CN108427497A - Method and system for calculating power loss exponent - Google Patents
Method and system for calculating power loss exponent Download PDFInfo
- Publication number
- CN108427497A CN108427497A CN201710082014.6A CN201710082014A CN108427497A CN 108427497 A CN108427497 A CN 108427497A CN 201710082014 A CN201710082014 A CN 201710082014A CN 108427497 A CN108427497 A CN 108427497A
- Authority
- CN
- China
- Prior art keywords
- processing system
- information processing
- components
- power
- load factor
- 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.)
- Pending
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
- G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3024—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a central processing unit [CPU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3037—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a memory, e.g. virtual memory, cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3051—Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
- G06F11/3062—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations where the monitored property is the power consumption
Abstract
Disclosed is a method of calculating a power loss exponent during operation of an information handling system, comprising the steps of: obtaining a utilization value for one or more components in an information handling system; multiplying the utilization values for the one or more components by the load factors of the one or more components, respectively, to calculate an individual power loss for each of the one or more components; and summing the individual power losses of the one or more components to calculate a power loss exponent for the information handling system. An apparatus for calculating a power loss exponent during operation of an information handling system is also disclosed.
Description
Technical field
The present invention relates to information processing systems, and more particularly to one or more of information processing system hardware group
The power consumption of part.
Background technology
Data center is a part for the computer network that key effect is played in tissue operation all over the world.Existing
In codes or data Center, it is necessary to reliable infrastructure is provided for IT operations, to minimize the possibility of any destruction.
Therefore, data center is fed to be supplied to the great power of its cabinet and system, to ensure that all hardware component is in always
Under operating condition, to provide 24/7 service.However, many power supplies may dissipate as power attenuation in use, and
Any calculating, processing or storage effect are not generated.Such power attenuation includes but not limited to VR (voltage regulator) damages
Consumption, power supply AC to DC losses and fan (FAN) power dissipation.
It is thus known that needing to calculate index associated with cabinet and/or system power dissipation so that from data center
From the point of view of viewpoint, entire workload can be suitably assigned in the case where minimizing total-power loss each cabinet and/
Or system.However, the method/system for being traditionally used for calculating power attenuation index needs telemetering of power, it means that must be
Specialized hardware sensing circuit is realized on system circuit board, for detecting the power attenuation of each component.Such realization method
Lead to the undesirable high hardware cost of data center.
Invention content
According to foregoing background, the purpose of the present invention is to provide the alternative sides for eliminating or at least mitigating above-mentioned technical problem
Method/device.
Above-mentioned purpose is met by the combination of the feature of main claim；From claims disclosing further sheet
The Advantageous embodiments of invention.
Those skilled in the art will be from being described below middle export other objects of the present invention.Therefore, the statement of foregoing purpose
It is not exhaustive, and is only for illustrating some purposes in many purposes of the present invention.
Therefore, in an aspect, the present invention be it is a kind of during the operation of information processing system calculate power attenuation refer to
Several methods.This approach includes the following steps：Obtain the utilization value for one or more of information processing system component；
To be multiplied respectively with the load factor of one or more components for the utilization value of one or more components, with calculate one or
The individual power attenuation of each in multiple components；And sum to the individual power attenuation of one or more components, in terms of
Calculate the power attenuation index of information processing system.
The real time monitoring utilized to one or more components is preferably based on to determine for one or more components
Utilization value.
It is highly preferred that being multiplied by the efficiency curve of component by the power data of component to obtain in one or more components
The utilization value of each.
In one embodiment, power data is based on calculating utilization rate (CUPS) the per second reading for component.
In another embodiment, CUPS is received from the CUPS sensors being integrated in the mainboard of information processing system to read
Number.
According to a modification of this invention, power data of the leading-out needle to the power supply unit of information processing system is read from software.
Another deformation according to the present invention is based on blower module for the power data of the blower module of information processing system
Rotary speed.
According to a modification of this invention, load factor is advance for information processing system before the operation of information processing system
Determining statistics load factor.
Optionally, load factor be the dynamic load that calculates automatically of the operation mean value utilized based on the system observed because
Son.
Preferably, load factor is adapted to be dynamically can adjust during the operation of information processing system.
According to another aspect of the present invention, it discloses a kind of for calculating power damage during the operation of information processing system
Consume the device of index.The device includes for obtaining for one or more of the information processing system utilization value of component
Component；For the utilization value for being directed to one or more components to be multiplied with the load factor of one or more components respectively, with
Calculate the component of the individual power attenuation of each in one or more components；And for one or more components
Body power attenuation is summed, to calculate the component of the power attenuation index of information processing system.
The real time monitoring utilized to one or more components is preferably based on to determine for one or more components
Utilization value.
It is highly preferred that being multiplied by the efficiency curve of component by the power data of component to obtain in one or more components
The utilization value of each.
In one implementation, power data is based on calculating utilization rate (CUPS) the per second reading for component.
In another implementation, CUPS is received from the CUPS sensors being integrated in the mainboard of information processing system to read
Number.
According to a modification of this invention, power data of the leading-out needle to the power supply unit of information processing system is read from software.
Another deformation according to the present invention is based on blower module for the power data of the blower module of information processing system
Rotary speed.
According to a modification of this invention, load factor is advance for information processing system before the operation of information processing system
Determining statistics load factor.
Optionally, load factor be the dynamic load that calculates automatically of the operation mean value utilized based on the system observed because
Son.
Preferably, load factor is adapted to be dynamically can adjust during the operation of information processing system.
The present invention has many advantages.By using the CUPS sensors being integrated into the chipset of information processing system,
The individual hardware circuit of the similar telemetering of power of the realtime power data of individual components need not be used to obtain.On the contrary, can be with
Power data is only obtained by software reading, this not only saves the manufacturing cost of information processing system, and reduces it
Power consumption and complexity.
Description of the drawings
It is of the invention aforementioned and into one from the description of the preferred embodiment provided below by way of only related with attached drawing example
Step feature will be apparent, wherein：
Fig. 1 is the block diagram for showing the power flow in data center according to an embodiment of the invention.
Fig. 2 is the stream for showing the method according to an embodiment of the invention for calculating total-power loss index in the data center
Cheng Tu.
Fig. 3 shows user circle according to an embodiment of the invention for adjusting the software of the load factor of component by user
Face.
In the accompanying drawings, run through several embodiments described herein, the identical component of identical digital representation.
Specific implementation mode
In the foregoing description of claim and the present invention, unless context is separately wanted due to representation language or necessary meaning
It asks, the deformation of word " comprising " or such as "comprising" or " containing " are used with the meaning of inclusive, i.e., the feature of specified explanation
In the presence of, but it is not excluded for presence or addition further feature in various embodiments of the present invention.
As herein and claims used in, " coupling " or " connection " refer to directly or indirectly via one or
Multiple electric components being electrically coupled or connecting, unless otherwise indicated.
Referring now to Figure 1, the first embodiment of the present invention is such as information of data center, desktop computer or server
Processing system.Information processing system includes power supply unit (PSU) 20, is configured as what reception was such as provided by main supply of electric power
Then such AC power is converted to DC power 40, DC power 40 by external AC power 26 for information processing system
It is provided to various hardware components as described in greater detail below.AC power 26 can be 110V or 220V, and DC power
40 be 12V.12V voltages can directly drive blower module 38.However, exist have the voltage of the operation less than 12V other are hard
Part component, such as memory 32, CPU 34 and input-output apparatus (I/O) 36.For this purpose, in PSU 20 and various hardware components
Between configuration corresponding voltage adjuster (VR).Voltage regulator includes the memory VR 22 for being connected to memory 32, is connected to
The CPU VR24 of the CPU 34 and I/O VR 26,28 for being connected to I/O 36.Voltage regulator is converted to for that will input D/C voltage
Different output dc voltages.In Fig. 1, some I/O equipment among I/O VR 26 are I/O 36 export the voltage of 3.3V, and I/O
VR 28 exports the voltage of 5V for some other I/O equipment among I/O 36.Finally, some I/O equipment in I/O 36 can be with
It directly works under 12V D/C voltages, and therefore in centre without being directly connected to PSU in the case of any voltage regulator
20。
The ratio for the power attenuation for being attributed to various hardware components is also shown in FIG. 1.Memory VR 22, CPU VR
Each in 24 and I/O VR 26,28 leads to certain power attenuations be directed toward by arrow 42, by being indicated with pattern block.Band
The area of pattern block shows the proportionate relationship of power attenuation in Fig. 1.Blower module 38 also has be directed toward by arrow 44, band figure
The power attenuation of case block instruction, but it can be seen that the area of block 44 is much larger than the area of block 42.This means that in information processing
In the total-power loss being subjected to during the operation of system, the power attenuation highest caused by blower module 38.
Referring now to Figure 2, the method as described above for calculating the total-power loss in information processing system includes base
The individual power attenuation of each hardware component is calculated in calculating utilization rate (CUPS) per second.CUPS is hard-wired reading,
The real-time monitoring utilized to CPU, memory and the I/O in information processing system is provided.CUPS unpacks i.e. independently of operating system
With, and do not consume cpu resource during monitoring process.For example, being provided in the board chip set of Intel's offerManagement engine (ME) technology is disclosed with lower sensor：
Platform CUPS sensors-provided in the form of platform " CUPS indexes " calculating,
Memory and I/O resource utilization values.
Kernel CUPS sensors-offer calculates utilization value.
Memory CUPS sensors-offer memory utilization rate value.
I/O CUPS sensors-offer I/O resource utilization values.
Therefore, it is based on above-mentioned CUPS to read, each individual hardware component (Hardware Subdivision such as shown in FIG. 1 will be calculated
Part) power attenuation.
Following formula is respectively used to calculate the utilization value of PSU, CPU VR, memory VR, I/O VR and blower module：
PSU_loss=PWR_psu_read × Eff_psu_curve [1]
VRcpu_loss=Wc × CPUconfig × Eff_cpu_curve [2]
VRmem_loss=Wm × MEMconfig × Eff_mem_curve [3]
VRio_loss=Wi × IOconfig × Eff_io_curve [4]
FAN_loss=FAN_rpm × Fan_curve [5]
In above-mentioned formula, Wc, Wm and Wi are to be directed to CPU VR, memory VR and I/O VR based on what CUPS was read respectively
Power consumption projection components.Then Wc, Wm and Wi are multiplied with corresponding CPUconfig, MEMconfig and IOconfig,
CPUconfig, MEMconfig and IOconfig are for being installed in information processing system and be built in systems
The configuration parameter of certain types of CPU, memory and I/O equipment.It is read note that only obtaining CUPS for CPU, memory and I/O
Number.For blower module, the sensor by being coupled to the motor in blower module obtains revolutions per minute (rpm) and reads.
For PSU, power reading is directly carried out by software.In formula [2]-[5], the power number that had previously been read by telemetering of power circuit
According to can be obtained now by using software, power data includes the utilization value based on various CUPS sensors.Work as acquisition
When power data, step 60 shown in Fig. 2 is completed.
Next, the power data of all components in formula [1]-[5] efficiency curve corresponding with them is multiplied, with
Obtain the utilization value in step 62.Such curve is intrinsic for each in various assemblies.Therefore, formula [1]-
[5] result is the utilization value of PSU, CPU, memory, I/O and blower module respectively.
Then this method goes to step 64, wherein calculating the power attenuation for individual components.In the step 66, then will
Individual power attenuation summation, to obtain the total-power loss of information processing system.Formula [6] is shown in step 64 and 66 the two
The calculating being related to.
Power attenuation index=PSU_loss × Lp+VRcpu_loss × Lc/ (Lc+Lm+Li)+VRmem_loss × Lm/
(Lc+Lm+Li)+VRio_loss×Li/(Lc+Lm+Li)+FAN_loss×Lf [6]
Wherein Lp, Lc, Lm, Li, Lf are load factors, and user can adjust the weight of each factor.
Here, the calculating for the power attenuation of CPU, memory and I/O respectively depends on three individual load factors
Lc, Lm and Li, characterized systematically is more CPU, I/O or memory intensive (or as two or the group of all three by this
It closes).In other words, Lc, Lm and Li depend on the property of workload performed in system.These are negative in order to better understand
Carry the factor, it may be considered that the workload run in data center's (example as information processing system) shows branch
Hold the behavior trend of specific resources.For example, user logs in and web services initial phase will generate more I/O streams on platform
Amount, and transaction operation is more to calculate (and memory) intensity.Then this can be based on using such factor
A little behavior trend read derived utilization value to adjust from CUPS, are dynamically calculated in the calculating of power attenuation and automatically
Use such factor.
Therefore, power attenuation is the available measurement for calculating surplus in information processing system.Therefore, if system has
Then there is the limited surplus for placing extra work amount on this system in big CUPS indexes.As resource consumption is reduced, system
Power attenuation reduce.There is a large amount of calculating surplus by low-power consumption in instruction, and server is for receiving new work
Load or the main target that workload is moved out, and information processing system is placed in lower power state to reduce
Power consumption.Then it can be monitored using such workload in entire information processing system, to provide the work of system
The advanced and whole understanding of load, can be used in combination with other PTAS power and thermal surveillance feature.
Select the process of appropriate load factor can be by the deployment website in information processing system for given system type
The experiment at place is completed.There are two kinds of load factors.First type is the static load factor, by standing in deployment
Experiment at point is established, and is arranged using spsFITC or IPMI interfaces.Second of type is the dynamic load factor, is based on
The system observed utilizes the operation mean value in (in 3 domains of CPU, memory and I/O) to be automatically calculated.
Once have determined that each (CPU, I/O and memory) in three load factors, then they can by with
In the individual power attenuation of foundation in spsFITC.How formula [6] can obtain for CPU, memory and I/ if being shown separately
The power attenuation of each in O.It is also important to note that these load factors can be at runtime by CUPS IPMI
Order dynamic is adjustable.
By the way that the corresponding load factor of the utilization value of each component is multiplied in formula [6], can calculate a
Body power attenuation, and therefore may then pass through and individual power attenuation is summed to obtain the total-power loss of system.It is counting
After can be regarded as the total-power loss for the exemplary data center of information processing system, for the viewpoint of management, entire work is negative
Load can be suitably assigned to each cabinet and/or system in the case where minimizing total-power loss.
For PSU and blower module, there is also the respective load factors associated with the two components.Fig. 3 shows fair
Family allowable adjusts graphical user circle of five load factors Lp, Lc, Lm, Li and Lf in formula [6] according to user preference dynamic
Face (GUI).In one example, each load factor is adjusted from rank 1 to rank 10.Obtained individual power attenuation
High value indicates that associated power attenuation (such as fan power loss) will be highlighted.
Therefore, exemplary embodiment of the present invention is fully described.Although description is related to specific embodiment, ability
Field technique personnel will be clear that, can put into practice the present invention using the deformation of these details.Therefore, the present invention should not be construed
To be limited to embodiments described herein.
Although being illustrated in detail in and describing the present invention in attached drawing and foregoing description, according to expectation of the present invention
To be considered illustrative rather than restrictive, it should be understood that exemplary embodiment only has been shown and describes, and
It does not limit the scope of the invention in any way.It is appreciated that any feature described herein can be together with any embodiment
It uses.Illustrative embodiment does not repel the other embodiment that do not record each other or herein.Therefore, the present invention also provides including above
The embodiment of the combination of one or more of the illustrative embodiment of description.In the feelings for not departing from the spirit and scope of the present invention
Under condition, can the invention described herein be modified and be changed, and therefore only as indicated by appended claims
Limitation should apply.
Any prior art publication is quoted in text if should be appreciated that, in Australia or any other country,
It is such to quote recognizing for the part for not constituting the common knowledge that the publication forms this field.
Claims (20)
1. a kind of method calculating power attenuation index during the operation of information processing system, includes the following steps：
Obtain the utilization value for one or more of described information processing system component；
It will be for the utilization value of the one or more of components load factor with one or more of components respectively
It is multiplied, to calculate the individual power attenuation of each of one or more of components part；And
The individual power attenuation summation to one or more of components, to calculate the work(of described information processing system
Rate loss index.
2. according to the method described in claim 1, wherein being come based on the real time monitoring utilized to one or more of components
Determine the utilization value for one or more of components.
3. according to the method described in claim 2, being wherein multiplied by the efficiency curve of the component by the power data of the component
To obtain the utilization value for each of one or more of components part.
4. according to the method described in claim 3, the wherein described power data is based on the calculating per second use for the component
Rate (CUPS) is read.
5. according to the method described in claim 4, wherein being passed from the CUPS in the mainboard for being integrated in described information processing system
Sensor receives the CUPS readings.
6. according to the method described in claim 3, wherein reading power supply list of the leading-out needle to described information processing system from software
The power data of member.
7. according to the method described in claim 3, being wherein directed to the power number of the blower module of described information processing system
According to the rotary speed based on the blower module.
8. method according to any one of the preceding claims, wherein the load factor is in described information processing system
The operation before be directed to the predetermined statistics load factor of described information processing system.
9. according to the described method of any one of claim 1-7, wherein the load factor is based on the system profit observed
Operation mean value and the dynamic load factor calculated automatically.
10. according to the described method of any one of claim 1-7, wherein the load factor is adapted to be at described information
It dynamically can adjust during the operation of reason system.
11. a kind of device calculating power attenuation index during the operation of information processing system, described device includes processor,
The processor is configured as：
Obtain the utilization value for one or more of described information processing system component；
It will be for the utilization value of the one or more of components load factor with one or more of components respectively
It is multiplied, to calculate the individual power attenuation of each of one or more of components part；And
The individual power attenuation summation to one or more of components, to calculate the work(of described information processing system
Rate loss index.
12. according to the devices described in claim 11, wherein based on the real time monitoring utilized to one or more of components
To determine the utilization value for one or more of components.
13. device according to claim 12, wherein being multiplied by effective song of the component by the power data of the component
Line obtains the utilization value for each of one or more of components part.
14. device according to claim 13, wherein the power data is made based on the calculating per second for the component
It is read with rate (CUPS).
15. device according to claim 14, wherein from the CUPS in the mainboard for being integrated in described information processing system
Sensor receives the CUPS readings.
16. device according to claim 13, wherein reading power supply of the leading-out needle to described information processing system from software
The power data of unit.
17. device according to claim 13, wherein the power of the blower module for described information processing system
Rotary speed of the data based on the blower module.
18. device according to any one of the preceding claims, wherein the load factor is in described information processing system
The predetermined statistics load factor of described information processing system is directed to before the operation of system.
19. according to the device described in any one of claim 11-17, wherein the load factor is based on the system observed
The operation mean value utilized and the dynamic load factor calculated automatically.
20. according to the device described in any one of claim 11-17, wherein the load factor is adapted to be in described information
It dynamically can adjust during the operation of processing system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710082014.6A CN108427497A (en) | 2017-02-15 | 2017-02-15 | Method and system for calculating power loss exponent |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710082014.6A CN108427497A (en) | 2017-02-15 | 2017-02-15 | Method and system for calculating power loss exponent |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108427497A true CN108427497A (en) | 2018-08-21 |
Family
ID=63155497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710082014.6A Pending CN108427497A (en) | 2017-02-15 | 2017-02-15 | Method and system for calculating power loss exponent |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108427497A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104065568A (en) * | 2014-07-07 | 2014-09-24 | 电子科技大学 | Web server cluster routing method |
| CN104778550A (en) * | 2015-04-15 | 2015-07-15 | 国网湖南省电力公司 | Power network quality analysis method based on real-time operating data |
| EP2947543A1 (en) * | 2014-05-20 | 2015-11-25 | Samsung Electronics Co., Ltd | Method for displaying battery information of electronic device and electronic device performing the same |
| CN106356859A (en) * | 2016-09-29 | 2017-01-25 | 大连海事大学 | Matlab-based rectangular coordinate newton power flow calculation method |
-
2017
- 2017-02-15 CN CN201710082014.6A patent/CN108427497A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2947543A1 (en) * | 2014-05-20 | 2015-11-25 | Samsung Electronics Co., Ltd | Method for displaying battery information of electronic device and electronic device performing the same |
| CN104065568A (en) * | 2014-07-07 | 2014-09-24 | 电子科技大学 | Web server cluster routing method |
| CN104778550A (en) * | 2015-04-15 | 2015-07-15 | 国网湖南省电力公司 | Power network quality analysis method based on real-time operating data |
| CN106356859A (en) * | 2016-09-29 | 2017-01-25 | 大连海事大学 | Matlab-based rectangular coordinate newton power flow calculation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ge et al. | Powerpack: Energy profiling and analysis of high-performance systems and applications | |
| Dargie | A stochastic model for estimating the power consumption of a processor | |
| US9304886B2 (en) | Associating energy consumption with a virtual machine | |
| US20040128100A1 (en) | Method and apparatus for thermal relief for critical tasks in multiple resources environment | |
| WO2002079959A2 (en) | Method and apparatus for determining throttle settings | |
| WO2012033909A2 (en) | Method and system of simulating a data center | |
| Mercati et al. | A linux-governor based dynamic reliability manager for android mobile devices | |
| TWI735531B (en) | Processor power monitoring and control with dynamic load balancing | |
| CN102063354A (en) | Pressure test method of server | |
| US20100106436A1 (en) | Power consumption calculation facility | |
| CN109684167A (en) | Calculation method, device, storage medium and the terminal of application program CPU usage | |
| CN110046118A (en) | A kind of compartment system of PCIE device, method and server system | |
| US20120151242A1 (en) | Apparatus for optimizing supply power of a computer component and methods therefor | |
| Goel et al. | Techniques to measure, model, and manage power | |
| Hsu et al. | Power measurement for high performance computing: State of the art | |
| Akgün et al. | Resource efficient dynamic voltage and frequency scaling on Xilinx FPGAs | |
| CN113900913B (en) | Method and device for determining power consumption, computer equipment and computer readable storage medium | |
| WO2022127574A1 (en) | Method and apparatus for simulating past and future data, and terminal device | |
| CN108427497A (en) | Method and system for calculating power loss exponent | |
| CN109241511B (en) | Electronic report generation method and equipment | |
| Piga et al. | Empirical and analytical approaches for web server power modeling | |
| CN111159038A (en) | Method for simulating CPU load and electronic equipment | |
| CN107239341B (en) | Resource conversion method and system and virtual machine resource scheduling system | |
| Ge et al. | ETune: A power analysis framework for data-intensive computing | |
| CN107451022A (en) | A kind of method and system for automatically adjusting linpack performance tests |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180821 |