CN106293009A - A kind of built-in energy-saving dispatching method divided based on region - Google Patents
A kind of built-in energy-saving dispatching method divided based on region Download PDFInfo
- Publication number
- CN106293009A CN106293009A CN201610675752.7A CN201610675752A CN106293009A CN 106293009 A CN106293009 A CN 106293009A CN 201610675752 A CN201610675752 A CN 201610675752A CN 106293009 A CN106293009 A CN 106293009A
- Authority
- CN
- China
- Prior art keywords
- task
- interval
- buttress
- deadline
- threshold value
- 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
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/329—Power saving characterised by the action undertaken by task scheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
- G06F9/4893—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues taking into account power or heat criteria
-
- 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 invention discloses a kind of built-in energy-saving dispatching method divided based on region, comprise the steps: to use task instances revision method that task-set to be scheduled is revised；According to task run cycle and the utilization rate of task-set, execution threshold value is set for each task；Task scheduling is carried out: when the length of the cycle of the task interval overlapping region interval with buttress is more than the execution threshold value of task, then dispatches this task and be allowed to complete in current buttress interval according to performing threshold value；Otherwise, will this task scheduling to next one buttress interval perform；This key method that the present invention provides, free time fragment medium and small for system is merged, to increase the average length of free time fragment, and then obtain sufficiently long free time and change for the resting state completing flush bonding processor, thus effectively reduce the live load of processor to reach energy-conservation purpose.
Description
Technical field
The invention belongs to embedded real time system task scheduling field, more particularly, to a kind of based on region division
Built-in energy-saving dispatching method.
Background technology
Embedded mobile terminal is own through gradually incorporating people's life, and most of embedded devices are generally with the lithium of fixing electricity
Batteries etc. are powered as power supply rather than direct current supply, and its application background also limit the possibility that embedded device charges at any time
Property.Low energy consumption becomes the important performance indexes determining that embedded system is worth already.Complete same task energy consumption less, the most permissible
The single charge greatly extending embedded mobile device uses the time, to improving Consumer's Experience and playing that embedded device is actual to be made
With being worth, all there is extremely important effect.
The application task run in embedded system needs to meet and performs time restriction, it is necessary in the previously given off period
Completing, how the relation between the scheduling of active balance real-time and energy-saving distribution is the research in embedded real-time calculating field before
Emphasis.The administration of energy conservation of existing embedded system has ERHS (Energy-saving Rate Hharmonized
And two kinds of typical technologies of DVFS (Dynamic Voltage and Frequency Scaling) Schedulers).
ERHS grows up on the basis of rate-monotonic, and the priority of task was distributed by deadline,
The task priority height that deadline is short is low by the task priority of time length；Defect is only with cycle of task certainly
Determine priority and be easily caused that some cycle is less but very important task misses the off period and the problem that cannot dispatch.
DVFS is the dynamic energy consumption being reduced processor by the execution frequency of regulation processor, the execution frequency of processor
Carrying change in voltage along with effect sheet on a processor and change, and the speed of processor is the slowest, the dynamic energy consumption of processor is more
Few；Utilize DVFS technology can adjust processor frequencies and carry out energy-conservation, but the drawback reducing processor speed is to cause to appoint in real time
The increase of business execution time, thus cause static energy consumption to increase.
Summary of the invention
For disadvantages described above or the Improvement requirement of prior art, the invention provides a kind of based on region division embedded
Energy-saving scheduling method, its object is to reduce the static energy consumption of embedded real-time processor.
For achieving the above object, according to one aspect of the present invention, it is provided that a kind of embedded joint divided based on region
Energy dispatching method, comprises the steps:
(1) task-set to be scheduled is revised, to shorten the cycle of all tasks and the work of task in task-set
Amount；
(2) according to task run cycle and the utilization rate of task-set, revised each task is arranged execution threshold value；Institute
State perform threshold value less than the 1/3 of the task instances cycle；
(3) task scheduling is carried out according to described execution threshold value: as the cycle of task interval (Crenel-interval with buttress
Interval, CI) the length of overlapping region more than the execution threshold value of task, then dispatch this task and be allowed to interval at current buttress
In complete；Otherwise, will this task scheduling to next one buttress interval perform；
Wherein, CI is the demarcation interval being shaped like buttress formed by schedule sequences is carried out interval division, CI's
Two ends are Crenel-Point (CP)；For single CI, task is scheduled for the two ends of CI and performs, so that the mesozone of CI
Territory has longer free time fragment and changes for processor state.
Preferably, the above-mentioned built-in energy-saving dispatching method divided based on region, its step (3) includes following sub-step:
(3.1) obtain the worst of task and complete moment tj；
(3.2) comparison task estimate deadline and the worst deadline, complete more than the worst when estimating the deadline
Time, then enter step (3.3)；
It is equal to the worst deadline when estimating the deadline, then enters step (3.4)；
It is less than the worst deadline when estimating the deadline, then enters step (3.5)；
(3.3) judge whether task can delay to process according to the time needed for tasks carrying, the most then enter step
(3.4)；If it is not, then enter step (3.6)；
(3.4) task is joined delayed tasks collection, and enter step (3.7)；
(3.5) judge to estimate whether remaining time is more than described execution threshold value, the most then enter step (3.4)；If it is not,
Then enter step (3.6)；
(3.6) increase can perform task summation, and enters step (3.7)；
(3.7) start time point of task is obtained.
In general, by the contemplated above technical scheme of the present invention compared with prior art, it is possible to obtaining following has
Benefit effect:
(1) the built-in energy-saving dispatching method divided based on region that the present invention provides, carries out task by performing threshold value
Scheduling, partial task example is postponed execution, free time fragment medium and small for system is merged, so that it is guaranteed that in each CI at most
Only one of which continuous print free time fragment；By increasing the average length of free time fragment, to obtain the sufficiently long free time
Time has been used for the resting state conversion of flush bonding processor, ensures that embedded system real time task scheduling meets the cycle simultaneously
Property and the constraint of off period, thus effectively reduce the live load of processor to reach energy-conservation purpose；
(2) the built-in energy-saving dispatching method divided based on region that the present invention provides, carries out interval drawing to schedule sequences
Point, and scheduler task collection in the interval after division, the complexity of scheduling can be simplified；And on the other hand, its schedule sequences is drawn
The CI number divided is the fewest, and the length of the free time fragment in whole scheduling process is the longest；When the free time fragment in system
Average length is the longest, can be used in energy-conservation free time fragment the most in system, and its energy-saving effect is the best；
Comparatively speaking, the DVFS method of prior art, although can reduce dynamic energy consumption, but the static merit of processor
Consumption is but increasing；And the present invention by by more free time fragment concentrate, formed big free time section for will at
Reason device is converted to low energy consumption state, and the quiescent dissipation under low energy consumption state also can reduce, and thus plays energy-conservation effect；
The ERHS method of prior art, the probability that the task that minimum period length is the least is scheduled is the least；Along with task
The increase of the utilization rate of collection, the schedulable verification condition of ERHS is less susceptible to meet；And this dispatching method that the present invention provides
Then not having these to limit, all of task-set can be scheduled, applicable more periodic task collection.
Accompanying drawing explanation
The schematic flow sheet of the built-in energy-saving dispatching method divided based on region that Fig. 1 provides for embodiment；
The schematic flow sheet of the task allocation step in Fig. 2 embodiment.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right
The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and
It is not used in the restriction present invention.If additionally, technical characteristic involved in each embodiment of invention described below
The conflict of not constituting each other just can be mutually combined.
The flow process of the built-in energy-saving dispatching method divided based on region provided in embodiment as schematically shown in Figure 1, bag
Include following steps:
(1) task-set to be scheduled is revised, to shorten the cycle of all tasks and the work of task in task-set
Amount；
(2) according to task run cycle and the utilization rate of task-set, revised each task is arranged execution threshold value；Should
Perform threshold value less than the 1/3 of the task instances cycle；
(3) task scheduling is carried out according to described execution threshold value: when the length of the cycle of the task interval overlapping region with CI
More than the execution threshold value of task, then dispatch this task and be allowed to complete in current buttress interval；Otherwise, this task scheduling is arrived down
One buttress interval performs.
By to task-set τ '={ τ '1,...τ'nRevise, it is converted into τ={ τ1,...τn}；By this task
Concentrating longer Task Switching is shorter task (cycle of task and the workload of task are shortened corresponding multiple simultaneously),
Thus ensure that the final utilization rate of whole task-set keeps constant；
In task-set after conversion, the cycle of taskDeadlineThe profit of task
By rate
The cycle of the task in the task-set that conversion obtains all meets condition: T1≤Ti≤2T1(1≤i≤n)；
Wherein, C'iRefer to task-set τ ' the worst execution time；T'iBe task-set τ ' cycle；T1Refer to task instances
τ1Cycle, be the multiple of periodic conversion, i refers to the numbering of task, and n refers to the quantity of the task instances of task-set；
Carry out task scheduling according to performing threshold value, calculate the moment point that all optional tasks in task τ start to perform, its stream
Journey is as in figure 2 it is shown, specific as follows
(1) the worst execution time t estimated of task is calculatedj；
(2) judge whether each task can be at interval CIjMiddle execution, specific as follows:
If (a) task τiEstimate the execution timeLess than its worst execution time tj；Further
Relatively estimate the execution timeWith execution threshold value UTiSize；If it is big to estimate the execution time, then judge task
Example τiFor intending postponing the example performed, otherwise, it is determined that example τiFor forcing task, should be at the interval CI of current buttressjMiddle execution；Note
Record can perform the total time that task needs to perform；
If (b) task τiEstimate the execution timeEqual to its worst execution time tj；Then judging should
Task τiFor performing of task can be postponed；
If (c) task τiEstimate the execution timeMore than its worst execution time tj, the most further
Judge whether this task is can delay task；If this task instances can not postpone, then must be at CIjMiddle execution；Can if task is
Delay task, then by this example delay to the interval CI of next buttressj+1Middle execution；
Obtain moment point δ=t that all optional tasks start to performj-SumOptional；Wherein, SumOptional is
Refer to interval CIjCao needs the summation of the workload of all optional tasks performed.
In embodiment, use shape such as (Ci, Di, Ti) tlv triple carry out quantificational description task；Wherein, CiRefer to task τi?
The worst execution time under big processor frequencies, DiAnd TiRefer respectively to relative off period and the Cycle Length of task；Use
τi,jRepresent task-set τiJth task instances.
For the optional task example in task-setIf metThen willScheduling is current
The interval C of buttressjMiddle execution, and make this optional task exampleAt time point tjComplete before；Otherwise, then optional task is real
ExampleIt is dispatched to the interval C of next buttressj+1Section start perform；In embodiment, the Cycle Length in each buttress interval is 20ms.
Task-set for the following periodic duty in embodiment: τ1=(1,10,10), τ2(unit is in=(6,16,16)
ms)；For each task-set τiArrange one and perform thresholds U*Ti；U is the utilization rate of task-set；According to performing, threshold decision is optional
Whether task instances postpones to perform；
Specifically, utilization rate U=1/10+6/16=0.47 of the whole task-set of acquisition is first calculated；
Then judge whether task instances is postponed, due to task instances τ2,2Discharge in moment point 16, and its off period
At t1After moment point, therefore, it is determined that task instances τ2,2At CI1In be optional example；The execution time L estimated due to it1
(τ2,2)=4 ＜ 0.475*16=7.52, by task instances τ2,2Postpone till the interval CI of next buttress2Middle execution；
Wherein, forcing task instances to refer to must performing in specifying CI of task, optional task example refers to be prolonged
Performing in late next CI of task.
At the interval CI of buttress2In, task instances τ2,2For forcing example, it is necessary to complete in this buttress interval；CI interval for buttress2
For, task instances τ2,3For optional task example, the execution time L estimated of this task2(τ2,3)=8 ＞ 0.47*16=
7.52, task τ2,3Must be at the interval CI of buttress2In complete.
As it will be easily appreciated by one skilled in the art that and the foregoing is only presently preferred embodiments of the present invention, not in order to
Limit the present invention, all any amendment, equivalent and improvement etc. made within the spirit and principles in the present invention, all should comprise
Within protection scope of the present invention.
Claims (2)
1. the built-in energy-saving dispatching method divided based on region, it is characterised in that comprise the steps:
(1) task-set to be scheduled is revised, to shorten the cycle of all tasks and the workload of task in task-set；
(2) according to task run cycle and the utilization rate of task-set, revised each task is arranged execution threshold value；
(3) task scheduling is carried out according to described execution threshold value: when the length of the cycle of the task interval overlapping region interval with buttress
More than the execution threshold value of task, then dispatch this task and be allowed to complete in current buttress interval；Otherwise, this task scheduling is arrived down
One buttress interval performs；
Described buttress interval is the demarcation interval formed by schedule sequences is carried out interval division, for single buttress interval, appoints
Business is scheduled for the interval two ends of buttress and performs, so that pile up neatly interval zone line have the length for processor state conversion
Free time fragment.
2. built-in energy-saving dispatching method as claimed in claim 1, it is characterised in that described step (3) includes following sub-step
Rapid:
(3.1) obtain the worst of task and complete moment tj；
(3.2) comparison task estimate deadline and the worst deadline, when estimating the deadline more than the worst deadline,
Then enter step (3.3)；
It is equal to the worst deadline when estimating the deadline, then enters step (3.4)；
It is less than the worst deadline when estimating the deadline, then enters step (3.5)；
(3.3) judge whether task can delay to process according to the time needed for tasks carrying, the most then enter step (3.4)；If
No, then enter step (3.6)；
(3.4) task is joined delayed tasks collection, and enter step (3.7)；
(3.5) judge to estimate whether remaining time is more than described execution threshold value, the most then enter step (3.4)；If it is not, then enter
Enter step (3.6)；
(3.6) increase can perform task summation, and enters step (3.7)；
(3.7) according to the start time point that can perform task summation acquisition task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610675752.7A CN106293009B (en) | 2016-08-16 | 2016-08-16 | A kind of built-in energy-saving dispatching method based on region division |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610675752.7A CN106293009B (en) | 2016-08-16 | 2016-08-16 | A kind of built-in energy-saving dispatching method based on region division |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106293009A true CN106293009A (en) | 2017-01-04 |
| CN106293009B CN106293009B (en) | 2018-01-26 |
Family
ID=57678307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610675752.7A Expired - Fee Related CN106293009B (en) | 2016-08-16 | 2016-08-16 | A kind of built-in energy-saving dispatching method based on region division |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106293009B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631526A (en) * | 2020-12-31 | 2021-04-09 | 国网浙江省电力有限公司营销服务中心 | Method for reading and writing same memory by main cycle and interrupt cycle of embedded system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1945498A (en) * | 2006-10-12 | 2007-04-11 | 浙江大学 | Low power consumption real time task parameter model dispatching method facing embedded system |
| CN101604198A (en) * | 2009-07-10 | 2009-12-16 | 杭州电子科技大学 | A kind of method that reduces power consumption of embedded system |
| CN103257900A (en) * | 2013-05-24 | 2013-08-21 | 杭州电子科技大学 | Real-time task set resource reservation method on multiprocessor for reducing CPU occupation |
| US20130305087A1 (en) * | 2012-05-09 | 2013-11-14 | Imec | Method and system for real-time error mitigation |
| CN103793207A (en) * | 2014-01-21 | 2014-05-14 | 上海爱数软件有限公司 | Intelligent scheduling method of single-thread and multi-priority system |
-
2016
- 2016-08-16 CN CN201610675752.7A patent/CN106293009B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1945498A (en) * | 2006-10-12 | 2007-04-11 | 浙江大学 | Low power consumption real time task parameter model dispatching method facing embedded system |
| CN101604198A (en) * | 2009-07-10 | 2009-12-16 | 杭州电子科技大学 | A kind of method that reduces power consumption of embedded system |
| US20130305087A1 (en) * | 2012-05-09 | 2013-11-14 | Imec | Method and system for real-time error mitigation |
| CN103257900A (en) * | 2013-05-24 | 2013-08-21 | 杭州电子科技大学 | Real-time task set resource reservation method on multiprocessor for reducing CPU occupation |
| CN103793207A (en) * | 2014-01-21 | 2014-05-14 | 上海爱数软件有限公司 | Intelligent scheduling method of single-thread and multi-priority system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631526A (en) * | 2020-12-31 | 2021-04-09 | 国网浙江省电力有限公司营销服务中心 | Method for reading and writing same memory by main cycle and interrupt cycle of embedded system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106293009B (en) | 2018-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105893148B (en) | A kind of accidental task low energy consumption dispatching method based on RM strategy | |
| CN101290585B (en) | Embedded system real time task scheduling method | |
| CN102929720B (en) | A kind of energy-conservation job scheduling system | |
| CN106970835B (en) | Hierarchical energy consumption optimization method for fixed priority resource-limited system | |
| CN103297505B (en) | The On-Line Control Method of data center's multiple-energy-source and system under dynamic cloud service request | |
| Lu et al. | Scheduling and mapping of periodic tasks on multi-core embedded systems with energy harvesting | |
| CN106445070B (en) | Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks | |
| CN110308977B (en) | Low-energy-consumption method for updating mixed key accidental tasks by dynamic utilization rate | |
| Li et al. | Energy-efficient scheduling in nonpreemptive systems with real-time constraints | |
| CN109324880A (en) | A kind of low-power consumption scheduling method suitable for real-time system periodic task model | |
| WO2013097313A1 (en) | Method and system for analyzing user loads in combination with time information | |
| CN109144680A (en) | A kind of clock ticktack interrupts setting method and device | |
| CN109324891A (en) | A kind of periodic duty low-power consumption scheduling method of ratio free time distribution | |
| CN106293009A (en) | A kind of built-in energy-saving dispatching method divided based on region | |
| Huang et al. | Adaptive power management for real-time event streams | |
| CN107329817A (en) | A kind of stand-by system mixing divides reliability and perceives energy consumption optimization method | |
| Huang et al. | Leakage-aware reallocation for periodic real-time tasks on multicore processors | |
| CN105117281A (en) | Task scheduling method based on task application signal and execution cost value of processor core | |
| CN102546999A (en) | Method, control device and system for reducing device power consumption based on business model | |
| CN106125901A (en) | A kind of dynamic energy-saving real-time scheduling method based on interval division | |
| CN102902878B (en) | A kind of energy cost perception dispatching method | |
| CN108874517A (en) | The stand-by system availability of fixed priority divides energy consumption optimization method | |
| Yokoyama et al. | Static task scheduling algorithms based on greedy heuristics for battery-powered DVS systems | |
| CN111078401B (en) | Periodic task temperature sensing energy consumption optimization method | |
| KR20170065080A (en) | Method for low power scheduling based TL-plane for a real time embedded system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180126 Termination date: 20190816 |