CN105893148A - Low-energy-consumption aporadic task scheduling method based on RM strategy - Google Patents
Low-energy-consumption aporadic task scheduling method based on RM strategy Download PDFInfo
- Publication number
- CN105893148A CN105893148A CN201610190341.9A CN201610190341A CN105893148A CN 105893148 A CN105893148 A CN 105893148A CN 201610190341 A CN201610190341 A CN 201610190341A CN 105893148 A CN105893148 A CN 105893148A
- Authority
- CN
- China
- Prior art keywords
- task
- speed
- processor
- resource requirement
- time
- 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
- 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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- 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
-
- 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 low-energy-consumption aporadic task scheduling method based on an RM strategy. The method comprises the steps that the initial priority and the execution priority of tasks are allocated before task set scheduling; the largest blocking time of the tasks and the processor demands under the worst condition are calculated; task sets are partitioned into the task set with the resource demands and the task set without the resource demands, the lowest running speed of the task set with the resource demands is calculated, and the lowest running speed of the task set without the resource demands is calculated; the running speed of the task sets in the off-line stage is calculated; an idle time management queue is built, the idle time of the tasks is calculated, the running speed of a processor is regulated through a DVS technique, and it is guaranteed that the final running speed is not lower than the key speed. According to the method, the condition that a universal power consumption model, the execution time of the tasks and the processor speed have the non-linear relationship and the processor speed switching expense are taken into account, the DVS technique and a DPM technique are combined, and therefore the energy consumption is greatly reduced.
Description
Technical field
The present invention relates to the Real-Time Scheduling of the accidental task in embedded system field, particularly to a kind of idol based on RM strategy
Send out task low energy consumption dispatching method.
Background technology
Embedded device application in life is more and more extensive, common embedded device such as mobile phone, MP3, IPAD, pen
Remembering this computer etc., these equipment are all to use battery to power, and due to limited battery capacity, the energy consumption provided also is limited, right
For these portable embedded devices, energy consumption problem highlights the most very much.Reduce energy consumption can not only the use of extension device time
Between, reduce the caloric value of equipment;And the replacement cycle of battery can be reduced.Therefore, low energy consumption becomes setting of embedded device
Meter target.Processor is the core of embedded device, and its power consumption overturns the dynamic power consumption and leakage current caused essentially from frequency
The quiescent dissipation formed.Dynamic power management (DPM) technology and dynamic voltage regulation (DVS) technology are to reduce system energy consumption at present
Common technology.DPM technology is mainly closed by the equipment that would sit idle for and is reduced system energy consumption.And DVS technology is mainly according to being
The load of system dynamically regulates processor speed and reduces system energy consumption.
Many researchers is by classical Real-Time Scheduling is theoretical and Low-power Technology combines at present, solves the energy consumption of system
Problem.But these researchs are primarily upon periodic task model and hybrid task model.At present accidental task model is ground
Study carefully relatively fewer.Existing accidental task low energy consumption dispatching method, utilizes EDF strategy scheduler task, it is adaptable to dynamic priority
System, ignores quiescent dissipation and the processor speed handover overhead of processor.Additionally, these methods assume the execution time of task
Linear with processor speed.
Summary of the invention
It is an object of the invention to overcome the deficiencies in the prior art, propose a kind of accidental task low energy consumption based on RM strategy
Dispatching method, RM (rate monotonic) strategy is Fixed-priority Scheduling Strategy strategy, and the priority of task by the cycle of task, (release by minimum
Put interval) determine, the cycle (minimum release interval) is the least, and its priority is the highest, and the priority of task of high priority performs.The party
Method is applicable to static priority tasks system, it is contemplated that general power consumption model, the execution time of task become with processor speed
Non-linear relation situation, processor speed handover overhead, in combination with DVS technology and DPM technology.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of accidental task low energy consumption dispatching method based on RM strategy, comprises the steps:
The initial priority of distribution task and execution priority;
Processor demand under the maximum obstruction time of calculating task and worst case;
Task-set being divided into the task-set of resource requirement and does not has the task-set of resource requirement, calculating respectively has resource
Demand task-set and do not have the minimum operation speed of resource requirement task-set;
Calculate the speed of service of task-set off-line phase;
Set up free time management queue;Recovery task completes the free time produced ahead of time, will distribute to free time
The task that current time queue medium priority is the highest, the computation processor current time speed of service;Processor current time is transported
Line speed compares with the critical speed preset, and utilizes the DVS technology regulation processor speed of service to guarantee that it is not less than key
Speed.
Initial priority and the execution priority of described distribution task include:
According to the initial priority of RM strategy distribution task, the minimum release interval of task is the least, and its initial priority is more
High;
Task-set is divided into resource requirement and there is no the task subset of resource requirement;
Calculate the maximum in all initial priority sharing same resource tasks;
If task belongs to not having in resource requirement task subset of task, distribute its execution priority equal to initial priority
Level;If task belongs to having in resource requirement task subset of task, distribute its execution priority all shared same equal to described
Maximum in the initial priority of resource tasks.
In processor demand step under the maximum obstruction time of described calculating task and worst case, task TiMaximum
Blocking time B (Ti) it is expressed as:
Wherein, IPiExpression task TiInitial priority, EPjAnd IPjRepresent task T respectivelyjExecution priority and initial
Priority, ejExpression task TjThe execution time under worst case, r under maximum processor speedjExpression task TjResource need
Asking, k is integer, and its span is 1≤k≤m, and m represents the number of task sharing resource;
Task TiProcessor demand D under interval [0, L] worst case0,LIt is expressed as:
Wherein, wherein eiExpression task TiThe execution time under worst case under maximum processor speed, L is more than 0
Real number, its value isRepresent and use resourceAll task minimums release interval, piExpression task Ti?
Little release interval, riExpression task TiResource requirement.
Described task-set it be divided into the task-set of resource requirement and there is no the task-set of resource requirement, and calculating respectively
There is resource requirement task-set and do not have the minimum operation speed of resource requirement task-set to include:
Task-set is divided into the task-set of resource requirement and there is no the task-set of resource requirement;
There is minimum operation speed LS of resource requirement task-setRTI () is according to the processor need under the worst case of task-set
Ask calculating, represent by equation below:
Wherein, SRT(i, L) indicates that the task-set of resource requirement meets the minimum of its deadline in all tasks of moment L
The speed of service, its value isExpression task TjHolding under worst case
The row time, pjExpression task TjMinimum release interval;
There is no minimum operation speed S of the task-set of resource requirementNRTI () calculates according to the load of task-set, with following public
Formula represents:
Wherein, DTS represents the task-set postponing release, and the interval of each task release time in DTS is both greater than it
Little release interval, FcN () represents the utilization rate upper bound that RM strategy scheduler task collection is feasible.
The speed of service equation below of described calculating task-set off-line phase represents:
Wherein, SlubI () represents the accidental task-set minimum speed of service in interval [0, L], represent by equation below:
Described free time represents by equation below:
Wherein, WiExpression task TiResidue under worst case performs time, UiExpression task TiResidue perform the time, τi
Represent processor speed handover overhead before this moment.
The described computation processor current time speed of service includes:
Computation processor current time speed of service S, wherein
By itself and speed SsfpsasrCompare;If S > Ssfpsasr, S=S is setsfpsasr;The most described S keeps constant.
Described speed SsfpsasrComputational methods are as follows:
If during the processor free time, arranging Ssfpsasr=Smin, described SminFor processor minimum operation speed;
When task TiDischarge an example, and when it belongs to DTS, improve Ssfpsasr, the amount of raising is
When task TiDo not discharge example, and time interval when it is not belonging to DTS and current exceedes between its minimum release
Every time, reduce Ssfpsasr, the amount of reduction is
The initial priority equation below of described task represents:
IPi=n-i+1
Wherein, n represents the number of accidental task in accidental task-set, and i is the integer more than or equal to 1 less than or equal to n.
Maximum equation below in described all initial priority sharing same resource tasks represents:
Wherein, 1≤i≤m, m represent the number of task sharing resource, rjExpression task TjResource requirement.
There is advantages that
(1) the accidental task low energy consumption dispatching method based on RM strategy that the present invention provides, not only allows for general merit
Consumption model, the execution time of task become non-linear relation situation and processor speed handover overhead with processor speed, and
The system of static priority tasks can be applicable to, utilize DVS technology and DPM technology simultaneously, thus than existing accidental task
Low energy consumption dispatching method saves the energy consumption of about 79.37%~82.94%;
(2) it is able to ensure that accidental task completes to perform in its deadline, and is able to ensure that resource is by the use of mutual exclusion;
(3) reduction of system energy consumption, can reduce the production cost of product, the use time of delay apparatus, reduces battery
Replacement cycle.
Below in conjunction with drawings and Examples, the present invention is described in further detail, but the one of the present invention is based on RM strategy
Accidental task low energy consumption dispatching method be not limited to embodiment.
Accompanying drawing explanation
Fig. 1 is the flow chart schematic diagram of the inventive method;
Fig. 2 is the simulation experiment result figure of embodiments of the invention normalization energy consumption and system availability;
Fig. 3 is the simulation experiment result figure of embodiments of the invention normalization energy consumption and task real load.
Detailed description of the invention
Below in conjunction with the accompanying drawings and embodiment the present invention is described in further detail.
See Fig. 1, a kind of based on RM scheduling strategy the resource-constrained accidental task low energy consumption dispatching party that the present invention provides
Method, comprises the steps:
Step 101: the initial priority of distribution task and execution priority;
Concrete, initial priority is allocated according to RM algorithm, and the minimum release interval of task is the least, and its priority is just
The highest;Execution priority obtains distribution when the resources such as CPU start to perform in task, and its priority completes to perform it in task
Before all keep constant;For not having the task of resource requirement, its initial priority is equal to its execution priority.
Further, task TiInitial priority IPiDistributing according to RM algorithm, its value is IPi=n-i+1, i represent and appoint
Business TiSubscript, its value is integer between 1≤i≤n, wherein n number of accidental task in being accidental task-set;Task initial
The numerical value of priority is the biggest, and its priority is the highest.
All use resources RiThe limit priority π of taskiRepresenting, its value is calculated by following formula:
Wherein, 1≤i≤m, m represent the number of task sharing resource, and n represents the number of accidental task in accidental task-set,
rjIt it is task TjResource requirement, IPjRepresent task T respectivelyjInitial priority.
The execution priority distribution method of task is as follows: task-set is divided into two disjoint subsets A and B.Set A
In task there is no a resource requirement, the task in set B has resource requirement.The execution priority of the task in set A is equal to it
Initial priority.The execution priority of the task in set B is equal in all initial priority sharing same resource tasks
Maximum.For task T in set Bi, its execution priority EPi=πi。
Step 102: the processor demand under the maximum obstruction time of calculating task and worst case;
Concrete, task TiMaximum obstruction time B (Ti) it is to be caused by the task that initial priority is lower than it, its
Value isWherein ejIt it is task TjUnder maximum processor speed during the execution under worst case
Between, rjIt it is task TjResource requirement, k is integer, and its span is 1≤k≤m, and m represents the number of task sharing resource,
EPjAnd IPjIt is task T respectivelyjExecution priority and initial priority.Task TiProcess under interval [0, L] worst case
Device demand D0,L, its value isWherein eiExpression task TiWorst case under maximum processor speed
Under the execution time, L is the real number more than 0, and its value isRepresent and use resourceAll tasks minimum
Release interval, piExpression task TiMinimum release interval, riExpression task TiResource requirement.
Step 103: task-set be divided into the task-set of resource requirement and do not have the task-set of resource requirement, counting respectively
Calculate and have resource requirement task-set and there is no the minimum operation speed of resource requirement task-set;
Concrete, task-set is divided into the task-set of resource requirement and there is no the task-set of resource requirement;There is resource
Minimum operation speed LS of demand task-setRTI () calculates according to the processor demand under the worst case of task-set, do not have resource to need
Task minimum operation speed S of the task-set askedNRTI () calculates according to the load of task-set, SNRTI () is that set NRT (i) is in the momentAll tasks meet the minimum speed of service of its deadline, and its value is
Wherein DTS is the task-set postponing release, and the interval of each task its release time in DTS is both greater than between its minimum release
Every, ejIt it is task TjThe execution time under worst case, pjIt it is task TjMinimum release interval, FcN () is that the scheduling of RM strategy is appointed
The utilization rate upper bound that business collection is feasible, its value is
SRT(i, L) is that the task-set having resource requirement is in the momentAll tasks meet its deadline
The minimum speed of service, its value isAll shared resources RiMinimum fortune
Line speed SRTMaximum LS of (i, L)RTI () represents, its value is
Step 104: calculate the speed of service of task-set off-line phase;
Concrete, task TiIn intervalBeing the only one task of having resource requirement, accidental task-set T is in interval
Minimum speed of service S of [0, L]lubI (), its value is Slub(i)=LSRT(i)+SNRT(i,L),If accidental
Business collection more than one task sharing resource, and its resource shared is different, speed of service L of task-set off-line phasesubMust expire
The processor demand of all tasks of foot, its value is calculated by following formula:
Wherein riIt it is task TiResource requirement, DTS be postpone release task-set,It is to use resourceAll
Business minimum release interval, piIt it is task TiMinimum release interval.
Step 105: set up free time management queue;Recovery task completes the free time produced ahead of time, during by the free time
Between distribute to the task that current time queue medium priority is the highest, the computation processor current time speed of service;Processor is worked as
The front moment speed of service compares with the critical speed preset, and utilizes the DVS technology regulation processor speed of service to guarantee that it is not
Less than critical speed.
Concrete, the method setting up free time management queue α is: by setting up a chained list, task is excellent according to it
First level order arrangement, is placed in this chained list, when the free time of task is equal to 0, it is removed from chained list, when task is complete
When becoming to perform, it is inserted in chained list according to its priority orders.
Recovery task completes the free time ST of generation specifically ahead of time: free time
Wherein WiFor task TiResidue under worst case performs time, UiFor task TiResidue perform the time, τiFor at this moment it
Front processor speed handover overhead.Concrete, Wi, UiE all it is arranged to during beginningi, eiWhen being the execution under task worst case
Between, along with the execution of task, Wi, UiAll reducing, if the time of t unit of tasks carrying, both reduces the unit of t accordingly,
By W when task completes to performiIt is set to 0.τiIt is processor speed handover overhead, τi=K | Si-Sj|, K is constant, Si、SjIt it is place
The speed of reason device.
Further, is distributed to the task that now queue medium priority is the highest free time, calculate now processor fortune
Line speed S, its value isBy itself and speed SsfpsasrCompare, SsfpsasrComputational methods as follows:
If during the processor free time, arranging Ssfpsasr=Smin, SminFor processor minimum operation speed;
When task TiDischarge an example, and when it belongs to DTS, improve Ssfpsasr, the amount of raising is
When task TiDo not discharge example, and time interval when it is not belonging to DTS and current exceedes between its minimum release
Every time, reduce Ssfpsasr, the amount of reduction is
When speed S > SsfpsasrTime, S=S is setsfpsasr;Otherwise speed S keeps constant.
Processor speed of service S is compared with the critical speed preset, utilizes DVS technology regulation processor to run speed
Degree S guarantees that it is not less than critical speed.Described critical speed is the speed of service that system energy consumption is minimum, and different processors is crucial
Speed is different.
Further, when processor is in idle condition, it is judged that whether idle interval now is more than processor state
The expense of switching, if more, utilizes DPM technology, switches the processor into low power consumpting state;Otherwise, do not process.Tool
Body, as long as idle interval is more than the expense of processor state switching, will automatically switch.
As in figure 2 it is shown, in the present embodiment, arrange under the execution time (WCET) under the worst case of task and best-case
The ratio of execution time (BCET) be 5, system availability is 0.1 to 0.6, and step-length is 0.05, it is assumed that processor speed switches
The constant of expense is 0.1.Comparing three kinds of methods, first, RM/DPP method in Fig. 2, the method task is all the time with maximum place
Reason device speed performs;Second, SFPSASR method, assumes in the method that task performed with the execution time under its worst case, no
Can reclaim the dynamic idle time, only utilize DVS technical energy saving;3rd, the method for the present invention, the method is in combination with DVS skill
Art and DPM technology consider processor speed handover overhead, it is possible to utilize the dynamic idle time to reduce energy consumption;Exist with RM/DPP method
System availability is to be normalized on the basis of energy consumption when 0.6.From figure 2 it can be seen that methodical energy consumption is all subject to
The impact of system availability, system availability rises, and the methodical normalization energy consumption of institute rises;This is because system availability is more
Height, the execution time increase of each task.The method of the present invention has obvious advantage compared with additive method, with RM/DPP method
Compare the energy consumption saving 73.68%~85.22%, save the energy consumption of 49.08%~64.89% compared with SFPSASR.
As it is shown on figure 3, in the present embodiment, it is 0.31812 that Fig. 3 arranges system availability, investigate task real load to energy
The impact of consumption, the ratio of WCET Yu BCET is from 1 to 10, and step-length is 1, it is assumed that the constant of processor speed handover overhead is 0.1, with
RM/DPP algorithm is normalized on the basis of the energy consumption that ratio is 1 of WCET with BCET.The method compared in Fig. 3 and Fig. 2
Identical.From figure 3, it can be seen that the energy consumption of RM/DPP method is affected by the ratio of WCET with BCET;WCET and BCET
Ratio rise, its normalization energy consumption decline;This is because the ratio of WCET with BCET is the highest, the true of task performs the time more
Little.The change of the ratio of WCET Yu BCET, the impact on SFPSASR method and the method for the present invention is little.This is primarily due to
The speed of service of both approaches task is limited by the speed of task-set off-line phase.The energy consumption of the method for the present invention is below
The energy consumption of additive method, saves the energy consumption of 79.37%~82.94%, saving compared with SFPSASR compared with RM/DPP method
42.14%~51.73% energy consumption about.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all spirit in the present invention and
Within principle, any modification, equivalent substitution and improvement etc. made, should be included within the scope of the present invention.
Claims (10)
1. an accidental task low energy consumption dispatching method based on RM strategy, it is characterised in that including:
The initial priority of distribution task and execution priority;
Processor demand under the maximum obstruction time of calculating task and worst case;
Task-set being divided into the task-set of resource requirement and does not has the task-set of resource requirement, calculating respectively has resource requirement
Task-set and do not have the minimum operation speed of resource requirement task-set;
Calculate the speed of service of task-set off-line phase;
Set up free time management queue;Recovery task completes the free time produced ahead of time, will distribute to currently free time
The task that moment queue medium priority is the highest, the computation processor current time speed of service;Processor current time is run speed
The critical speed spent and preset compares, and utilizes the DVS technology regulation processor speed of service to guarantee that it is not less than critical speed.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 1, it is characterised in that described point
Initial priority and the execution priority of joining task include:
According to the initial priority of RM strategy distribution task, the minimum release interval of task is the least, and its initial priority is the highest;
Task-set is divided into resource requirement and there is no the task subset of resource requirement;
Calculate the maximum in all initial priority sharing same resource tasks;
If task belongs to not having in resource requirement task subset of task, distribute its execution priority equal to initial priority;If
Task belongs to having in resource requirement task subset of task, distributes its execution priority and appoints equal to described all shared same resources
Maximum in the initial priority of business.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 1, it is characterised in that described meter
In processor demand step under the maximum obstruction time of calculation task and worst case, task TiMaximum obstruction time B (Ti) table
It is shown as:
Wherein, IPiExpression task TiInitial priority, EPjAnd IPjRepresent task T respectivelyjExecution priority and initial priority
Level, ejExpression task TjThe execution time under worst case, r under maximum processor speedjExpression task TjResource requirement, k
For integer, its span is 1≤k≤m, and m represents the number of task sharing resource;
Task TiProcessor demand D under interval [0, L] worst case0,LIt is expressed as:
Wherein, wherein eiExpression task TiThe execution time under worst case under maximum processor speed, L is the reality more than 0
Number, its value is Represent and use resourceAll task minimums release interval, piExpression task TiMinimum
Release interval, riExpression task TiResource requirement.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 3, it is characterised in that described general
Task-set is divided into the task-set of resource requirement and does not has the task-set of resource requirement, and calculating has resource requirement task respectively
Collect and do not have the minimum operation speed of resource requirement task-set to include:
Task-set is divided into the task-set of resource requirement and there is no the task-set of resource requirement;
There is minimum operation speed LS of resource requirement task-setRTI () is according to the processor demand meter under the worst case of task-set
Calculate, represent by equation below:
Wherein, SRT(i, L) indicates that the task-set of resource requirement meets the minimum operation of its deadline in all tasks of moment L
Speed, its value isejExpression task TjDuring execution under worst case
Between, pjExpression task TjMinimum release interval;
There is no minimum operation speed S of the task-set of resource requirementNRTI () calculates according to the load of task-set, use equation below table
Show:
Wherein, DTS represents the task-set postponing release, and the interval of each task release time in DTS is both greater than its minimum and releases
Put interval, FcN () represents the utilization rate upper bound that RM strategy scheduler task collection is feasible.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 4, it is characterised in that described meter
The speed of service equation below calculating task-set off-line phase represents:
Wherein, SlubI () represents the accidental task-set minimum speed of service in interval [0, L], represent by equation below:
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 1, it is characterised in that described sky
Represent by equation below between idle:
Wherein, WiExpression task TiResidue under worst case performs time, UiExpression task TiResidue perform the time, τiRepresent
Processor speed handover overhead before this moment.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 6, it is characterised in that described meter
Calculate the processor current time speed of service to include:
Computation processor current time speed of service S, wherein
By itself and speed SsfpsasrCompare;If S > Ssfpsasr, S=S is setsfpsasr;The most described S keeps constant.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 7, it is characterised in that described speed
Degree SsfpsasrComputational methods are as follows:
If during the processor free time, arranging Ssfpsasr=Smin, described SminFor processor minimum operation speed;
When task TiDischarge an example, and when it belongs to DTS, improve Ssfpsasr, the amount of raising is
When task TiDo not discharge example, and when time interval when it is not belonging to DTS and current exceedes its minimum release interval,
Reduce Ssfpsasr, the amount of reduction is
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 2, it is characterised in that described
The initial priority equation below of business represents:
IPi=n-i+1
Wherein, n represents the number of accidental task in accidental task-set, and i is the integer more than or equal to 1 less than or equal to n.
Accidental task low energy consumption dispatching method based on RM strategy the most according to claim 2, it is characterised in that described
Maximum equation below in all initial priority sharing same resource tasks represents:
Wherein, 1≤i≤m, m represent the number of task sharing resource, rjExpression task TjResource requirement.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610190341.9A CN105893148B (en) | 2016-03-30 | 2016-03-30 | A kind of accidental task low energy consumption dispatching method based on RM strategy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610190341.9A CN105893148B (en) | 2016-03-30 | 2016-03-30 | A kind of accidental task low energy consumption dispatching method based on RM strategy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105893148A true CN105893148A (en) | 2016-08-24 |
CN105893148B CN105893148B (en) | 2019-01-22 |
Family
ID=57014326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610190341.9A Active CN105893148B (en) | 2016-03-30 | 2016-03-30 | A kind of accidental task low energy consumption dispatching method based on RM strategy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105893148B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106445070A (en) * | 2016-09-12 | 2017-02-22 | 华侨大学 | Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks |
CN106970835A (en) * | 2017-03-20 | 2017-07-21 | 华侨大学 | Fixed priority resource limited system level energy consumption optimization method |
CN107728466A (en) * | 2017-09-28 | 2018-02-23 | 华侨大学 | One kind is applied to digital control system fixed priority reliability and perceives energy consumption optimization method |
CN107995660A (en) * | 2017-12-18 | 2018-05-04 | 重庆邮电大学 | Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading |
CN108845659A (en) * | 2018-01-30 | 2018-11-20 | 武汉大学 | A kind of embeded processor real-time task distribution method that power consumption is preferential |
CN108845870A (en) * | 2018-05-29 | 2018-11-20 | 大连理工大学 | A kind of probability real-time task scheduling method based on pWCET shaping |
CN109586971A (en) * | 2018-12-14 | 2019-04-05 | 广东外语外贸大学 | A kind of load resource demand appraisal procedure based on linear relationship |
CN109597378A (en) * | 2018-11-02 | 2019-04-09 | 华侨大学 | A kind of resource-constrained hybrid task energy consumption cognitive method |
CN109799805A (en) * | 2019-01-17 | 2019-05-24 | 湖南大学 | A kind of high-performing car electronic schedule algorithm of reliability perception |
CN110275770A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Task balance dispatching method, system, node and electronic equipment |
CN110850954A (en) * | 2019-10-28 | 2020-02-28 | 华侨大学 | Energy consumption optimization method based on fixed priority event triggering mixed key accidental tasks |
CN111597030A (en) * | 2020-05-21 | 2020-08-28 | 华侨大学 | Adaptive factor energy consumption optimization method based on task attributes |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109739332B (en) * | 2019-01-25 | 2022-05-03 | 华侨大学 | Multi-task general energy consumption optimization method |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104424017A (en) * | 2013-08-21 | 2015-03-18 | 中国科学院沈阳计算技术研究所有限公司 | Fault-tolerant low-power-consumption dispatching method applicable to numerical control system |
-
2016
- 2016-03-30 CN CN201610190341.9A patent/CN105893148B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104424017A (en) * | 2013-08-21 | 2015-03-18 | 中国科学院沈阳计算技术研究所有限公司 | Fault-tolerant low-power-consumption dispatching method applicable to numerical control system |
Non-Patent Citations (2)
Title |
---|
张忆文,郭锐锋,邓昌义: "常带宽服务器混合任务低功耗调度算法", 《计算机研究与发展》 * |
邓昌义,郭锐锋,张忆文,韩文业: "面向硬实时系统零星任务低调度算法", 《小型微型计算机系统》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106445070A (en) * | 2016-09-12 | 2017-02-22 | 华侨大学 | Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks |
CN106970835A (en) * | 2017-03-20 | 2017-07-21 | 华侨大学 | Fixed priority resource limited system level energy consumption optimization method |
CN106970835B (en) * | 2017-03-20 | 2021-03-09 | 华侨大学 | Hierarchical energy consumption optimization method for fixed priority resource-limited system |
CN107728466A (en) * | 2017-09-28 | 2018-02-23 | 华侨大学 | One kind is applied to digital control system fixed priority reliability and perceives energy consumption optimization method |
CN107995660A (en) * | 2017-12-18 | 2018-05-04 | 重庆邮电大学 | Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading |
CN108845659A (en) * | 2018-01-30 | 2018-11-20 | 武汉大学 | A kind of embeded processor real-time task distribution method that power consumption is preferential |
CN108845659B (en) * | 2018-01-30 | 2021-06-04 | 武汉大学 | Embedded processor real-time task allocation method with priority on power consumption |
CN110275770A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Task balance dispatching method, system, node and electronic equipment |
CN110275770B (en) * | 2018-03-15 | 2023-09-22 | 阿里巴巴集团控股有限公司 | Task balanced scheduling method, system, node and electronic equipment |
CN108845870A (en) * | 2018-05-29 | 2018-11-20 | 大连理工大学 | A kind of probability real-time task scheduling method based on pWCET shaping |
CN108845870B (en) * | 2018-05-29 | 2021-05-07 | 大连理工大学 | Probabilistic real-time task scheduling method based on pWCET shaping |
CN109597378B (en) * | 2018-11-02 | 2021-03-09 | 华侨大学 | Resource-limited hybrid task energy consumption sensing method |
CN109597378A (en) * | 2018-11-02 | 2019-04-09 | 华侨大学 | A kind of resource-constrained hybrid task energy consumption cognitive method |
CN109586971B (en) * | 2018-12-14 | 2021-06-15 | 广东外语外贸大学 | Load resource demand evaluation method based on linear relation |
CN109586971A (en) * | 2018-12-14 | 2019-04-05 | 广东外语外贸大学 | A kind of load resource demand appraisal procedure based on linear relationship |
CN109799805A (en) * | 2019-01-17 | 2019-05-24 | 湖南大学 | A kind of high-performing car electronic schedule algorithm of reliability perception |
CN110850954A (en) * | 2019-10-28 | 2020-02-28 | 华侨大学 | Energy consumption optimization method based on fixed priority event triggering mixed key accidental tasks |
CN110850954B (en) * | 2019-10-28 | 2023-03-28 | 华侨大学 | Energy consumption optimization method based on fixed priority event triggering mixed key accidental tasks |
CN111597030A (en) * | 2020-05-21 | 2020-08-28 | 华侨大学 | Adaptive factor energy consumption optimization method based on task attributes |
CN111597030B (en) * | 2020-05-21 | 2023-03-24 | 华侨大学 | Adaptive factor energy consumption optimization method based on task attributes |
Also Published As
Publication number | Publication date |
---|---|
CN105893148B (en) | 2019-01-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105893148A (en) | Low-energy-consumption aporadic task scheduling method based on RM strategy | |
Li et al. | Unrelated parallel machine scheduling problem with energy and tardiness cost | |
CN103576827B (en) | The method of the online energy-saving distribution of a kind of cloud computation data center | |
CN106445070B (en) | Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks | |
CN105225017A (en) | A kind of GROUP OF HYDROPOWER STATIONS Short-term Optimal Operation method of multi-Agent | |
CN105975049B (en) | A kind of accidental task low energy consumption dispatching method of tasks synchronization | |
CN107861796B (en) | Virtual machine scheduling method supporting energy consumption optimization of cloud data center | |
CN106201718A (en) | A kind of cloud computing resources dynamic retractility method based on load estimation | |
CN103399626B (en) | Towards Parallel application dispatching system and the method for the power-aware of hybrid compute environment | |
CN103916438B (en) | Cloud testing environment scheduling method and system based on load forecast | |
WO2020239057A1 (en) | Distributed mobile energy storage power distribution network control method and system | |
CN110308977B (en) | Low-energy-consumption method for updating mixed key accidental tasks by dynamic utilization rate | |
CN102981890A (en) | Computing task and virtual machine deploying method within a virtual data center | |
CN110633152A (en) | Method and device for realizing horizontal scaling of service cluster | |
CN103577265A (en) | Method and device of offline energy-saving dispatching in cloud computing data center | |
CN107728466A (en) | One kind is applied to digital control system fixed priority reliability and perceives energy consumption optimization method | |
CN106970835A (en) | Fixed priority resource limited system level energy consumption optimization method | |
CN103023802B (en) | A kind of low energy consumption dispatching patcher towards web cluster and method | |
Song et al. | An efficient scheduling algorithm for energy consumption constrained parallel applications on heterogeneous distributed systems | |
CN105847385A (en) | Cloud computing platform virtual machine dispatching method based on operation duration | |
CN103414784B (en) | Support the cloud computing resource scheduling method of contingency mode | |
CN105225016B (en) | Based on the energy distributing method of cooperative game in the cloud computing system of renewable energy supply | |
CN104063282A (en) | Management method, device and server for IaaS cloud variable scale resource pool | |
CN102567076A (en) | Selection method of server integration scheme based on energy consumption | |
CN105630126A (en) | Low-power scheduling method for mixed task based on constant bandwidth server |
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 |