CN101216727A - Embedded type low-power consumption operating system dynamic frequency regulation mapping method - Google Patents
Embedded type low-power consumption operating system dynamic frequency regulation mapping method Download PDFInfo
- Publication number
- CN101216727A CN101216727A CNA2008100591862A CN200810059186A CN101216727A CN 101216727 A CN101216727 A CN 101216727A CN A2008100591862 A CNA2008100591862 A CN A2008100591862A CN 200810059186 A CN200810059186 A CN 200810059186A CN 101216727 A CN101216727 A CN 101216727A
- Authority
- CN
- China
- Prior art keywords
- processor
- frequency
- operating system
- pmu
- energy
- 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
- 238000013507 mapping Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000004134 energy conservation Methods 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 6
- 230000008034 disappearance Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 14
- 230000006870 function Effects 0.000 description 4
- 239000012141 concentrate Substances 0.000 description 3
- 210000004940 nucleus Anatomy 0.000 description 3
- 210000004027 cell Anatomy 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 229910001416 lithium ion Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Landscapes
- Power Sources (AREA)
Abstract
The invention discloses a mapping method for adjusting the dynamic frequency in an imbedded low-power operating system and provides a new method and technology for modeling, addressing and optimizing the mapping issue during the dynamic adjustment of the processor frequency in an embedded low-power operating system. The invention uses the method that the operating frequency of the processor maps the most energy saving frequency of the theoretic processor upwards and downwards to the actual adjustable voltage in order to serve the energy saving purpose, and meanwhile to ensure the performance requirement for the system. The invention is based on a processor frequency mapping model, through which the most energy saving frequency of the theoretic processor is mapped to the actual adjustable processor frequency. When in operation, the operating system uses dynamic frequency and voltage modulation technology to determine the theoretically most energy saving operating frequency under current performance requirement according to the performance requirement and the operating status of the processor, and maps the most energy saving frequency to the actual processor frequency in order to serve the energy saving purpose.
Description
Technical field
The present invention relates to based on the built-in system software power-saving technology, particularly relate to a kind of mapping method that dynamic frequency is adjusted in the embedded type low-power consumption operating system.
Background technology
In portable embedded device field of power management, present difficulty is to satisfy the requirement of portable terminal to the power supply power supply, accomplishes to take up room little, in light weight again and power-on time is longer.The power solution emphasis of consumer electronics product of future generation should concentrate on hardware and software two aspect technology, comprise: (1) is under small and exquisite physical dimension, how to realize the technology and the technology of required power source performance, relate to technology such as heat management, noise reduction, battery management and function integration; (2) dynamic power management technology, it depends on that cpu performance, software, middleware and user are to changing the requirements such as the time interval of battery; (3) the dynamic power management technology is to operating system nucleus and driver, and application programming interface (API) is to the influence of driver, middleware and application itself.
The function of embedded device becomes from strength to strength now, and function is also more and more abundanter.Along with the embedded device function is more and more, the user is also more and more higher to the energy requirement of embedded device battery, and existing lithium ion battery more and more has been difficult to satisfy the requirement of consumer to normal service time.To this, industry is mainly taked two kinds of methods, and the one, exploitation possesses the more novel battery technology of high-energy-density, and as fuel cell, in predictable 5 years, battery technology can not have very big breakthrough; The 2nd, the energy conversion efficiency of battery and energy-conservation aspect work hard.Under the at present new still jejune situation of high-energy battery technology (as fuel cell), the power management of handheld device of future generation can only be from improving the power utilization rate and reducing these two aspects of power consumption and set about.
How to prolong the serviceable life of battery, and the consumption that as far as possible reduces the energy content of battery has become a research focus of built-in field.Mainly concentrate on hardware design and software optimization two aspects now.Wherein software optimization aspect mainly comprises system software and application software two aspects now.System software mainly concentrates on two of compiler and operating system nucleuss.
In the operating system field, main method for managing power supply is to utilize operating system nucleus now, the frequency of dynamic Adjustment System processor and bus, the whole energy consumption of reduction system.And system can be that system is in low power consumpting state, to reach purpose of energy saving by dynamic frequency instruction change system state.Aspect compiler, now mainly by compiler in the compiling application program stage, code is optimized, compactness and the access means that code tries one's best concentrated as far as possible, to reach purpose of energy saving.
Because processor is generally all only supported discrete frequency adjustment now, be that processor is only supported a series of characteristic frequency, and the optimum processor frequencies that obtains by above method for managing power supply and necessarily be exactly the characteristic frequency that processor is supported can not directly apply on the processor.Need optimum processor frequencies to be mapped on the corresponding processor actual frequency by certain mapping model.This just needs the corresponding mapping method of exploitation so, when guaranteeing that system performance requires, optimum processor frequencies can be mapped to corresponding processor actual frequency.
Summary of the invention
The object of the present invention is to provide the mapping method that dynamic frequency is adjusted in a kind of embedded type low-power consumption operating system.
The step that the present invention solves the technical scheme that its technical matters adopts is as follows:
1) dynamic collect system incident:
Hardware platform all has performance monitoring module PMU, and PMU gathers run case generation number in the total system operational process, frequency when comprising the processor operation, processor memory disappearance;
Operating system is provided with PMU by instruction, and the type of PMU collection event and the higher limit of collection event generation number are set, and when PMU whenever collects the incident that needs are gathered, PMU has a corresponding counter, and this counter will add 1; When the incident generation number that collects reaches the higher limit of setting, PMU will send an interruption, interrupt the program of moving; Interrupt routine reads the variety of event generation number that PMU collects and is added on the incident generation numerical value that operating system preserves;
In the present invention, require PMU to gather two kinds of system events, comprising: the instruction number of processor operation, the idle instruction NOP number of processor operation;
2) the most energy-conservation frequency of computation processor theory:
Operating system is interrupted current operation by timer, and operating system comes regularly to interrupt current operation, the theory the most energy-conservation frequency of timer interrupt handling routine computation processor under current performance requirement by the timing length t that timer is set; The incident generation numerical value that operating system is obtained according to PMU calculates the free time of current processor and compares m; And the incident generation number clear 0 that operating system is preserved; Obtain in the most energy-conservation frequency of current performance requirement processor theory by following mapping formula:
f＝P(m，x)＝m×f
max×x
Wherein:
M is the free time ratio of current processor, the instruction number of m=NOP instruction number/processor operation,
X is a current performance requirement, and span is 0～1,
F is the most energy-conservation frequency of current performance requirement processor theory,
P (m x) is the mapping formula, and this mapping formula will obtain the most energy-conservation frequency according to current x and m parameter,
f
MaxHighest running speed for processor;
3) theoretical running frequency is mapped to the actual processor frequency:
Because processor is generally all only supported discrete frequency adjustment now, be that processor is only supported a series of characteristic frequency, so after operating system calculates the most energy-conservation frequency f of current performance requirement processor theory, processor might not be supported, if the theoretic frequency f that calculates just in time is certain characteristic frequency that processor is supported, operating system directly is set to f by the running frequency of dynamic flat rate adjustment instruction processor so, otherwise will f be mapped to actual frequency guaranteeing not influence under the prerequisite of runnability by following formula:
N/f＝N
H/f
H+(N-N
H)/f
L
Wherein
N finishes the clock number that can move in the time of timer length t under processor frequencies f, i.e. N=f * t,
F is the most energy-conservation frequency of current performance requirement processor theory,
N
HBe that processor operates in frequency f
HUnder periodicity,
f
HBe the characteristic frequency that processor is supported, and f
HGreater than f,
f
LBe the characteristic frequency that processor is supported, and f
LLess than f,
Be that f just in time falls into by f
HAnd f
LIn the interval of forming, and f
HAnd f
LBetween the difference minimum, the characteristic frequency according to processor is supported can find corresponding f by the size of f value
HAnd f
LCalculate N by formula then
HValue, so just can know that task need be at f for f being mapped to actual motion frequency guaranteed performance simultaneously
HThe periodicity N that moves under the frequency
H, also just can calculate simultaneously at f
LThe periodicity N-N that moves under the frequency
H
By above-mentioned formula, just f can be mapped to the actual processor frequency f
HAnd f
L, guaranteed performance is constant simultaneously; Processor just can be adjusted to f with processor frequencies earlier
HMove N under the frequency
HIn the cycle, then processor frequencies is adjusted to f
LMove N-N under the frequency
HCycle.
The present invention compares with background technology, and the useful effect that has is:
The present invention arrives actual adjustable voltage by utilizing the upwards downward method of shining upon of processor running frequency with the most energy-conservation frequency map of theoretical treatment device, to reach purpose of energy saving, guarantees the performance requirement of system simultaneously.This invention is arrived actual adjustable processor frequencies by mapping model with the most energy-conservation frequency map of theoretical treatment device based on a processor frequencies mapping model.Operating system in the process of implementation, the dynamic frequency voltage-regulating technique, determine the energy-saving run frequency of the theory of processor under current performance requirement according to performance requirement and processor ruuning situation, be mapped to the actual processor frequency by mapping model then, to reach energy-conservation purpose.
(1) performance guarantees.When application program is moved, the most energy-conservation frequency map of theoretical treatment device is arrived actual adjustable processor frequencies, the performance that it is exactly the assurance system that this model has a prerequisite by the frequency map model in system.
(2) stability.Operating system is grasped the right of dynamic debugging system state in own hand, rather than transfers to application program, and system just can guarantee the stable of system at the state of taking into account dynamic debugging system under the overall situation like this.
(3) practicality.The present invention can be few as far as possible the code of modification system, reach energy-conservation purpose, mapping model is very simple, is easy to realize.Through repeatedly checking, this method can well reduce programmer's workload, and is very practical.
Description of drawings
Accompanying drawing is the process flow diagram of total system work.
Embodiment
As shown in drawings, during dynamic frequency is adjusted in implementing embedded type low-power consumption operating system mapping method, operating system has been considered performance and energy-conservation two constraints in scheduling process, and the energy-efficient treatment device frequency f of theory is mapped to the actual processor frequency.
1) dynamic collect system incident
The hardware platform of current trend all has performance monitoring module PMU, and PMU can gather run case in the total system operational process, frequency when comprising the processor operation, processor memory disappearance etc.; PMU has plural register and is used for preserving the incident generation quantity that collects; Can gather two incidents simultaneously as the PMU on the Intel Xscale processor.
Operating system can be provided with PMU by special instruction, the type of PMU collection event and the higher limit of collection event generation number are set, when PMU whenever collects the incident that needs are gathered, PMU has a corresponding counter, this counter will add 1 also can subtract 1, and this main with used PMU characteristic is relevant; When the incident generation number that collects reaches the higher limit of setting, PMU will send an interruption, interrupt the program of moving; Interrupt routine reads the variety of event generation number that PMU collects and is added on the incident generation numerical value that operating system preserves;
With the PMU on the Intel Xscale processor is example, it has two performance monitoring counter PMN0 and PMN1, be used for counting at system's class incident, the incident that is monitored is provided with in register PMNC the inside, incident generation number is set in PMNO and PMN1 simultaneously, the every generation of events corresponding once, corresponding counter adds 1, system will produce an interruption when PMN0 and PMN1 overflow, PMNC also is one 32 a register, it can carry out rezero operation to all counters, can also the incident that PMN0 and PMN1 will monitor be provided with.
In the present invention, require PMU to gather two kinds of system events, comprising: processor memory disappearance, the instruction number of processor operation;
2) the most energy-conservation frequency of computation processor theory
Operating system is interrupted current operation by timer, and operating system comes regularly to interrupt current operation, the theory the most energy-conservation frequency of timer interrupt handling routine computation processor under current performance requirement by the timing length t that timer is set; The incident generation number that operating system is obtained according to PMU calculates the free time of current processor and compares m; And the incident generation number clear 0 that operating system is preserved; Obtain in the most energy-conservation frequency of current performance requirement processor theory by following mapping formula:
f＝P(m，x)＝m×f
max×x
Wherein
M is the free time ratio of current processor, the instruction number of m=NOP instruction number/processor operation,
X is a current performance requirement, and span is 0～1,
F is the most energy-conservation frequency of current performance requirement processor theory,
P (m x) is the mapping formula, and this mapping formula will obtain the most energy-conservation frequency according to current x and m parameter,
f
MaxHighest running speed for processor;
(m x) is the mapping formula to P, and this mapping formula will obtain the most energy-conservation frequency according to current x and m parameter, be example with Intel XcalePXA255, the maximum running frequency f of processor
MaxBe 400MHZ, can obtain f=400 * m * x according to m and x so, suppose that present m is 0.4 and x is 1 to be exactly under the highest situation of performance, so f=160MHZ.
3) theoretical running frequency is mapped to the actual processor frequency
Because processor is generally all only supported discrete frequency adjustment now, be that processor is only supported a series of characteristic frequency, so after operating system calculates the most energy-conservation frequency f of current performance requirement processor theory, processor might not be supported, if the theoretic frequency f that calculates just in time is certain characteristic frequency that processor is supported, so operating system can be directly running frequency by dynamic flat rate adjustment instruction processor be set to f, otherwise will f be mapped to actual frequency guaranteeing not influence under the prerequisite of runnability by following formula:
N/f＝N
H/f
H+(N-N
H)/f
L
Wherein
N finishes the clock number that can move in the time of timer length t under processor frequencies f, i.e. N=f * t,
F is the most energy-conservation frequency of current performance requirement processor theory,
N
HBe that processor operates in frequency f
HUnder periodicity,
f
HBe the characteristic frequency that processor is supported, and f
HGreater than f,
f
LBe the characteristic frequency that processor is supported, and f
LLess than f,
Be that f just in time falls into by f
HAnd f
LIn the interval of forming, and f
HAnd f
LBetween the difference minimum, the characteristic frequency according to processor is supported can find corresponding f by the size of f value
HAnd f
LCalculate N by formula then
HValue, so just can know that task need be at f for f being mapped to actual motion frequency guaranteed performance simultaneously
HThe periodicity N that moves under the frequency
H, also just can calculate simultaneously at f
LThe periodicity N-N that moves under the frequency
H
By above-mentioned formula, just f can be mapped to the actual processor frequency f
HAnd f
L, guaranteed performance is constant simultaneously; Processor just can be adjusted to f with processor frequencies earlier
HMove N under the frequency
HIn the cycle, then processor frequencies is adjusted to f
LMove N-N under the frequency
HCycle.With Intel Xscale is example, and Xscale PXA255 chip is supported 4 characteristic frequency, is respectively 400MHZ, 300MHZ, 200MHZ, 100MHZ.Suppose present f=340MHZ, t=10us, N=3400 cycle so.Because 300＜f＜400, and 300 and 400 near f, can obtain f so
H=400, f
L=300.By formula N/f=N
H/ f
H+ (N-N
H)/f
L, can calculate N
H=1600 cycles, and N-N
HIn=1800 cycles, promptly processor moves 1600 cycles under 400MHZ, then 1800 cycles of operation under 300MHZ.
Claims (1)
1. the mapping method that dynamic frequency is adjusted in the embedded type low-power consumption operating system is characterized in that the step of this method is as follows:
1) dynamic collect system incident:
Hardware platform all has performance monitoring module PMU, and PMU gathers run case generation number in the total system operational process, frequency when comprising the processor operation, processor memory disappearance;
Operating system is provided with PMU by instruction, and the type of PMU collection event and the higher limit of collection event generation number are set, and when PMU whenever collects the incident that needs are gathered, PMU has a corresponding counter, and this counter will add 1; When the incident generation number that collects reaches the higher limit of setting, PMU will send an interruption, interrupt the program of moving; Interrupt routine reads the variety of event generation number that PMU collects and is added on the incident generation numerical value that operating system preserves;
In the present invention, require PMU to gather two kinds of system events, comprising: the instruction number of processor operation, the idle instruction NOP number of processor operation;
2) the most energy-conservation frequency of computation processor theory:
Operating system is interrupted current operation by timer, and operating system comes regularly to interrupt current operation, the theory the most energy-conservation frequency of timer interrupt handling routine computation processor under current performance requirement by the timing length t that timer is set; The incident generation numerical value that operating system is obtained according to PMU calculates the free time of current processor and compares m; And the incident generation number clear 0 that operating system is preserved; Obtain in the most energy-conservation frequency of current performance requirement processor theory by following mapping formula:
f＝P(m，x)＝m×f
max×x
Wherein:
M is the free time ratio of current processor, the instruction number of m=NOP instruction number/processor operation,
X is a current performance requirement, and span is 0～1,
F is the most energy-conservation frequency of current performance requirement processor theory,
P (m x) is the mapping formula, and this mapping formula will obtain the most energy-conservation frequency according to current x and m parameter,
f
MaxHighest running speed for processor;
3) theoretical running frequency is mapped to the actual processor frequency:
Because processor is generally all only supported discrete frequency adjustment now, be that processor is only supported a series of characteristic frequency, so after operating system calculates the most energy-conservation frequency f of current performance requirement processor theory, processor might not be supported, if the theoretic frequency f that calculates just in time is certain characteristic frequency that processor is supported, operating system directly is set to f by the running frequency of dynamic flat rate adjustment instruction processor so, otherwise will f be mapped to actual frequency guaranteeing not influence under the prerequisite of runnability by following formula:
N/N
H/f
H+(N-N
H)/f
L
Wherein
N finishes the clock number that can move in the time of timer length t under processor frequencies f, i.e. N=f * t,
F is the most energy-conservation frequency of current performance requirement processor theory,
N
HBe that processor operates in frequency f
HUnder periodicity,
f
HBe the characteristic frequency that processor is supported, and f
HGreater than f,
f
HBe the characteristic frequency that processor is supported, and f
LLess than f,
Be that f just in time falls into by f
HAnd f
LIn the interval of forming, and f
HAnd f
LBetween the difference minimum, the characteristic frequency according to processor is supported can find corresponding f by the size of f value
HAnd f
LCalculate N by formula then
HValue, so just can know that task need be at f for f being mapped to actual motion frequency guaranteed performance simultaneously
HThe periodicity N that moves under the frequency
H, also just can calculate simultaneously at f
LThe periodicity N-N that moves under the frequency
H
By above-mentioned formula, just f can be mapped to the actual processor frequency f
HAnd f
L, guaranteed performance is constant simultaneously; Processor just can be adjusted to f with processor frequencies earlier
HMove N under the frequency
HIn the cycle, then processor frequencies is adjusted to f
LMove N-N under the frequency
HCycle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100591862A CN101216727A (en) | 2008-01-16 | 2008-01-16 | Embedded type low-power consumption operating system dynamic frequency regulation mapping method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008100591862A CN101216727A (en) | 2008-01-16 | 2008-01-16 | Embedded type low-power consumption operating system dynamic frequency regulation mapping method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101216727A true CN101216727A (en) | 2008-07-09 |
Family
ID=39623170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008100591862A Pending CN101216727A (en) | 2008-01-16 | 2008-01-16 | Embedded type low-power consumption operating system dynamic frequency regulation mapping method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101216727A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010091598A1 (en) * | 2009-02-11 | 2010-08-19 | 上海芯豪微电子有限公司 | Method and device for automatic real-time adjustment of clock frequency or supply voltage of processor system |
| CN101604198B (en) * | 2009-07-10 | 2010-09-29 | 杭州电子科技大学 | Method for reducing power consumption of embedded system |
| CN101866213A (en) * | 2010-07-23 | 2010-10-20 | 上海交通大学 | Energy-saving method of embedded system with SPM-DMA (Sequential Processing Machine-Direct Memory Access) structure |
| CN104011621A (en) * | 2011-12-22 | 2014-08-27 | 英特尔公司 | A method, apparatus, and system for energy efficiency and energy conservation including enhanced temperature based voltage control |
| CN104424031A (en) * | 2013-08-23 | 2015-03-18 | 中兴通讯股份有限公司 | Working frequency control method and device for processor |
| CN105677482A (en) * | 2015-12-31 | 2016-06-15 | 联想(北京)有限公司 | Frequency adjusting method and electronic equipment |
| CN108509255A (en) * | 2017-02-24 | 2018-09-07 | 龙芯中科技术有限公司 | The treating method and apparatus of hardware interrupts |
| CN109085911A (en) * | 2018-07-11 | 2018-12-25 | 深圳超多维科技有限公司 | A kind of dynamic frequency method, apparatus, equipment and storage medium |
| CN110045951A (en) * | 2019-04-22 | 2019-07-23 | 国网山西省电力公司电力科学研究院 | A kind of developing instrument that neural network hardware low-power consumption customizes |
-
2008
- 2008-01-16 CN CNA2008100591862A patent/CN101216727A/en active Pending
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010091598A1 (en) * | 2009-02-11 | 2010-08-19 | 上海芯豪微电子有限公司 | Method and device for automatic real-time adjustment of clock frequency or supply voltage of processor system |
| CN101604198B (en) * | 2009-07-10 | 2010-09-29 | 杭州电子科技大学 | Method for reducing power consumption of embedded system |
| CN101866213A (en) * | 2010-07-23 | 2010-10-20 | 上海交通大学 | Energy-saving method of embedded system with SPM-DMA (Sequential Processing Machine-Direct Memory Access) structure |
| CN104011621B (en) * | 2011-12-22 | 2016-12-14 | 英特尔公司 | Based on temperature voltage-controlled for efficiency with energy-conservation methods, devices and systems including strengthen |
| CN104011621A (en) * | 2011-12-22 | 2014-08-27 | 英特尔公司 | A method, apparatus, and system for energy efficiency and energy conservation including enhanced temperature based voltage control |
| CN104424031A (en) * | 2013-08-23 | 2015-03-18 | 中兴通讯股份有限公司 | Working frequency control method and device for processor |
| CN104424031B (en) * | 2013-08-23 | 2021-01-26 | 中兴通讯股份有限公司 | Method and device for controlling working frequency of processor |
| CN105677482A (en) * | 2015-12-31 | 2016-06-15 | 联想(北京)有限公司 | Frequency adjusting method and electronic equipment |
| CN105677482B (en) * | 2015-12-31 | 2020-12-18 | 联想(北京)有限公司 | Frequency adjusting method and electronic equipment |
| CN108509255A (en) * | 2017-02-24 | 2018-09-07 | 龙芯中科技术有限公司 | The treating method and apparatus of hardware interrupts |
| CN108509255B (en) * | 2017-02-24 | 2021-01-22 | 龙芯中科技术股份有限公司 | Method and device for processing hardware interrupt |
| CN109085911A (en) * | 2018-07-11 | 2018-12-25 | 深圳超多维科技有限公司 | A kind of dynamic frequency method, apparatus, equipment and storage medium |
| CN110045951A (en) * | 2019-04-22 | 2019-07-23 | 国网山西省电力公司电力科学研究院 | A kind of developing instrument that neural network hardware low-power consumption customizes |
| CN110045951B (en) * | 2019-04-22 | 2022-04-15 | 国网山西省电力公司电力科学研究院 | Development tool for low-power-consumption customization of neural network hardware |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101216727A (en) | Embedded type low-power consumption operating system dynamic frequency regulation mapping method | |
| Ge et al. | Performance-constrained distributed dvs scheduling for scientific applications on power-aware clusters | |
| CN100465857C (en) | Simplifying method facing to embedded system low-power consumption real time task scheduling | |
| Ge et al. | Improvement of power-performance efficiency for high-end computing | |
| CN101965039B (en) | Ultra-low-power design method for wireless sensor network node | |
| Duan et al. | Optimizing cloud data center energy efficiency via dynamic prediction of CPU idle intervals | |
| CN101510115B (en) | Method for controlling processor frequency when program operation by using dynamic compiler | |
| CN101853066A (en) | Method and device for automatically adjusting clock frequency of system in real time | |
| CN1945498A (en) | Low power consumption real time task parameter model dispatching method facing embedded system | |
| Mikhaylov et al. | Optimization of microcontroller hardware parameters for wireless sensor network node power consumption and lifetime improvement | |
| Xiang et al. | Run-time management for multicore embedded systems with energy harvesting | |
| Menon et al. | Thermal aware automated load balancing for hpc applications | |
| Haj-Yahya et al. | Energy efficient high performance processors: recent approaches for designing green high performance computing | |
| Stangaciu et al. | Energy efficiency in real-time systems: A brief overview | |
| CN100590571C (en) | MSR method for real time embedded system EDF low-power consumption scheduling | |
| CN203276161U (en) | Low-power-consumption embedded intelligent core system | |
| Li et al. | Chameleon: Adapting throughput server to time-varying green power budget using online learning | |
| CN102761423A (en) | Control method of rectifying modules in communication power supply system | |
| CN202476201U (en) | Wind-solar-electricity hybrid energy-saving and water-saving irrigation device | |
| Huang et al. | A energy efficient scheduling base on dynamic voltage and frequency scaling for multi-core embedded real-time system | |
| CN205986319U (en) | Supply circuit is used at little consumption data transmission terminal | |
| Wang et al. | Task migration for energy conservation in real-time multi-processor embedded systems | |
| Xiang et al. | Harvesting-aware energy management for multicore platforms with hybrid energy storage | |
| Ge et al. | Performance and energy modeling for cooperative hybrid computing | |
| CN204118792U (en) | A kind of UEM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080709 |