CN103810043B - A kind of energy-saving scheduling method being applicable to digital control system periodic duty - Google Patents

A kind of energy-saving scheduling method being applicable to digital control system periodic duty Download PDF

Info

Publication number
CN103810043B
CN103810043B CN201210445655.0A CN201210445655A CN103810043B CN 103810043 B CN103810043 B CN 103810043B CN 201210445655 A CN201210445655 A CN 201210445655A CN 103810043 B CN103810043 B CN 103810043B
Authority
CN
China
Prior art keywords
task
speed
time
service
execution
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.)
Active
Application number
CN201210445655.0A
Other languages
Chinese (zh)
Other versions
CN103810043A (en
Inventor
张忆文
郭锐锋
王鸿亮
刘娴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Institute of Computing Technology of CAS
Original Assignee
Shenyang Institute of Computing Technology of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenyang Institute of Computing Technology of CAS filed Critical Shenyang Institute of Computing Technology of CAS
Priority to CN201210445655.0A priority Critical patent/CN103810043B/en
Publication of CN103810043A publication Critical patent/CN103810043A/en
Application granted granted Critical
Publication of CN103810043B publication Critical patent/CN103810043B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The present invention discloses a kind of energy-saving scheduling method being applicable to digital control system periodic duty, and its step is as follows: calculate task T in digital control systemiBest execution speed under off-line stateCalculate task T againiIn best execution speedUnder the execution time, according to earliest-deadline-first principle, task-set is ranked up;When certain task completes, reclaim task TiFree time Slack_Time, utilize this free time Slack_Time to calculate this task speed of service S under processor specifies continuous voltage;Final speed of service S determining each task forward part according to speed of service SLSpeed of service S with rear sectionH, thus calculate task speed of service S in forward partLUnder execution time ex.LWith speed of service S in rear sectionHExecution time ex.H, by required actual execution time ex.L、ex.HDispatch execution task;So make the present invention make full use of the free time Slack_Time of system, reduce the speed of service of processor, reach to save the purpose of energy consumption.

Description

A kind of energy-saving scheduling method being applicable to digital control system periodic duty
Technical field
The present invention relates to the Real-Time Scheduling of real-time system field task, a kind of energy-saving scheduling method being applicable to digital control system periodic duty.
Background technology
Real-time system is more and more extensive in the application of fields of numeric control technique and embedded product, seems the most important in the application especially complicated in function, system is huge.This will ask real-time system can provide highly efficient computing capability, and to meet the requirement of radio communication, multimedia application, but high performance cost is exactly high energy consumption.Meanwhile, along with precision, the developing rapidly of footprint day by day of manufacturing process, result in steeply rising of system power dissipation, the high temperature that high power consumption is brought can cause the possibility of system jam to increase, and reduces the reliability of whole system;Accordingly, it is capable to consumption has become the bottleneck of real-time system design.
Dynamic voltage regulation (DVS) technology is a kind of effective low-power consumption and energy optimization technology.Its main thought is for different system loads, on the premise of meeting system real time, reduces operating voltage and the clock frequency of CPU.
Existing low-power consumption scheduling algorithm uses DVS technology, on the premise of meeting system real time, by the free time of recovery system, utilizes free time to reduce working voltage and the clock frequency of processor, to reduce system energy consumption;But it assumes that processor provides continuous print frequency and voltage, and the commercial processor of reality provides only discrete frequency and voltage, in order to ensure task not missed deadlines, its actual motion speed selected is bigger than the ideal velocity calculated in the case of cline frequency, so can cause the waste of system resource.
Summary of the invention
For the weak point of existing low-power consumption scheduling algorithm, the present invention proposes a kind of energy-saving scheduling method being applicable to digital control system periodic duty, and this algorithm can effectively utilize the free time of system, reduces the energy consumption of system.
For realizing above-mentioned algorithm, the present invention uses following technical scheme:
A kind of energy-saving scheduling method being applicable to digital control system periodic duty, it is characterised in that: comprise the steps:
Before task set dispatching, calculate task T in digital control systemiBest execution speed under off-line state
Calculate task T againiIn best execution speedUnder the execution time, according to earliest-deadline-first principle, task-set is ranked up;
When certain task completes, reclaim this task free time Slack_Time, utilize this free time Slack_Time to calculate speed of service S under processor specifies continuous voltage of this task;
According to speed of service S, determine speed of service S of each task forward partLSpeed of service S with rear sectionH, finally calculate task speed of service S in forward partLUnder, the execution time e of taskx.LWith speed of service S in rear sectionHThe execution time e of taskx.H, finally with required actual execution time ex.L、ex.HDispatch execution task.
Earliest-deadline-first principle: deadline is the shortest, priority is the highest, when task TiDeadline identical time, early arriving of task TiPriority is high;When task TiDeadline and arrival time homogeneous while, task TiWhat serial subscript i was little has higher priority.
Calculate each task TiBest execution speed under off-line stateIts calculation procedure is as follows:
Utilize execution time CiWith cycle PiRatio sum formulaCalculate utilization rate U of systemtot, then by utilization rate UtotMinimum speed S with processorminComparing, the greater is the best execution speed under off-line state
When certain task completes, reclaiming this task free time Slack_Time, utilize this free time Slack_Time to calculate speed of service S under processor provides continuous voltage of this task, it is as follows that its speed of service S processes step:
Setting up a data structure α queue and record fulfiling ahead of schedule of task, α queue is in off-line state best execution speedUnder ready queue, the arrival time of logger task, deadline and residue perform the time;
Reclaim this task free time Slack_Time, find out fulfiling ahead of schedule in α queue of task, utilize each residue fulfiling task ahead of schedule to perform the time, calculate each free time Slack_Time fulfiling task ahead of schedule, utilize total free time Slack_Time to calculate speed of service S.
According to speed of service S, determine speed of service S of each task forward partLSpeed of service S with rear sectionH, it is as follows that it processes step:
Value according to speed of service S and two adjacent speed S that processor specifiesLowAnd SHighCompare, when speed of service S is more than SLowAnd S is less than or equal to SHigh(SLow< S≤SHigh) time, determine and think speed of service S of task forward partLIt is speed S that processor specifiesLow, speed of service S of task rear sectionHIt is speed S that processor specifiesHigh
The invention have the advantages that and advantage:
(1) use the inventive method, owing to making full use of the free time Slack_Time of system, reduce the speed of service of processor, thus save the energy consumption of 30.69% than existing low-power consumption scheduling algorithm.
(2) high temperature brought due to high power consumption can cause the possibility of system jam to increase, and can reduce the reliability of whole system, and the present invention makes full use of the free time Slack_Time of system, reduce system energy consumption, therefore improve the reliability of system.
(3) system encapsulation and the cost of cooling are reduced.Power consumption increase can significantly improve the caloric value of system, and in order to make product normally to dispel the heat, producer inherently increases the investment studied for new encapsulation and cooling technology, and the cost that thus be accordingly used in encapsulation and cooling device can improve.
Accompanying drawing explanation
Fig. 1 is that the inventive method processes flow chart of steps;
Fig. 2 is the simulation experiment result figure of the present invention.
Detailed description of the invention
With embodiment, the present invention program is described in further detail below in conjunction with the accompanying drawings.
See accompanying drawing 1, Fig. 2, be a kind of energy-saving scheduling method being applicable to digital control system periodic duty, it is characterised in that: comprise the steps:
Before task set dispatching, calculate task T in digital control systemiBest execution speed under off-line state
Calculate task T againiIn best execution speedUnder the execution time, according to earliest-deadline-first principle, task-set is ranked up;
When certain task completes, reclaim this task free time Slack_Time, utilize this free time to calculate speed of service S under processor provides continuous voltage of this task;
According to speed of service S, determine speed of service S of each task forward partLSpeed of service S with rear sectionH, finally calculate task speed of service S in forward partLUnder execution time ex.LWith speed of service S in rear sectionHUnder execution time ex.H, finally with required actual execution time ex.L、ex.HDispatch execution task.
The inventive method is further illustrated:
Calculate task TiBest execution speed in off-line state task
Utilization rate U of system is calculated by formula (1)tot
U tot = Σ i = 1 n C i P i - - - ( 1 )
In formula, Ci、Pi, i be respectively periodic duty TiExecution time, cycle, the identifier of task under worst case;N is the number of task.
Task T is calculated by formula (2)iBest execution speed in off-line state task
S = max { S min , U tot } - - - ( 2 )
Wherein SminIt it is the minimum speed that specifies of processor.
When there being new task to arrive, according to earliest-deadline-first principle, new task is inserted in the relevant position in α queue.
Earliest-deadline-first principle: deadline is the shortest, priority is the highest, when task TiDeadline identical time, early arriving of task TiPriority is high;When task TiDeadline and arrival time homogeneous while, task TiWhat serial subscript i was little has higher priority.
When certain task completes, reclaiming the Slack_Time time this task free time, utilize this free time Slack_Time to calculate speed of service S under processor provides continuous voltage of this task, it is as follows that its speed of service S processes step:
Setting up a data structure α queue and record fulfiling ahead of schedule of task, α queue is in off-line state best execution speedUnder ready queue, the arrival time of logger task, deadline and residue perform the time;
Reclaim this task free time Slack_Time, find out fulfiling ahead of schedule in α queue of task, utilize each residue fulfiling task ahead of schedule to perform the time, calculate each free time Slack_Time fulfiling task ahead of schedule, utilize total free time Slack_Time to calculate speed of service S.
Each free time Slack_Time fulfiling task ahead of schedule is calculated by formula (3):
Slack _ Time = Σ i | d i ≤ d x rem i ( t ) - w x S x ( t ) - - - ( 3 )
In formula, i is the identifier of task, di、dxTask T of being respectivelyi、TxDeadline;remiT () is task TiResidue at moment t performs the time, and the residue execution time of task gradually decreases along with the execution of task;When task completes, the remaining time of task is 0.For task TxAt moment t with speed SxResidue under the worst case run performs the time. be can be seen that free time Slack_Time is that the summation of the residue execution time by fulfiling task ahead of schedule calculates with the difference of the residue execution time under worst case by formula (3).
Free time Slack_Time is utilized to calculate speed of service S of task:
Speed of service S of task is calculated by formula (4)
S = w x S x ( t ) rem x ( t ) + Slack _ Time - - - ( 4 )
In formula, remxT () is task TxResidue at moment t performs the time, and Slack_Time is the free time of task.When minimum speed S that speed of service S of task specifies than processorminValue hour, minimum speed S that speed of service S specifies equal to processormin
According to speed of service S, determine speed of service S of each task forward partLSpeed of service S with rear sectionH, finally calculate task speed of service S in forward partLUnder execution time ex.LWith speed of service S in rear sectionHExecution time ex.H, it is as follows that it processes step:
Value according to speed of service S and two adjacent speed S that processor specifiesLowAnd SHighCompare, when speed of service S is more than SLowAnd S is less than or equal to SHigh(SLow< S≤SHigh) time, determine and think speed of service S of task forward partLIt is speed S that processor specifiesLow, speed of service S of task rear sectionHIt is speed S that processor specifiesHigh
Speed of service S in forward partLUnder, task TxThe execution time be execution time ex.L;Speed of service S in rear sectionHUnder, task TxThe execution time be execution time ex.H
Task TxThe execution time meet (5) formula:
ex=ex.L+ex.H(5)
Task TxExecution time when just arriving to meets (6) formula:
S·ex=SL·ex.L+SH·ex.H(6)
Execution time e is learnt by formula (5)x, by execution time exSubstitute into formula (6) and calculate speed of service S in forward partLUnder, task TxExecution time ex.L:
e x . L = e x · S H - S S H - S L - - - ( 7 )
Execution time e is learnt by formula (7)x.L, by execution time ex.LSubstitute into formula (5) and calculate the S of the speed of service in rear sectionHUnder, task TxExecution time ex.H, finally with required actual execution time ex.L、ex.HDispatch execution task.

Claims (5)

1. the energy-saving scheduling method being applicable to digital control system periodic duty, it is characterised in that: include as Lower step:
Before task set dispatching, calculate task T in digital control systemiOptimum operation speed under off-line state Degree
Calculate task T againiIn best execution speedUnder the execution time, according to earliest-deadline-first Task-set is ranked up by principle;
When certain task completes, reclaim this task free time Slack_Time, utilize this free time Slack_Time calculates this task speed of service S under processor provides continuous voltage;
According to speed of service S, determine speed of service S of each task forward partLOperation with rear section Speed SH;Calculate task speed of service S in forward partLUnder actual execution time ex.LWith rear Speed of service S of partHExecution time ex.H, finally with required actual execution time ex.L、ex.HCome Scheduling execution task.
2. according to a kind of energy-saving distribution side being applicable to digital control system periodic duty described in claim 1 Method, it is characterised in that: earliest-deadline-first principle: deadline is the shortest, priority is the highest, when Task TiDeadline identical time, early arriving of task TiPriority is high;When task TiDeadline The while of homogeneous with arrival time, task TiWhat serial subscript i was little has higher priority.
3. according to a kind of energy-saving distribution side being applicable to digital control system periodic duty described in claim 1 Method, it is characterised in that: calculate each task TiBest execution speed under off-line stateIt calculates step Rapid as follows:
Utilize execution time CiWith cycle PiRatio sum formulaCalculate the utilization rate of system Utot, then by utilization rate UtotMinimum speed S with processorminComparing, the greater is under off-line state Best execution speed
4. according to a kind of energy-saving distribution side being applicable to digital control system periodic duty described in claim 1 Method, it is characterised in that: when certain task completes, reclaim the Slack_Time time this task free time, profit The speed of the operation under processor provides continuous voltage of this task is calculated with this free time Slack_Time Degree S, it is as follows that its speed of service S processes step:
Setting up a data structure α queue and record fulfiling ahead of schedule of task, α queue is in off-line state Ready queue under best execution speed, when the arrival time of logger task, deadline and residue perform Between;
Reclaim this task free time Slack_Time, find out fulfiling ahead of schedule in α queue of task, utilize Each residue fulfiling task ahead of schedule performs the time, calculates each free time fulfiling task ahead of schedule Slack_Time, utilizes total free time Slack_Time to calculate speed of service S.
5. according to a kind of energy-saving distribution being applicable to digital control system periodic duty described in claim 1 or 4 Method, it is characterised in that: according to speed of service S, determine speed of service S of each task forward partLWith Speed of service S of rear sectionH, it is as follows that it processes step:
Value according to speed of service S and two adjacent speed S that processor specifiesLowAnd SHighCompare, When speed of service S is more than SLowAnd S is less than or equal to SHighTime, determine and think that the operation of task forward part is fast Degree SLIt is speed S that processor specifiesLow, speed of service S of task rear sectionHIt is processor regulation Speed SHigh
CN201210445655.0A 2012-11-09 2012-11-09 A kind of energy-saving scheduling method being applicable to digital control system periodic duty Active CN103810043B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210445655.0A CN103810043B (en) 2012-11-09 2012-11-09 A kind of energy-saving scheduling method being applicable to digital control system periodic duty

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210445655.0A CN103810043B (en) 2012-11-09 2012-11-09 A kind of energy-saving scheduling method being applicable to digital control system periodic duty

Publications (2)

Publication Number Publication Date
CN103810043A CN103810043A (en) 2014-05-21
CN103810043B true CN103810043B (en) 2016-09-07

Family

ID=50706852

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210445655.0A Active CN103810043B (en) 2012-11-09 2012-11-09 A kind of energy-saving scheduling method being applicable to digital control system periodic duty

Country Status (1)

Country Link
CN (1) CN103810043B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105629873B (en) * 2014-11-07 2018-08-24 中国科学院沈阳计算技术研究所有限公司 A kind of hybrid tasks scheduling method suitable for digital control system
CN105677449B (en) * 2014-11-17 2019-02-26 中国科学院沈阳计算技术研究所有限公司 A kind of low-power consumption scheduling method suitable for digital control system
CN106569884B (en) * 2015-10-12 2020-03-10 沈阳高精数控智能技术股份有限公司 Reliable low-power-consumption scheduling method based on dynamic voltage regulation technology
CN106815073B (en) * 2015-11-27 2020-04-28 沈阳高精数控智能技术股份有限公司 Balance factor-based dynamic accidental task low-power-consumption scheduling method
CN106933325B (en) * 2017-02-10 2019-10-11 华侨大学 A kind of fixed priority I/O device energy consumption management method
CN109613971B (en) * 2018-12-04 2022-03-04 华侨大学 Low-energy-consumption method for allocating idle time with lowest utilization rate
CN111324197B (en) * 2020-02-07 2023-03-07 华侨大学 Method for reducing system energy consumption based on three-speed periodic task
CN112541121A (en) * 2020-12-21 2021-03-23 深圳市彬讯科技有限公司 Page data pushing method and device, computer equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7174468B2 (en) * 2002-08-01 2007-02-06 Texas Instruments Incorporated Methodology for coordinating and tuning application power
CN101923489A (en) * 2010-08-06 2010-12-22 西华师范大学 Embedded real-time task scheduling method based on fuzzy importance and idle time
CN102135906A (en) * 2011-03-18 2011-07-27 深圳市民德电子科技有限公司 Power consumption control method and system orienting to embedded real-time operating system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2133793B1 (en) * 2008-06-10 2015-08-12 Barcelona Supercomputing Center-Centro Nacional de Supercomputación A multithreaded processor and a mechanism and a method for executing one hard real-time task in a multithreaded processor

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7174468B2 (en) * 2002-08-01 2007-02-06 Texas Instruments Incorporated Methodology for coordinating and tuning application power
CN101923489A (en) * 2010-08-06 2010-12-22 西华师范大学 Embedded real-time task scheduling method based on fuzzy importance and idle time
CN102135906A (en) * 2011-03-18 2011-07-27 深圳市民德电子科技有限公司 Power consumption control method and system orienting to embedded real-time operating system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
一种基于增强型空闲时间回收算法的实时系统低功耗策略;张哲 等;《电路与系统学报》;20090215;第14卷(第1期);第18-21页 *
丁万夫 等.面向系统负载的分段式实时调度算法及其实现.《小型微型计算机系统》.2009,第30卷(第3期),第569-572页. *

Also Published As

Publication number Publication date
CN103810043A (en) 2014-05-21

Similar Documents

Publication Publication Date Title
CN103810043B (en) A kind of energy-saving scheduling method being applicable to digital control system periodic duty
Li et al. Unrelated parallel machine scheduling problem with energy and tardiness cost
CN104424017B (en) A kind of fault-tolerant low-power consumption scheduling method suitable for digital control system
CN106445070B (en) Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks
CN105893148A (en) Low-energy-consumption aporadic task scheduling method based on RM strategy
CN101290585A (en) Embedded system real time task scheduling method
CN101840353A (en) Dynamic timing method for real-time embedded operating system
CN103076870A (en) Application fusing scheduling and resource dynamic configuring method of energy consumption drive in data center
CN103810026A (en) Mixing scheduling method suitable for real-time system periodic tasks
CN104808770A (en) Data center energy consumption management method and system based on dynamic frequency modulation
CN103902016A (en) Server power consumption management method oriented to scene prediction
CN105630126B (en) One kind is based on normal bandwidth server hybrid task low-power consumption scheduling method
CN103914346A (en) Group-based dual-priority task scheduling and energy saving method for real-time operating system
CN106970835A (en) Fixed priority resource limited system level energy consumption optimization method
CN109739332A (en) A kind of general energy consumption optimization method of multitask
CN100590571C (en) MSR method for real time embedded system EDF low-power consumption scheduling
Quan et al. Guaranteed scheduling for repetitive hard real-time tasks under the maximal temperature constraint
CN104537505B (en) A kind of data center's real-time task scheduling method based on market guidance
Xiong et al. Energy consumption optimization of processor scheduling for real-time embedded systems under the constraints of sequential relationship and reliability
CN110008515B (en) Renewable energy data center management method and device
CN101697140A (en) Optimized dispatching method of operating system
CN105117281A (en) Task scheduling method based on task application signal and execution cost value of processor core
CN105677449B (en) A kind of low-power consumption scheduling method suitable for digital control system
CN101847044B (en) Low-power consumption management method for petascale computer cluster
CN103064741B (en) A kind of method of the divided load scheduling based on energy model

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant