number is greater than setting power consumption threshold value;If more than otherwise execution step 204 executes step
Rapid 205.
204, in the power consumption adjusting stage, the cpu frequency of equipment to be processed is reduced, the power consumption of equipment to be processed is dropped to and is set
Power consumption threshold value is determined hereinafter, and entering step 205.
205, it waits and enters next power consumption forecast period, and when next power consumption forecast period reaches, return step
201。
In the present embodiment, cpu frequency is determined as power consumption adjusting parameter, is then utilized the variation with cpu frequency and is become
Change and changing sensitivity treats processing equipment greater than the power consumption Prediction Parameters of respective threshold and carries out power consumption prediction, to obtain pre- measurement of power
Consumption value.Then, cpu frequency is adaptively adjusted according to prediction power consumption number, to achieve the purpose that control the power consumption of equipment to be processed.
In the present embodiment, it is desirable that the power consumption of equipment to be processed is less than setting power consumption threshold value.Based on this, pre- measurement of power is being obtained
After consumption value, prediction power consumption number can be compared with preset power consumption threshold value;And whether function is entered by comparison result automatic trigger
Consume the adjusting stage.It when predicting that power consumption number is greater than setting power consumption threshold value, determines and enters the power consumption adjusting stage, adjust rank in power consumption
Section, reduces the cpu frequency of equipment to be processed, and the power consumption of equipment to be processed is dropped to setting power consumption threshold value hereinafter, reaching power consumption
The purpose of control.Conversely, the function of equipment to be processed can not be adjusted when predicting that power consumption number is less than or equal to setting power consumption threshold value
Consumption, and wait into next power consumption forecast period, accomplish the purpose for monitoring the power consumption of equipment to be processed on demand.
Here setting power consumption threshold value refer to equipment to be processed expectation power consumption number or the permitted power consumption number upper limit, it is unlimited
Its fixed numerical value, can be according to application scenarios adaptive settings.
It is worth noting that in addition to whether being entered by prediction power consumption number and the comparison result automatic trigger of preset power consumption threshold value
Except the power consumption adjusting stage, can also trigger whether enter the power consumption adjusting stage by external command.For example, when prediction power consumption number is big
When setting power consumption threshold value, outside can issue power consumption adjustment instruction;After receiving from external power consumption adjustment instruction, really
Surely enter the power consumption adjusting stage, and in the power consumption adjusting stage, the cpu frequency of equipment to be processed is reduced, by equipment to be processed
Power consumption drops to setting power consumption threshold value hereinafter, achieving the purpose that power consumption control.
In the present embodiment, using cpu frequency as the power consumption adjusting parameter of power consumption adjusting stage, so that power consumption adjustment is more straight
It is connected to effect.In addition to this, variation is generated using the variation with cpu frequency, and changing sensitivity is greater than the equipment of respective threshold
It adopts parameter prediction and obtains the prediction power consumption number of equipment to be processed, and the foundation that the prediction power consumption number is adjusted as power consumption, so that
The accuracy of power consumption adjustment is higher, and the effect of power consumption adjustment is more preferable.
In each embodiment of the application, processing equipment can be treated according to power consumption Prediction Parameters and carries out power consumption prediction.Optionally,
It can train in advance and obtain a power consumption prediction model, in this way after obtaining power consumption Prediction Parameters, power consumption Prediction Parameters can be made
Enter ginseng operation power consumption prediction model for model, to obtain the prediction power consumption number of equipment to be processed.
In some exemplary embodiments, it is contemplated that the type of power consumption Prediction Parameters will be different, such as some are function
Shape parameter is consumed, some are non-power consumption shape parameters, to be convenient for power consumption prediction, can carry out parameter type conversion in advance.It is based on
This, it is a kind of based on power consumption prediction model treat processing equipment carry out power consumption prediction embodiment include: according to preset parameter
Non- power consumption shape parameter in power consumption Prediction Parameters is converted to power consumption shape parameter by transformational relation, for example, by temperature shape parameter and
The system parameter of non-power consumption type is converted to power consumption shape parameter;Later, by power consumption Prediction Parameters original power consumption shape parameter and
Target power consumption shape parameter enters ginseng operation power consumption prediction model as model, to obtain the prediction power consumption number of equipment to be processed.At this
In for the ease of distinguish power consumption Prediction Parameters originally include power consumption shape parameter and by non-power consumption shape parameter convert Lai power consumption type
Parameter, by by non-power consumption shape parameter convert Lai power consumption shape parameter be known as target power consumption shape parameter, and will be in power consumption Prediction Parameters
Originally the power consumption shape parameter for including is known as original power consumption shape parameter.
Optionally, for the energy-consuming parts in equipment to be processed, the corresponding relationship of its temperature and power consumption can be pre-established.With
For the server of a certain model, CPU rated disspation range be 65~155W, memory rated disspation range be 4~7W, it is hard
The rated disspation range of disk is 15~35W, the rated disspation range of network interface card is 2~25W, the rated disspation range of fan be 5~
35W.In the present embodiment, the power consumption of above-mentioned component at different temperatures can be measured in advance, for example, the temperature of CPU is 23 degree or so
When, power consumption 65W;When being 30 degree, power consumption 70W;When being 50 degree, power consumption 155W.For another example the temperature of hard disk exists
At 30 degree, power consumption 15W;At 55 degree, power consumption 35W.The present embodiment can take multiple measurements, and based on measurement knot
Fruit establishes the temperature of different energy-consuming parts and the transformational relation of power consumption.Above-mentioned transformational relation can be indicated by transfer function, obtained
After taking temperature parameter, the corresponding target power consumption shape parameter of temperature parameter can be obtained according to transfer function.
Likewise, the transfer function of other types of performance parameter and power consumption shape parameter can also be established through the above way,
And accordingly converted after obtaining performance parameter, it repeats no more.
The power consumption prediction model used in above-described embodiment can be collected previously according to power consumption Prediction Parameters sample and BMC
The actual power loss value of equipment to be processed be trained.It is pre- for each power consumption in advance in the power consumption prediction model
It surveys parameter (including original power consumption shape parameter and the target power consumption shape parameter being converted to) and distributes corresponding model parameter, such as
Weighted value, the model parameter can indicate that different degree accounting of the different power consumption Prediction Parameters in power consumption prediction, the different degree account for
Than indicating that the power consumption Prediction Parameters treat the contribution amount of the prediction power consumption number of processing equipment.For a power consumption Prediction Parameters,
Changing sensitivity with power consumption adjusting parameter (such as cpu frequency) is higher, and the power consumption Prediction Parameters are important in power consumption prediction
Spend the bigger of accounting.The corresponding model parameter of different power consumption Prediction Parameters can be preset based on experience value, can also be by mould
Type adaptive learning obtains.
It is worth noting that regardless of whether being set using power consumption Prediction Parameters to be processed using above-mentioned power consumption prediction model
Standby progress power consumption prediction, which may is that, treats processing equipment in conjunction with power consumption Prediction Parameters and the corresponding weighted value of power consumption Prediction Parameters
Carry out the process of model prediction.Wherein, the size Yu power consumption Prediction Parameters of the corresponding weighted value of power consumption Prediction Parameters are with power consumption tune
The height of the changing sensitivity of whole parameter is directly proportional, i.e., power consumption Prediction Parameters are higher with the changing sensitivity of power consumption adjusting parameter,
Corresponding weighted value is bigger;Conversely, power consumption Prediction Parameters are lower with the changing sensitivity of power consumption adjusting parameter, corresponding weight
Value is just smaller.
Further, on the basis of power consumption prediction model, the process that can predict with power consumption constantly corrects power consumption prediction mould
Type, to improve the precision of prediction of power consumption prediction model.In a kind of optional embodiment, it can be combined during power consumption is predicted
The actual power loss value of equipment to be processed constantly corrects power consumption prediction model.Wherein, mountable substrate management control in equipment to be processed
Device (Baseboard Management Controller, BMC) processed, BMC is generally mounted on the mainboard of equipment to be processed, main
To be used to acquire the actual power loss value of mainboard.Based on this, in model predictive process, can pass through according to the setting period to be processed
BMC in equipment acquires the actual power loss value of equipment to be processed, and according to collected actual power loss value of each period and the function
Consumption prediction model corresponds to each modulus of periodicity type prediction result, the model parameter in power consumption prediction model is constantly adjusted, until function
Consume the model prediction result of prediction model with by the difference of the collected actual power loss value of MBC less than set difference threshold as
Only.Wherein, power consumption prediction model, which corresponds to each modulus of periodicity type prediction result, can be power consumption prediction model according to the corresponding period
Interior power consumption Prediction Parameters treat the result that processing equipment carries out power consumption prediction.
Based on above-mentioned, the application another exemplary embodiment provides a kind of power consumption control method, as shown in figure 3, this method
The following steps are included:
301, from the device parameter of equipment to be processed, the power consumption adjusting parameter of power consumption adjusting stage is determined.
302, it from other device parameters in addition to power consumption adjusting parameter, determines the variation with power consumption adjusting parameter and becomes
Change, and changing sensitivity is greater than the device parameter of respective threshold as power consumption Prediction Parameters.
303, according to preset Parameter Switch relationship, the non-power consumption shape parameter in power consumption Prediction Parameters is converted into target function
Consume shape parameter.
304, the original power consumption shape parameter in target power consumption shape parameter and power consumption Prediction Parameters is entered ginseng as model to run
Power consumption prediction model, to obtain the prediction power consumption number of equipment to be processed.
If 305, collecting the actual power loss value of equipment to be processed by BMC when obtaining prediction power consumption number, then obtain pre-
Survey the difference of power consumption number and actual power loss value.
306, judge to predict the difference threshold whether difference of power consumption number and actual power loss value is greater than setting;If it has, then
Step 307 is executed, if it has not, thening follow the steps 308.
307, the model parameter in power consumption prediction model is adjusted, to improve the precision of power consumption prediction model, and return step
302。
308, judge whether to receive external power consumption adjustment instruction;If the determination result is YES, 309 are thened follow the steps;If
Judging result is no, then return step 302.
309, power consumption adjusting parameter is adjusted according to prediction power consumption number, to control the power consumption of equipment to be processed.
In the present embodiment, equipment to be processed includes BMC, and BMC can be with lower frequency continuous collecting equipment to be processed
Actual power loss value.In the power consumption forecast period of step 302-307 description, if when obtaining the prediction power consumption number of equipment to be processed,
BMC collects equipment to be processed in the actual power loss value at corresponding moment, then may compare the size of the two power consumption numbers, and by the two
Difference be compared with the difference threshold of setting.Wherein, which can characterize the actual power loss value of equipment to be processed
With the degree of closeness of prediction power consumption number.The difference threshold is empirical value, and the embodiment of the present application is with no restrictions.
When the difference for predicting power consumption number and actual power loss value is greater than the difference threshold of setting, it is believed that prediction power consumption number goes out
Biggish deviation is showed, has not conformed to the actual conditions.Then, the model parameter in adjustable power consumption prediction model constantly predicts mould to power consumption
Type optimizes, and the prediction result of the power consumption prediction model after optimization and the difference of the collected actual power loss value of BMC contract
It is small in the difference threshold of setting, provide more accurate prediction power consumption number, for the power consumption adjusting stage in order to further increase
Power consumption control effect.
Optionally, when adjusting the model parameter in power consumption prediction model, Variable Control method can be used.By control wherein certain
The corresponding model parameter of power consumption Prediction Parameters is constant a bit, changes the mode of other corresponding model parameters of power consumption Prediction Parameters, reaches
To the purpose of optimization power consumption prediction model.Wherein, change the model parameter of which power consumption Prediction Parameters, and keep which power consumption pre-
The model parameter for surveying parameter is constant, can determine, not limit this depending on concrete application demand, or at random.Certainly, in reality
In the application of border, other parameter regulation means can also be used, the present embodiment is with no restrictions.
When receiving external power consumption adjustment instruction, the power consumption adjusting stage can be entered, and according to power consumption forecast period
Prediction power consumption number power consumption adjusting parameter is adjusted, to control the power consumption of equipment to be processed.
In some processes of the description in above-described embodiment and attached drawing, the multiple behaviour occurred according to particular order are contained
Make, but it should be clearly understood that these operations can not be executed according to its sequence what appears in this article or be executed parallel, behaviour
Serial number of work such as 101,102 etc. is only used for distinguishing each different operation, and it is suitable that serial number itself does not represent any execution
Sequence.In addition, these processes may include more or fewer operations, and these operations can be executed in order or be held parallel
Row.It should be noted that the description such as herein " first ", " second ", be for distinguishing different message, equipment, module etc.,
Sequencing is not represented, " first " and " second " is not also limited and is different type.
Embodiment above describes the processes of power consumption control method.When disposing implementation, the above method be can be applied to wait locate
It manages and is realized inside equipment;Or it also can be applied in a power consumption control system realize.
In some optional embodiments, power consumption prediction provided by the above embodiment or power consumption control method are applied to be processed
It is realized inside equipment.Wherein, the internal frame diagram of equipment to be processed as shown in fig. 4 a, below with reference to internal frame diagram pair shown in Fig. 4 a
Power consumption control process is described in detail.
As shown in fig. 4 a, equipment to be processed includes: power consumption prediction module 41 and power consumption adjustment module 42.Wherein, power consumption
Prediction module 41 may include parameter collection submodule 411, computational submodule 412 and power consumption prediction model.In addition, such as Fig. 4 a institute
Show, equipment to be processed further includes MSR, system file etc..MSR be stored with the temperature of CPU, the RAPL value of CPU, it is memory temperature, interior
Deposit the register parameters such as power consumption;The system parameters such as memory usage, network interface card I/O performance parameter are stored in system file.
It is alternatively possible to which but it is not limited to this using cpu frequency as the power consumption adjusting parameter of power consumption adjusting stage.
Parameter collection submodule 411 is mainly used for reading cpu temperature, memory temperature, CPU power consumption, memory function from MSR
The parameters such as consumption, and/or, the property such as memory usage, magnetic disc i/o performance parameter, network interface card I/O performance parameter are read from system file
Energy parameter, the power consumption Prediction Parameters as power consumption forecast period.
Wherein, power consumption Prediction Parameters can be supplied to meter after obtaining power consumption Prediction Parameters by parameter collection submodule 411
Operator module 412.Computational submodule 412 is mainly used for being calculated according to the power consumption Prediction Parameters that parameter collection submodule 411 provides
The prediction power consumption number of equipment to be processed, and prediction power consumption number is supplied to power consumption adjustment module 42.Optionally, as shown in fig. 4 a,
First interface 43 is equipped between power consumption prediction module 41 and power consumption control module 42.Based on this, power consumption control module 42 can lead to
It crosses first interface 43 and obtains the calculated prediction power consumption number of computational submodule 412.Wherein, first interface 43 can be hardware and connect
Mouthful, it is also possible to software interface, the embodiment of the present application is with no restrictions.
Power consumption adjustment module 42 is mainly used for the prediction power consumption number provided according to computational submodule 412 and adjusts ginseng to power consumption
Number, such as cpu frequency are adjusted, to control the power consumption of equipment to be processed.Optionally, power consumption adjustment module 42 can predicted
When power consumption number is greater than setting power consumption threshold value, the cpu frequency of equipment to be processed is reduced, to reduce the power consumption of equipment to be processed.
In above process, computational submodule 412 is particularly used in: entering ginseng operation function for power consumption Prediction Parameters as model
Prediction model is consumed, to obtain the prediction power consumption number of equipment to be processed.
Further, computational submodule 412 can be according to preset Parameter Switch relationship, by the non-power consumption in power consumption Prediction Parameters
Shape parameter is converted to target power consumption shape parameter;Later, by the original power consumption shape parameter and target power consumption in power consumption Prediction Parameters
Shape parameter enters ginseng operation power consumption prediction model as model, to obtain the prediction power consumption number of equipment to be processed.
Still optionally further, as shown in fig. 4 a, which further includes BMC 40, BMC 40 be mainly used to acquisition to
The actual power loss value of processing equipment.As shown in fig. 4 a, power consumption adjustment module 42 can be connect with the BMC 40 in equipment to be processed, and
The actual power loss value of equipment to be processed can be acquired with lower frequency (set the period) by BMC 40.Correspondingly, power consumption adjusts
Module 42 can also provide the collected actual power loss value of BMC 40 to computational submodule 412, in order to 412 basis of computational submodule
Actual power loss value is trained and corrects to power consumption prediction model.As shown in fig. 4 a, power consumption prediction module 41 and power consumption control mould
Second interface 44 is additionally provided between block 42.Wherein, computational submodule 412 can be by second interface 44 from power consumption control module 42
Read the actual power loss value that BMC 40 is got.Optionally, second interface 44 can be hardware interface, is also possible to software and connects
Mouthful, the embodiment of the present application is with no restrictions.
For computational submodule 412, in the model prediction stage, when calculating the prediction power consumption number of equipment to be processed,
The collected equipment to be processed of BMC 40 is read in the practical function at corresponding moment from power consumption control module 42 by second interface 44
Then consumption value obtains the difference of prediction power consumption number and actual power loss value;When the difference is greater than the difference threshold of setting, function is adjusted
The model parameter in prediction model is consumed, to optimize power consumption prediction model, improves the precision of prediction of power consumption prediction model.
Wherein, the interaction flow in internal frame diagram shown in Fig. 4 a between each module, as shown in Figure 4 b:
In power consumption forecast period, parameter collection submodule 411 inputs collected power consumption prediction ginseng to computational submodule 412
Number;Computational submodule 412 exports power consumption Prediction Parameters to power consumption prediction model;Power consumption prediction model is according to power consumption Prediction Parameters
Power consumption prediction result is obtained, and result is inputted into computational submodule 412, so that computational submodule 412 determines prediction power consumption number.
Optionally, after computational submodule 412 obtains prediction power consumption number, the prediction power consumption number directly can be written to first
Interface 43 reads prediction power consumption number from first interface 43 so that power consumption adjusts module 42.Or
As shown in Figure 4 b, in the power consumption forecast period of the above process, power consumption adjusts module 42 can be collected by BMC 40
The actual power loss value of equipment to be processed is written to second interface 44, and parameter collection submodule 411 is read to from from second interface 44
The actual power loss value of equipment is managed, and the actual power loss value is inputted into computational submodule 412.Computational submodule 412 can be according to the reality
Border power consumption number optimizes power consumption prediction model, and regains prediction power consumption number using the power consumption prediction model after optimization,
Function prediction consumption value is write into value first interface 43, reads prediction power consumption number from first interface 43 so that power consumption adjusts module 42.
In the power consumption adjusting stage, power consumption adjustment module 42 can receive the power consumption adjustment instruction of user's input, and receive
When power consumption adjustment instruction, prediction power consumption number is read from first interface 43, to carry out power consumption adjustment according to the prediction power consumption number.At this
The prediction power consumption number being calculated constantly can be written to first interface 43 by a stage, computational submodule 412, so that power consumption adjusts mould
Block 42 is read.
In other alternative embodiments, power consumption control method provided by the above embodiment is applied to a function shown in Fig. 4 c
Consumption control system.As illustrated in fig. 4 c, which includes: the first electronic equipment 401 and the second electronic equipment 402.Second electronics is set
Standby 402 be equipment to be processed, and the first electronic equipment 401 can carry out power consumption control to the second electronic equipment 402.
During each power consumption control, the first electronic equipment 401 initially enters power consumption forecast period, predicts rank in power consumption
Section needs the parameter adjusted, i.e. power consumption adjusting parameter in conjunction with the power consumption adjusting stage, from the device parameter of the second electronic equipment 402
The middle variation obtained to power consumption adjusting parameter is than more sensitive some device parameters as power consumption Prediction Parameters;Then, according to function
It consumes Prediction Parameters and power consumption prediction is carried out to the second electronic equipment 402, obtain prediction power consumption number.Further, electric to first when needing
When sub- equipment 401 carries out power consumption adjustment, such as the power consumption adjustment period of setting reaches, or receives external power consumption adjustment and refer to
When enabling, into the power consumption adjusting stage;In the power consumption adjusting stage, the first electronic equipment 401 is adaptively adjusted according to prediction power consumption number
Power consumption adjusting parameter, to achieve the purpose that control the power consumption of the second electronic equipment 402.
Wherein, the first electronic equipment 401 selection power consumption Prediction Parameters, calculating prediction power consumption number and adjustment power consumption adjustment ginseng
The detailed realization of the operations such as number, reference can be made to the description in previous embodiment, details are not described herein.
Fig. 5 is the structural schematic diagram for the electronic equipment that the application another exemplary embodiment provides.As shown in figure 5, the electricity
Sub- equipment includes: memory 51 and processor 52.
Memory 51 can be configured to store various other data for storing one or more computer instruction to prop up
Hold operation on an electronic device.The example of these data includes for any application program operated on an electronic device or side
The instruction of method.
Memory 51 can be by any kind of volatibility or non-volatile memory device or their combination realization, such as
Static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only to be deposited
Reservoir (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or
CD.
In some exemplary embodiments, it includes the memory remotely located relative to processor 52 that memory 51 is optional,
These remote memories can be communicated to connect by network and processor 52.The example of above-mentioned network includes but is not limited to interconnect
Net, intranet, local area network, mobile radio communication and combinations thereof.
Processor 52 is coupled with memory 51, for executing one or more computer instruction to be used for: from electronic equipment
Device parameter in, obtain power consumption Prediction Parameters;Power consumption Prediction Parameters refer to the variation of power consumption adjusting parameter and change, and become
Change the device parameter that sensitivity is greater than respective threshold;Processing equipment is treated according to power consumption Prediction Parameters and carries out power consumption prediction, to obtain
Obtain the prediction power consumption number of equipment to be processed.
In some exemplary embodiments, processor 52 is also used to before obtaining power consumption Prediction Parameters, is set to be processed
In standby device parameter, power consumption adjusting parameter is determined.
Still optionally further, processor 52 is specifically used for: from the device parameter of electronic equipment, determining cpu frequency conduct
Power consumption adjusting parameter;And from other device parameters in addition to cpu frequency, acquisition change with the variation of cpu frequency and
Changing sensitivity is greater than the device parameter of respective threshold as power consumption Prediction Parameters.
Still optionally further, as shown in figure 5, including MSR 57 inside the processor 52.Based on this, processor 52 is specifically used
In: at least one register parameters are read from the MSR 57 for including inside processor 52;And/or the system from electronic equipment
At least one system parameter is read in file, at least one system parameter includes other system parameters in addition to cpu busy percentage,
Or it is not maximum that at least one system parameter, which includes the different degree accounting of cpu busy percentage and cpu busy percentage in power consumption prediction,
And be less than default accounting threshold value.
In some exemplary embodiments, processor 52 is specifically used for when obtaining prediction power consumption number: power consumption is predicted to join
Number enters ginseng operation power consumption prediction model as model, to obtain prediction power consumption number.
Further, processor 52 is specifically used for: according to preset Parameter Switch relationship, by the NOT function in power consumption Prediction Parameters
Consumption shape parameter is converted to target power consumption shape parameter;By the original power consumption shape parameter and target power consumption type ginseng in power consumption Prediction Parameters
Number enters ginseng operation power consumption prediction model as model, to obtain prediction power consumption number.
Still optionally further, as shown in figure 5, the electronic equipment further include: BMC 58.Based on this, processor 52 is also used to:
The actual power loss value of equipment to be processed is acquired to set the period by baseboard management controller BMC;It is collected according to each period
Actual power loss value and power consumption prediction model correspond to each modulus of periodicity type prediction result, constantly adjust power consumption prediction model in
Model parameter, until the model prediction result of power consumption prediction model is less than with the difference by the collected actual power loss value of MBC
Until setting difference threshold.
In some exemplary embodiments, processor 52 is also used to: being carried out according to prediction power consumption number to power consumption adjusting parameter
Adjustment, to control the power consumption of equipment to be processed.
Still optionally further, processor 52 is specifically used for when being adjusted to power consumption adjusting parameter: if prediction power consumption number
Greater than setting power consumption threshold value, then the cpu frequency of electronic equipment is reduced.
Further, as shown in figure 5, the electronic equipment further include: communication component 53, display 54, power supply module 55, audio
Other components such as component 56.Members are only schematically provided in Fig. 5, are not meant to that electronic equipment only includes group shown in Fig. 5
Part.
Power consumption prediction technique or power consumption control provided by the embodiment of the present application can be performed in electronic equipment provided in this embodiment
Method processed, working principle and beneficial effect, reference can be made to the description in above method embodiment, details are not described herein.
The embodiment of the present application also provides a kind of computer readable storage medium for being stored with computer program, the computer
Program is performed the step that can be realized in power consumption prediction technique provided by the embodiments of the present application.
Fig. 6 is the structural schematic diagram for the electronic equipment that the application another exemplary embodiment provides.As shown in fig. 6, the electricity
Sub- equipment includes: memory 61 and processor 62.
Memory 61 can be configured to store various other data for storing one or more computer instruction to prop up
Hold operation on an electronic device.The example of these data includes for any application program operated on an electronic device or side
The instruction of method.
Memory 61 can be by any kind of volatibility or non-volatile memory device or their combination realization, such as
Static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only to be deposited
Reservoir (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or
CD.
In some exemplary embodiments, it includes the memory remotely located relative to processor 62 that memory 61 is optional,
These remote memories can be communicated to connect by network and processor 62.The example of above-mentioned network includes but is not limited to interconnect
Net, intranet, local area network, mobile radio communication and combinations thereof.
Processor 62 is coupled with memory 61, for executing one or more computer instruction to be used for:
From the device parameter of electronic equipment, power consumption Prediction Parameters are obtained;Power consumption Prediction Parameters, which refer to adjust with power consumption, joins
Several variations and change, and changing sensitivity be greater than respective threshold device parameter;It is set according to power consumption Prediction Parameters to be processed
It is standby to carry out power consumption prediction, to obtain the prediction power consumption number of equipment to be processed;Power consumption adjusting parameter is carried out according to prediction power consumption number
Adjustment, to control the power consumption of equipment to be processed.
In some exemplary embodiments, processor 62 is also used to before obtaining power consumption Prediction Parameters, is set to be processed
In standby device parameter, power consumption adjusting parameter is determined.
Still optionally further, processor 62 is specifically used for: from the device parameter of electronic equipment, determining cpu frequency conduct
Power consumption adjusting parameter;And from other device parameters in addition to cpu frequency, acquisition change with the variation of cpu frequency and
Changing sensitivity is greater than the device parameter of respective threshold as power consumption Prediction Parameters.
Still optionally further, as shown in fig. 6, including MSR 67 inside the processor 62.Based on this, processor 62 is specifically used
In: at least one register parameters are read from the MSR 67 for including inside processor 62;And/or the system from electronic equipment
At least one system parameter is read in file, at least one system parameter includes other system parameters in addition to cpu busy percentage,
Or it is not maximum that at least one system parameter, which includes the different degree accounting of cpu busy percentage and cpu busy percentage in power consumption prediction,
And be less than default accounting threshold value.
In some exemplary embodiments, processor 62 is specifically used for when obtaining prediction power consumption number: power consumption is predicted to join
Number enters ginseng operation power consumption prediction model as model, to obtain prediction power consumption number.
Further, processor 62 is specifically used for: according to preset Parameter Switch relationship, by the NOT function in power consumption Prediction Parameters
Consumption shape parameter is converted to target power consumption shape parameter;By the original power consumption shape parameter and target power consumption type ginseng in power consumption Prediction Parameters
Number enters ginseng operation power consumption prediction model as model, to obtain prediction power consumption number.
Still optionally further, as shown in fig. 6, the electronic equipment further include: BMC 68.Based on this, processor 62 is also used to:
The actual power loss value of equipment to be processed is acquired to set the period by baseboard management controller BMC;It is collected according to each period
Actual power loss value and power consumption prediction model correspond to each modulus of periodicity type prediction result, constantly adjust power consumption prediction model in
Model parameter, until the model prediction result of power consumption prediction model is less than with the difference by the collected actual power loss value of MBC
Until setting difference threshold.
Still optionally further, processor 62 is specifically used for when being adjusted to power consumption adjusting parameter: if prediction power consumption number
Greater than setting power consumption threshold value, then the cpu frequency of electronic equipment is reduced.
Further, as shown in fig. 6, the electronic equipment further include: communication component 63, display 64, power supply module 66, audio
Other components such as component 66.Members are only schematically provided in Fig. 6, are not meant to that electronic equipment only includes group shown in Fig. 6
Part.
Power consumption control method provided by the embodiment of the present application can be performed in electronic equipment provided in this embodiment, and work is former
Reason and beneficial effect, reference can be made to the description in above method embodiment, details are not described herein.
The embodiment of the present application also provides a kind of computer readable storage medium for being stored with computer program, the computer
Program is performed the step that can be realized in power consumption control method provided by the embodiments of the present application.
Communication component in Fig. 5 or Fig. 6 can be configured to convenient for wired between communication component place equipment and other equipment
Or the communication of wireless mode.Equipment where communication component can access the wireless network based on communication standard, such as WiFi, 2G or
3G or their combination.In one exemplary embodiment, communication component receives via broadcast channel and comes from external broadcasting management
The broadcast singal or broadcast related information of system.In one exemplary embodiment, communication component further includes near-field communication (NFC)
Module, to promote short range communication.For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) can be based in NFC module
Technology, ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies are realized.
Display in Fig. 5 or Fig. 6 includes screen, and screen may include liquid crystal display (LCD) and touch panel
(TP).If screen includes touch panel, screen may be implemented as touch screen, to receive input signal from the user.Touching
Touching panel includes one or more touch sensors to sense the gesture on touch, slide, and touch panel.Touch sensor can
Not only to sense the boundary of a touch or slide action, but also detect the duration associated with the touch or slide operation and
Pressure.
Power supply module in Fig. 5 or Fig. 6, the various assemblies of equipment provide electric power where power supply module.Power supply module can
To include power-supply management system, one or more power supplys and other with for equipment where power supply module generate, manage, and distribute electricity
The associated component of power.
Audio component in Fig. 5 or Fig. 6, can be configured to output and/or input audio signal.For example, audio component packet
A microphone (MIC) is included, the equipment where audio component is in operation mode, as call model, logging mode and voice are known
When other mode, microphone is configured as receiving external audio signal.The received audio signal can be further stored in and deposit
Reservoir is sent via communication component.In some embodiments, audio component further includes a loudspeaker, for exporting audio letter
Number.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (20)
1. a kind of power consumption prediction technique characterized by comprising
Power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed;The power consumption Prediction Parameters, which refer to adjust with power consumption, joins
Several variations and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the pre- of the equipment to be processed
Survey power consumption number.
2. the method according to claim 1, wherein it is pre- to obtain power consumption in the device parameter from equipment to be processed
It surveys before parameter, the method also includes:
The power consumption adjusting parameter is determined from the device parameter of the equipment to be processed.
3. according to the method described in claim 2, it is characterized in that, described in being determined from the device parameter of the equipment to be processed
Power consumption adjusting parameter, comprising:
From the device parameter of the equipment to be processed, determine cpu frequency as the power consumption adjusting parameter;
It is described that power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed, comprising:
From other device parameters in addition to the cpu frequency, acquisition changes with the variation of the cpu frequency and changes spirit
Sensitivity is greater than the device parameter of respective threshold as the power consumption Prediction Parameters.
4. according to the method described in claim 3, it is characterized in that, other equipment ginseng from addition to the cpu frequency
In number, acquisition changes with the variation of the cpu frequency and changing sensitivity is greater than described in the device parameter conduct of respective threshold
Power consumption Prediction Parameters, comprising:
At least one register parameters are read from the special module register of the equipment to be processed;
And/or
At least one system parameter is read from the system file of the equipment to be processed, at least one system parameter includes
Other system parameters or at least one system parameter in addition to cpu busy percentage include cpu busy percentage and other systems
System parameter, and different degree accounting of the cpu busy percentage in power consumption prediction is not the largest and is less than default accounting threshold value.
5. the method according to claim 1, wherein it is described according to the power consumption Prediction Parameters to described to be processed
Equipment carries out power consumption prediction, to obtain the prediction power consumption number of the equipment to be processed, comprising:
Enter ginseng operation power consumption prediction model for the power consumption Prediction Parameters as model, to obtain the prediction power consumption number.
6. according to the method described in claim 5, it is characterized in that, described enter ginseng fortune for the power consumption Prediction Parameters as model
Row power consumption prediction model, to obtain the prediction power consumption number, comprising:
According to preset Parameter Switch relationship, the non-power consumption shape parameter in the power consumption Prediction Parameters is converted into target power consumption type
Parameter;
Using in the power consumption Prediction Parameters original power consumption shape parameter and the target power consumption shape parameter as model enter ginseng fortune
The row power consumption prediction model, to obtain the prediction power consumption number.
7. according to the method described in claim 5, it is characterized in that, being run the power consumption Prediction Parameters are entered ginseng as model
Power consumption prediction model, before obtaining the prediction power consumption number, the method also includes:
The actual power loss value of the equipment to be processed is acquired to set the period by baseboard management controller BMC;
Each modulus of periodicity type prediction knot is corresponded to according to collected actual power loss value of each period and the power consumption prediction model
Fruit constantly adjusts the model parameter in the power consumption prediction model, until the power consumption prediction model model prediction result with
Until being less than setting difference threshold by the difference of the collected actual power loss value of the MBC.
8. method according to claim 1-7, which is characterized in that in the pre- measurement of power for obtaining the equipment to be processed
After consumption value, the method also includes:
The power consumption adjusting parameter is adjusted according to the prediction power consumption number, to control the power consumption of the equipment to be processed.
9. according to the method described in claim 8, it is characterized in that, according to the prediction power consumption number to the power consumption adjusting parameter
It is adjusted, to control the power consumption of the equipment to be processed, comprising:
In response to power consumption adjustment instruction, the prediction power consumption number is compared with setting power consumption threshold value;
If the prediction power consumption number be greater than the setting power consumption threshold value, adjust the power consumption adjusting parameter, with reduce described in
The power consumption of processing equipment.
10. a kind of electronic equipment characterized by comprising memory and processor;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
From the device parameter of the electronic equipment, power consumption Prediction Parameters are obtained;The power consumption Prediction Parameters refer to power consumption tune
The variation of whole parameter and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the pre- of the equipment to be processed
Survey power consumption number.
11. electronic equipment according to claim 10, which is characterized in that the processor is also used to:
Before obtaining the power consumption Prediction Parameters, from the device parameter of the equipment to be processed, the power consumption adjustment is determined
Parameter.
12. electronic equipment according to claim 11, which is characterized in that the processor is specifically used for:
From the device parameter of the electronic equipment, determine cpu frequency as the power consumption adjusting parameter;And
From other device parameters in addition to the cpu frequency, acquisition changes with the variation of the cpu frequency and changes spirit
Sensitivity is greater than the device parameter of respective threshold as the power consumption Prediction Parameters.
13. electronic equipment according to claim 12, which is characterized in that the processor is specifically used for:
At least one register parameters are read from the special module register for including inside the processor;
And/or
At least one system parameter is read from the system file of the electronic equipment, at least one system parameter includes removing
Other system parameters or at least one system parameter except cpu busy percentage include cpu busy percentage and CPU benefit
It is not the largest with different degree accounting of the rate in power consumption prediction and is less than default accounting threshold value.
14. the described in any item electronic equipments of 0-13 according to claim 1, which is characterized in that the processor is specifically used for:
Enter ginseng operation power consumption prediction model for the power consumption Prediction Parameters as model, to obtain the prediction power consumption number.
15. electronic equipment according to claim 14, which is characterized in that the processor is specifically used for:
According to preset Parameter Switch relationship, the non-power consumption shape parameter in the power consumption Prediction Parameters is converted into target power consumption type
Parameter;
Using in the power consumption Prediction Parameters original power consumption shape parameter and the target power consumption shape parameter as model enter ginseng fortune
The row power consumption prediction model, to obtain the prediction power consumption number.
16. electronic equipment according to claim 14, which is characterized in that the processor is also used to:
The actual power loss value of the equipment to be processed is acquired to set the period by baseboard management controller BMC;
Each modulus of periodicity type prediction knot is corresponded to according to collected actual power loss value of each period and the power consumption prediction model
Fruit constantly adjusts the model parameter in the power consumption prediction model, until the power consumption prediction model model prediction result with
Until being less than setting difference threshold by the difference of the collected actual power loss value of the MBC.
17. a kind of computer readable storage medium for being stored with computer program, which is characterized in that the computer program is held
It can be realized the step in any one of claim 1-9 the method when row.
18. a kind of power consumption control method characterized by comprising
Power consumption Prediction Parameters are obtained from the device parameter of equipment to be processed;The power consumption Prediction Parameters, which refer to adjust with power consumption, joins
Several variations and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the pre- of the equipment to be processed
Survey power consumption number;
The power consumption adjusting parameter is adjusted according to the prediction power consumption number, to control the power consumption of the equipment to be processed.
19. a kind of electronic equipment characterized by comprising memory and processor;
The memory, for storing computer program;
The processor is coupled with the memory, for executing the computer program, to be used for:
From the device parameter of the electronic equipment, power consumption Prediction Parameters are obtained;The power consumption Prediction Parameters refer to power consumption tune
The variation of whole parameter and change, and changing sensitivity be greater than respective threshold device parameter;
Power consumption prediction is carried out to the equipment to be processed according to the power consumption Prediction Parameters, to obtain the pre- of the equipment to be processed
Survey power consumption number;
The power consumption adjusting parameter is adjusted according to the prediction power consumption number, to control the power consumption of the equipment to be processed.
20. a kind of computer readable storage medium for being stored with computer program, which is characterized in that the computer program is held
It can be realized the step in claim 18 the method when row.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810241512.5A CN110308782B (en) | 2018-03-22 | 2018-03-22 | Power consumption prediction and control method and device and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810241512.5A CN110308782B (en) | 2018-03-22 | 2018-03-22 | Power consumption prediction and control method and device and computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110308782A true CN110308782A (en) | 2019-10-08 |
CN110308782B CN110308782B (en) | 2023-09-08 |
Family
ID=68073443
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810241512.5A Active CN110308782B (en) | 2018-03-22 | 2018-03-22 | Power consumption prediction and control method and device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110308782B (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111026328A (en) * | 2019-11-13 | 2020-04-17 | 华为技术有限公司 | Energy consumption control method and device and storage system |
CN111026603A (en) * | 2019-10-25 | 2020-04-17 | 南京大学 | On-chip network temperature prediction method and device, equipment and storage medium |
CN111478824A (en) * | 2020-03-20 | 2020-07-31 | 苏州浪潮智能科技有限公司 | Network card power consumption testing method, device and system |
CN111652407A (en) * | 2020-04-13 | 2020-09-11 | 北京旷视机器人技术有限公司 | Method, device, medium, electronic equipment and system for processing tasks in warehouse |
CN112100910A (en) * | 2020-09-07 | 2020-12-18 | 海光信息技术有限公司 | Power consumption model training method, power consumption testing method and device for processor |
CN112243060A (en) * | 2020-10-30 | 2021-01-19 | Oppo广东移动通信有限公司 | Power consumption estimation method of processor, mobile terminal and computer storage medium |
CN112486683A (en) * | 2020-11-27 | 2021-03-12 | 中国科学技术大学先进技术研究院 | Processor control method, control device, and computer-readable storage medium |
CN112665191A (en) * | 2020-12-01 | 2021-04-16 | 珠海格力电器股份有限公司 | Control method and device of heat pump water heater and heat pump water heater |
CN113741674A (en) * | 2021-09-13 | 2021-12-03 | 西安超越申泰信息科技有限公司 | Power consumption control method, system, terminal and storage medium for notebook computer |
CN114138095A (en) * | 2022-01-29 | 2022-03-04 | 阿里巴巴(中国)有限公司 | Power consumption processing method and device for internet data center IDC and readable medium |
CN114489307A (en) * | 2022-02-14 | 2022-05-13 | 阿里巴巴(中国)有限公司 | Energy efficiency optimization method and device for internet data center |
CN114924637A (en) * | 2022-04-02 | 2022-08-19 | 联想(北京)有限公司 | Control method and electronic equipment |
CN115993943A (en) * | 2023-03-23 | 2023-04-21 | 苏州宽温电子科技有限公司 | Memory characterization method, chip design method, medium and device |
CN116204387A (en) * | 2023-04-26 | 2023-06-02 | 之江实验室 | Chip current prediction method and device, medium and electronic equipment |
CN116635813A (en) * | 2020-11-10 | 2023-08-22 | 超威半导体公司 | Application coverage for power estimation mechanism |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2264794A (en) * | 1992-03-06 | 1993-09-08 | Intel Corp | Low power consumption chip controller |
US20100286974A1 (en) * | 2009-05-05 | 2010-11-11 | Sun Microsystems, Inc. | Technique using power macromodeling for register transfer level power estimation |
US20110106286A1 (en) * | 2009-11-05 | 2011-05-05 | Qualcomm Incorporated | Generating A Power Model For An Electronic Device |
US20120053925A1 (en) * | 2010-08-31 | 2012-03-01 | Steven Geffin | Method and System for Computer Power and Resource Consumption Modeling |
US20120159201A1 (en) * | 2010-12-21 | 2012-06-21 | Eric Distefano | Method and apparatus to configure thermal design power in a microprocessor |
CN105243068A (en) * | 2014-07-09 | 2016-01-13 | 华为技术有限公司 | Database system query method, server and energy consumption test system |
CN105988918A (en) * | 2015-02-26 | 2016-10-05 | 阿里巴巴集团控股有限公司 | GPU fault prediction method and device |
JPWO2016031065A1 (en) * | 2014-08-29 | 2017-04-27 | 三菱電機株式会社 | Power consumption estimation device, device management system, power consumption estimation method and program |
US20170192485A1 (en) * | 2016-01-04 | 2017-07-06 | International Business Machines Corporation | Providing a power optimized design for a device |
-
2018
- 2018-03-22 CN CN201810241512.5A patent/CN110308782B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2264794A (en) * | 1992-03-06 | 1993-09-08 | Intel Corp | Low power consumption chip controller |
US20100286974A1 (en) * | 2009-05-05 | 2010-11-11 | Sun Microsystems, Inc. | Technique using power macromodeling for register transfer level power estimation |
US20110106286A1 (en) * | 2009-11-05 | 2011-05-05 | Qualcomm Incorporated | Generating A Power Model For An Electronic Device |
US20120053925A1 (en) * | 2010-08-31 | 2012-03-01 | Steven Geffin | Method and System for Computer Power and Resource Consumption Modeling |
US20120159201A1 (en) * | 2010-12-21 | 2012-06-21 | Eric Distefano | Method and apparatus to configure thermal design power in a microprocessor |
CN105243068A (en) * | 2014-07-09 | 2016-01-13 | 华为技术有限公司 | Database system query method, server and energy consumption test system |
JPWO2016031065A1 (en) * | 2014-08-29 | 2017-04-27 | 三菱電機株式会社 | Power consumption estimation device, device management system, power consumption estimation method and program |
CN105988918A (en) * | 2015-02-26 | 2016-10-05 | 阿里巴巴集团控股有限公司 | GPU fault prediction method and device |
US20170192485A1 (en) * | 2016-01-04 | 2017-07-06 | International Business Machines Corporation | Providing a power optimized design for a device |
Non-Patent Citations (2)
Title |
---|
陈志韬等: "商用处理器上针对能耗优化的DVFS调节机制", 《计算机工程》 * |
陈志韬等: "商用处理器上针对能耗优化的DVFS调节机制", 《计算机工程》, no. 03, 15 March 2017 (2017-03-15) * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111026603A (en) * | 2019-10-25 | 2020-04-17 | 南京大学 | On-chip network temperature prediction method and device, equipment and storage medium |
CN111026603B (en) * | 2019-10-25 | 2021-04-23 | 南京大学 | On-chip network temperature prediction method and device, equipment and storage medium |
CN111026328A (en) * | 2019-11-13 | 2020-04-17 | 华为技术有限公司 | Energy consumption control method and device and storage system |
CN111478824A (en) * | 2020-03-20 | 2020-07-31 | 苏州浪潮智能科技有限公司 | Network card power consumption testing method, device and system |
CN111652407A (en) * | 2020-04-13 | 2020-09-11 | 北京旷视机器人技术有限公司 | Method, device, medium, electronic equipment and system for processing tasks in warehouse |
CN111652407B (en) * | 2020-04-13 | 2023-09-05 | 北京旷视机器人技术有限公司 | Task processing method, device, medium, electronic equipment and system in warehouse |
CN112100910A (en) * | 2020-09-07 | 2020-12-18 | 海光信息技术有限公司 | Power consumption model training method, power consumption testing method and device for processor |
CN112243060B (en) * | 2020-10-30 | 2022-10-21 | Oppo广东移动通信有限公司 | Power consumption estimation method of processor, mobile terminal and computer storage medium |
CN112243060A (en) * | 2020-10-30 | 2021-01-19 | Oppo广东移动通信有限公司 | Power consumption estimation method of processor, mobile terminal and computer storage medium |
CN116635813A (en) * | 2020-11-10 | 2023-08-22 | 超威半导体公司 | Application coverage for power estimation mechanism |
CN112486683B (en) * | 2020-11-27 | 2023-05-19 | 中国科学技术大学先进技术研究院 | Processor control method, control apparatus, and computer-readable storage medium |
CN112486683A (en) * | 2020-11-27 | 2021-03-12 | 中国科学技术大学先进技术研究院 | Processor control method, control device, and computer-readable storage medium |
CN112665191A (en) * | 2020-12-01 | 2021-04-16 | 珠海格力电器股份有限公司 | Control method and device of heat pump water heater and heat pump water heater |
CN113741674A (en) * | 2021-09-13 | 2021-12-03 | 西安超越申泰信息科技有限公司 | Power consumption control method, system, terminal and storage medium for notebook computer |
CN114138095A (en) * | 2022-01-29 | 2022-03-04 | 阿里巴巴(中国)有限公司 | Power consumption processing method and device for internet data center IDC and readable medium |
CN114489307A (en) * | 2022-02-14 | 2022-05-13 | 阿里巴巴(中国)有限公司 | Energy efficiency optimization method and device for internet data center |
CN114924637A (en) * | 2022-04-02 | 2022-08-19 | 联想(北京)有限公司 | Control method and electronic equipment |
CN114924637B (en) * | 2022-04-02 | 2024-02-27 | 联想(北京)有限公司 | Control method and electronic equipment |
CN115993943A (en) * | 2023-03-23 | 2023-04-21 | 苏州宽温电子科技有限公司 | Memory characterization method, chip design method, medium and device |
CN116204387A (en) * | 2023-04-26 | 2023-06-02 | 之江实验室 | Chip current prediction method and device, medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN110308782B (en) | 2023-09-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110308782A (en) | Power consumption prediction, control method, equipment and computer readable storage medium | |
US10915156B2 (en) | Wearable device-aware supervised power management for mobile platforms | |
Pramanik et al. | Power consumption analysis, measurement, management, and issues: A state-of-the-art review of smartphone battery and energy usage | |
US9823673B2 (en) | Energy efficiency aware thermal management in a multi-processor system on a chip based on monitored processing component current draw | |
US20170139462A1 (en) | Datacenter power management optimizations | |
US20130097415A1 (en) | Central Processing Unit Monitoring and Management Based On A busy-Idle Histogram | |
US20170147055A1 (en) | Systems and methods for providing local hardware limit management and enforcement | |
US20130046967A1 (en) | Proactive Power Management Using a Power Management Unit | |
CN106415296A (en) | Adaptive battery life extension | |
EP2673993A1 (en) | Method and apparatus of smart power management for mobile communication terminals | |
CN107924225A (en) | System and method for dynamically adjusting memory state transition timers | |
CN109992370A (en) | Applied program processing method and device, electronic equipment, computer readable storage medium | |
CN103125132A (en) | Systems and methods for measuring the effectiveness of a workload predictor on a mobile device | |
CN108292157A (en) | Processor core energy management | |
CN106455024A (en) | Method for automatically adjusting communication frequency, and terminal equipment | |
CN110032321A (en) | Applied program processing method and device, electronic equipment, computer readable storage medium | |
CN101365192A (en) | Method and apparatus regulating frequency of embedded processor and mobile terminal processor | |
CN109710330A (en) | Method for determining operation parameters, device, terminal and the storage medium of application program | |
Obukhova et al. | Diagnostics of power consumption of a mobile device multi-core processor with detail of each core utilization | |
CN102866767A (en) | Semiconductor device, radio communication terminal using same, and clock frequency control method | |
CN104246653B (en) | For the method and apparatus making the power consumption of fixed frequency processing unit operation minimize | |
CN104699224A (en) | Power-saving control method | |
Nithi et al. | Smart power management for mobile handsets | |
CN108491000A (en) | device temperature adjusting method, device and panel type wireless access point | |
CN106407143A (en) | Method and device for controlling hot plug operation of CPU in mobile terminal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |