CN104182180B - Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system - Google Patents
Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system Download PDFInfo
- Publication number
- CN104182180B CN104182180B CN201410369550.0A CN201410369550A CN104182180B CN 104182180 B CN104182180 B CN 104182180B CN 201410369550 A CN201410369550 A CN 201410369550A CN 104182180 B CN104182180 B CN 104182180B
- Authority
- CN
- China
- Prior art keywords
- task
- time
- edf
- pcm
- real
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a low-energy EDF (earliest deadline first) real-time task scheduling method for a mixed main memory embedded system. With the adoption of advantages that a PCM (phase-change memory) is non-volatile, low in energy and high in performance and in combination of a dynamic EDF algorithm, real-time constraint of the whole task set is guaranteed, accordingly, power consumption of the whole system is reduced, and the real-time constraint of tasks is not influenced. The method comprises the steps as follows: 1), tasks in the task set T are arranged according to a (Wpi-Wdi)Nwi descending order; 2), all the tasks are initialized; 3), the tasks are placed in the PCM one by one according to a task sequence of the task set T, if the task set still can be scheduled, the task is marked as P-task, Ci is equal to Wpi, and operation is performed until all the tasks in the task set T are checked; 4), the system starts to execute the tasks; 5), free time which is assigned to all the tasks with the dynamic EDF algorithm is calculated; 6), the free time is assigned with the dynamic EDF algorithm according to priority; and 7), the step 6 is repeated until the whole task set T is finished.
Description
Technical field
It is the present invention relates to real time embedded system (Real-Time Embedded System) field, more particularly to a kind of
The low energy consumption EDF real-time task scheduling methods of embedded system are hosted towards mixing.
Background technology
Embedded system is a computer system towards some application-specifics.In view of the safety of embedded system
With the factor such as reliability, its application generally has real-time constraint.In recent years, embedded system is developed rapidly, various
Smart machine is quietly into the life of people.However, as function and application becomes to become increasingly complex, the service life of battery into
Limit for the maximum on these equipment.Research shows, in modern embedded system, the energy consumption of main memory is in whole system energy consumption
Shared ratio is increasing.Therefore, it is the effective method for extending battery pot life to reduce the energy consumption for hosting, and such as
What reduces the key issue that the energy consumption of main storage system is a urgent need to resolve.
Phase transition storage (Phase-change memory, PCM), as which is non-volatile, the characteristic such as low-power consumption and high-performance
Cause the extensive concern of academia and industrial quarters.Compared with tradition hosts DRAM, PCM possesses low energy consumption and non-volatile excellent
Point.Although PCM possesses higher readwrite performance compared with FLASH, but compared with DRAM, still prolong with higher read-write
When, particularly write time delay.Meanwhile, PCM greatly limit its service life with number of times restriction, the feature is write.
The pluses and minuses of comprehensive PCM and DRAM, academia propose the mixing based on PCM and DRAM and host framework (Hybrid
Main Memory Architecture), i.e., high-performance (the low read-write time delay of DRAM) is obtained using DRAM, while utilizing PCM
Obtain larger energy consumption and save (low energy consumption of PCM).
However, the introducing of mixing main storage system causes real-time task scheduling problem to become more complicated:As smart machine,
It would be desirable to provide higher performance but while consume more energy, and as embedded system, its should maximum electrochemical cell make
With the life-span but result in the execution time delay of task, or even the real-time constraint for destroying task.Therefore, performance and both energy consumptions
Balance be one need solve major issue.
Although the research in recent years for mixing main storage system is more, in terms of hosting real-time task scheduling for mixing
Research it is less.Research to mixing main storage system is concentrated mainly on the support of operating system, variable from task different at present
The aspects such as distribution, energy optimizing model, main memory controller in main memory medium.Current research only considered the distribution of task and
Do not consider the scheduling of task, optimize just for main memory controller and do not consider specific real-time scheduling.Therefore, study
It is a major issue for being worth research that mixing hosts the real-time task scheduling under framework.
The content of the invention
For solving the above problems, it is contemplated that embedded system is hosted for mixing, propose the real-time tune of a task
Degree method is farthest to save energy consumption, while ensureing the real-time constraint of whole task-set.It is mixed what is the present invention relates to
Close in storage architecture, PCM and DRAM takes unified addressing mode, and CPU can be to the direct access of various pieces.Operating system distinguishes this
Two parts address space, and which is managed.Real-time scheduling proposed by the present invention is intended to operating system aspect and ensures mixed
Close high-performance and the low energy consumption of main storage system.The technical solution adopted in the present invention is as follows:
A kind of low energy consumption EDF real-time task scheduling methods for hosting embedded system towards mixing, comprise the following steps:
1) by the task in task-set T according to (Wpi-Wdi)/NwiDescending is arranged, wherein WpiRepresent the task in PCM
Worst condition performs time, WdiRepresent that worst condition of the task in DRAM performs time, NwiRepresent the execution of the task
Number of times is write in journey;
2) all tasks are initialized:It is D-task, and C by all task flaggingsi=Wdi, wherein CiRepresent the task most
Difference situation performs the time;
3) task is put in PCM one by one according to the task order of task-set T, if task-set still schedulable, marks
The task is remembered for P-task, and Ci=Wpi, until all task inspections are finished in task-set T;
4) system starts execution task:Wherein D-task is performed in DRAM, and P-task is performed in PCM;
5) calculate " free time " that dynamic EDF algorithms distribute to all tasks;
6) according to EDF prioritizations, team's head element TiPossess the deadline of minimum, be expressed as di, dynamic EDF algorithms
The D-task task instances that will perform " free time " will be distributed to according to priority, until the task terminates;
7) repeat step 6 is until whole task-set T terminates.
In the step 3, the schedulable sufficient and necessary condition of task-set is:
Wherein CiFor task TiWorst condition perform the time, PiFor task TiDuty cycle.
In the step 5, " free time " is calculated as follows:
Wherein, diExpression task TiDeadline, dxFor task TxDeadline, CxExpression task TxWorst condition
The execution time.
When dynamic EDF algorithms " free time " will distribute to task t in the step 6i, data structure Preempt-
Queue is non-NULL, and now dynamic EDF algorithms do not carry out redistributing for free time.
If the D-task is converted to P-task by free time enough in the step 6, then the task is put in PCM
Perform, until the task terminates.
If free time is insufficient to for the D-task to be converted to P-task in the step 6, then according to maximum transport number
According to amount SiWith migration rate computation migration time of the data in different main memory, it is assumed that migration time is migTimei, free time
For slack time, the task is put into dynamic EDF algorithms the time performed in PCM i.e. slack time-migTimeiIf,
The task has been not carried out within this time, then move in DRAM task from PCM, until the tasks carrying is complete.
When a D-task task is assigned the extra time in the step 6, but before its transition process occurs
Be preempted, when task from seize middle recovery when, dynamic EDF algorithms will carry out second time distribution to the task, i.e.,
When which is when middle recovery is seized, as seizing for task may produce new " free time ", the part-time can be distributed to
Being preempted for the task.
If task migrates back DRAM from PCM in the step 6, then the task can have been performed in DRAM always,
Extra time will not be allocated.
Beneficial effect:
What the present invention was obtained has the beneficial effect that:
1) reduce power consumption;
2) ensure that the real-time constraint of whole task-set;
3) ensure that as few as possible write operation in PCM;
4) ensure that task immigration can only from PCM to DRAM in migrate, without moving back to, and each task is at most only migrated
Once.
Description of the drawings
The mixing main storage system framework that Fig. 1 present invention is adopted;
The flow chart of Fig. 2 present invention;
Fig. 3 a dynamics EDF (Dynamic-EDF) dispatching algorithm statistics whole free times;
Fig. 3 b tasks T3Terminate in advance to produce 40 free times;
Fig. 3 c all free times distribute to task T1, cause task T2Property constraint when losing;
The execution flow chart of a D-task task in Fig. 4 dynamic EDF (Dynamic-EDF) algorithms.
Wherein, D1, D2, D3Respectively T1, T2, T3Deadline.
Specific embodiment
Framework is hosted for mixing, as shown in Figure 1, storage both included the mixing main storage system framework that the present invention is adopted
DRAM storages are stored including PCM again.The present invention proposes 2 Real-Time Task Schedule Algorithms based on EDF algorithms, including static EDF
Dispatching algorithm (static-EDF) and dynamic EDF dispatching algorithms (dynamic-EDF).
Fig. 2 is the flow chart of the present invention.A kind of low energy consumption EDF real-time task schedulings for hosting embedded system towards mixing
Method, comprises the following steps:
1) by the task in task-set T according to (Wpi-Wdi)/NwiDescending is arranged, wherein WpiRepresent the task in PCM
Worst condition performs time, WdiRepresent that worst condition of the task in DRAM performs time, NwiRepresent the execution of the task
Number of times is write in journey;
2) all tasks are initialized:It is D-task, and C by all task flaggingsi=Wdi, wherein CiRepresent the task most
Difference situation performs the time;
3) task is put in PCM one by one according to the task order of task-set T, if task-set still schedulable, marks
The task is remembered for P-task, and Ci=Wpi, until all task inspections are finished in task-set T;
4) system starts execution task:Wherein D-task is performed in DRAM, and P-task is performed in PCM;
5) calculate " free time " that dynamic EDF algorithms distribute to all tasks;
6) according to EDF prioritizations, team's head element TiPossess the deadline of minimum, be expressed as di, dynamic EDF algorithms
The D-task task instances that will perform " free time " will be distributed to according to priority, until the task terminates;
7) repeat step 6 is until whole task-set T terminates.
In traditional real-time system model, EDF algorithms assume that each task must terminate before once calling on which, and
Before scheduling, all tasks are simultaneously ready.Same hypothesis is taken in the present invention.Additionally, all of task is all independent in the present invention
, there is no dependence between task, and all tasks do not have not preemptible part.For the sake of simplicity, present invention assumes that
The expense of scheduling is ignored.
In the present invention, for a cycle task-set T={ T1,T2,…,Tn, wherein T1, T2..., TnFor task, n is
Natural number, each periodic duty TiRepresented by a five-tuple<Wdi,Wpi,Pi,Nwi,Si>, wherein WdiRepresent the task in DRAM
In worst condition perform the time (WCET), WpiRepresent WCET (in the present invention, W of the task in PCMdi<Wpi), PiFor
The cycle of the task, NwiNumber of times, S are write in for the implementation procedure of the taskiRepresent when task needs migration, need migration
Maximum amount of data (code segment, data segment, stack segment including the task etc.), i is the natural number less than or equal to n.It is of the invention false
If all tasks are not needed and user mutual in the process of implementation, this is rational for real time embedded system.
EDF dispatching algorithms proposed by the invention ensure that task immigration can only from PCM to DRAM in migrate, without moving
Return, and each task is at most only migrated once.
1st, static state EDF dispatching algorithms
It is in static EDF dispatching algorithms (static-EDF), for periodic duty collection, as far as possible few in PCM in order to ensure
Write operation (extends the PCM life-spans), by all tasks according to (Wpi-Wdi)/NwiDescending sort, one by one task trial are held in being put into PCM
OK.As task to be put in PCM the execution time of the task that increased, it is therefore desirable to which whether inspection now destroys task-set
Schedulability, if the task is put in PCM, the real-time constraint of all tasks is unaffected, then be arranged in the task
Perform in PCM, and the labelling task is P-task, otherwise the task is D-task.
For EDF algorithms, the schedulable necessary and sufficient condition of task-set is not higher than 100% for cpu busy percentage, i.e.,:
CiFor task TiWorst condition perform time WCET, PiFor task TiCycle, n is natural number, i be less than etc.
In the natural number of n.
Static EDF (static-EDF) algorithm steps are as follows:
(1) for the task in task-set T is according to (Wpi-Wdi)/NwiDescending sort;
(2) all tasks are initialized for D-task, and Ci=Wdi;
(3) attempt one by one putting it in PCM according to the task order in task-set T, according to aforementioned schedulability rule
The schedulability for destroying task-set is checked whether, if task-set still schedulable, the labelling task is P-task, and Ci
=Wpi;
(4) (3) are repeated until all task inspections are finished in T;
(5) task-set is dispatched according to EDF dispatching algorithms, wherein D-task is performed in DRAM, and P-task is held in PCM
OK.
2nd, dynamic EDF (dynamic-EDF) dispatching algorithm
In the present invention, dynamic dispatching algorithm is the optimization to static scheduling algorithm.In static scheduling algorithm, algorithm is
Each task has been reserved its worst condition and has performed the time (WCET), but task, in actual execution, its actual execution time is past
Past is far smaller than its WCET.Therefore, when task is fulfiled ahead of schedule, compared with reserved WCET, can produce and " be not used
Time ", in the present invention, these times are called " free time " (Slack Time).Obviously, if scheduling is reasonable, protecting
In the case of card task-set is schedulable, slack time can be reassigned to having not carried out for task.Thus, can be part
D-task distributes more times so as to can perform in PCM, and then obtains more preferable energy consumption saving.
However, when the slack time for reclaiming are not enough to for whole D-task to be completely converted into P-task, in order to ensure
The real-time constraint of task, sometimes has to migrate task in PCM and DRAM, i.e., task front portion is performed in PCM, and
Rear portion is performed in having to move to DRAM.The migration of task result in extra time, control and energy expense, therefore
The present invention avoids and minimizes the migration of task, dynamic dispatching algorithm to ensure that all tasks at most can be migrated once as far as possible.
In dynamic dispatching algorithm, key issue is the calculating and distribution of slack time, therefore, the present invention is solved emphatically
Determine the problem, and then realize the scheduling of task-set.
For optimizing static state EDF algorithms, as EDF algorithms are dynamic priority scheduling algorithm, therefore, the present invention is from the overall situation
Angle, is managed and distributes to global " free time " i.e. slack time, and be not limited solely to nearest
Deadline (nearest deadline).
However, being not blindly safe using all of slack time.Consider 3 periodic duty T1, T2And T3,
Its task parameters is Wd1=100, Wp1=140, P1=300;Wd2=180, P2=300;Wd3=60, P3=900.All tasks are all
For D-task, such as shown in accompanying drawing 3 (a).If T3Fulfil ahead of schedule when t=300, result in 40 slack time, it is such as attached
Shown in Fig. 3 (b).If all this 40 unit interval are fully allocated to T1, then T1It is changed into P-task, but but result in and appoint
Business T2Do not completed before its deadline, destroy its real-time and constrain (if T1And T2Which is consumed in the process of implementation
Respective WCET), such as shown in accompanying drawing 3 (c).
Therefore, in order to avoid above-mentioned situation, before to task distribution slack time, need to calculate its " can use "
slack time.For this purpose, invention introduces a new data structure, Ready-Queue, to record static-EDF algorithms
Distribute to " free time " of all tasks.To a cycle task-set T={ T1,T2,…,Tn, as task is had at which
Complete before calling next time, therefore in Ready-Queue, at most include n element, n is natural number.To in Ready-Queue
Element is according to EDF prioritizations, team's head element TiPossess the deadline of minimum, be expressed as di, i is the nature less than or equal to n
Number.
Using Ready-Queue, the present invention can easily its available slack time to each task computation.To i.e.
To perform for task Ti, it is higher than T that its available slack time is priorityiAnd the summation of the free time of completed task,
Due to now with high priority task it is actual execution in completed, and its show in Ready-Queue its
Now go back in static-EDF scheduling and do not complete.Its available slack time is calculated as follows:
As EDF algorithms are dynamic priority task scheduling algorithms, therefore in dynamic EDF algorithms, low priority task
Can be seized by high-priority task at any time.If task TiBy task TjSeize, if being distributed to originally TiSlack
Time distributes to Tj, task T may be destroyediReal-time constraint, because before seizing, task TiIn low speed internal memory
(PCM) a period of time has been run in.Therefore, needs when seizing of task avoid seizing its allocated slack
Time, for this purpose, present invention introduces another data structure, Preempt-Queue, safeguard that the task of being preempted is the allocated
slack time.When Preempt-Queue non-NULLs, dynamic EDF algorithms do not carry out redistributing for free time.
The principle of dynamic EDF algorithms distribution slack time is as follows:If the D-task is changed by slack time enough
For P-task, then this time calling for the task is put into into execution (attention in PCM:The periodic duty remains as D-task, is this
The secondary example for calling is performed in being put into PCM).If slack time are not enough to the task is put in PCM completely, dynamic EDF
Algorithm is put into the task part in PCM, and another part is performed in being put into DRAM.
When a D-task is assigned the extra time, but it was preempted before its transition process occurs.When which is from robbing
When accounting for middle recovery, as seizing for task may produce new slack time, the part-time can be distributed to this and be preempted
Task, the migration of the task may be avoided.I.e. when task from seize middle recovery when, dynamic EDF algorithms may be right
The task carries out second time distribution.If task migrates back DRAM from PCM, the task can be held in DRAM always
Go, extra time will not be allocated.
In order to provide the detailed implementation process of the present invention, with reference to pseudo-code of the algorithm to static EDF proposed by the present invention
Algorithm and dynamic EDF algorithms do and further describe in detail.
1st, static state EDF dispatching algorithms
Static scheduling algorithm proposed by the present invention is put into PCM by task one by one, then tests whole task-set schedulable
Property the static storage medium for determining tasks carrying of mode.When initial, all tasks are D-task, if a task
Perform during PCM can be put into completely, then the task flagging is P-task by static scheduling algorithm.In static scheduling algorithm, appoint
The attribute of business is once it is determined that (D-task or P-task), then, in the implementation procedure of whole task-set, the attribute of the task is not
Can change.The implementation procedure of static EDF dispatching algorithms is as follows:
Obviously, if a task-set is schedulable (pure DRAM is hosted in framework) in EDF algorithms, the task-set
It is also schedulable (mixing PCM/DRAM is hosted in framework) in static EDF (static-EDF) algorithm.
2nd, dynamic EDF dispatching algorithms
Before statement dynamic EDF algorithms in detail, each symbol and its implication that table 1 is used in listing dynamic EDF algorithms.
Each symbol and description in 1. dynamic EDF algorithms of table
Symbol | Description |
ti | Task TiTask instances (or task TiOnce call) |
allocationi | Distribute to tiTask time |
migTimei | Task TiThe maximum migration time in DRAM is moved to from PCM |
migFlagi | Task tiMigration flag bit, 1 represent tiNeeds move to DRAM from PCM |
exedTimei | Example tiTime (task T performed in PCMiFor D-task) |
PTimei | Example tiAllocated PCM time (tasks TiFor D-task) |
Ci | Task tiWCET, if TiFor P-task, then Ci=Wpi, otherwise Ci=Wdi |
The present invention adopts CiLogger task tiWorst condition perform the time, if task be P-task, initialize Ci=
Wpi, otherwise, initialize Ci=Wdi.And if only if CiWhen=0, the task is removed from Ready-Queue.
In dynamic-EDF algorithms, constantly dynamic updates queue Ready-Queue.First element in queue
CiSuccessively decrease over time, work as CiFor 0 when first element delete from queue, hereafter queue the next one element repeat this mistake
Journey.For the consideration of efficiency, the present invention is only reached in task, task is completed, task immigration when perform Ready-Queue's
Update, its renewal process UpdateQueue (t) is as follows, and wherein symbol t is represented from elapsed time after last time event.
The concrete implementation procedure of dynamic EDF dispatching algorithms (dynamic-EDF) is as follows:
The execution flow chart of a D-task task in Fig. 4 dynamic EDF (Dynamic-EDF) algorithms.Dynamic EDF algorithms are given
Unfinished task distributes the extra time, but allocation algorithm will not be seized static EDF algorithms and be initially allocated to other tasks
Time.If the D-task is converted to P-task by slack time enough, this time calling the task to be put in PCM and hold
Row (notes:The periodic duty remains as D-task, and the example for simply this time calling is performed in being put into PCM).If slack
Time is not enough to the task is put in PCM completely, then algorithm checks whether free time can ensure that task is performed in PCM
At least threshold% (such as 50%).Reason for doing so is that, the actual execution time of task is far smaller than its WCET,
Threshold value threshold% may can meet the task and perform in PCM completely, so as to avoid the migration of task.Ready-
Queue ensure that the safety distribution of free time, and the introducing of Preempt-Queue queues guarantees that high-priority task will not be seized
The execution for having distributed to the time of low priority task, i.e. task can be preempted, but distribute to the free time of the task
Can not be preempted.Dynamic EDF algorithms proposed by the present invention ensure that:In any time of EDF algorithm performs, all tasks it is complete
Will not be more late than the deadline in static EDF algorithms into the time.Similar theory is in paper " Dynamic and
Give in aggressive scheduling techniques for power-aware real-timesystems " in detail
It is thin to prove, therefore, if a task-set is schedulable under static-EDF algorithms, which is in dynamic-EDF algorithms
Under be still schedulable, i.e., dynamic EDF algorithms ensure that the schedulability of task-set.
Additionally, in dynamic EDF dispatching methods, if current idle deficiency of time is held so that whole task is put in PCM
OK, the task will be moved in DRAM from PCM.After task immigration, if now seized by high-priority task, the task is not
Can enter enqueue Preempt-Queue, thus when the task allocationi≤Ci(dynamic the 4th row of EDF algorithms).When this
When middle recovery is seized, it will not be reallocated the time for business, because only in recovering for task is Preempt-Queue
Only element when just can by again distribute slack time (dynamic the 6th row of EDF algorithms).Therefore, the task after migration will
Perform in DRAM, until the tasks carrying is complete, i.e., each task is at most migrated once.
Although the above-mentioned accompanying drawing that combines is described to the specific embodiment of the present invention, not to present invention protection model
The restriction enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
The various modifications made by needing to pay creative work or deformation are still within protection scope of the present invention.
Claims (7)
1. a kind of low energy consumption EDF real-time task scheduling methods for hosting embedded system towards mixing, is characterized in that, including following
Step:
1) by the task in task-set T according to (Wpi-Wdi)/NwiDescending is arranged, wherein WpiRepresent that the task is worst in PCM
Situation performs time, WdiRepresent that worst condition of the task in DRAM performs time, NwiIn representing the implementation procedure of the task
Write number of times, in the implementation procedure of the task to write number of times in the PCM or in DRAM be identical;
2) all tasks are initialized:It is D-task, and C by all task flaggingsi=Wdi, wherein CiRepresent the worst feelings of the task
Condition performs the time;
3) task is put in PCM one by one according to the task order of task-set T, if task-set still schedulable, labelling should
Task is P-task, and Ci=Wpi, until all task inspections are finished in task-set T;
4) system starts execution task:Wherein D-task is performed in DRAM, and P-task is performed in PCM;
5) calculate the free time that dynamic EDF algorithms distribute to all tasks;The free time is
Wherein diExpression task TiDeadline, dxFor task TxDeadline, CxExpression task TxWorst condition perform
Time;
6) according to EDF prioritizations, team's head element ti possesses the deadline of minimum, is expressed as di, dynamic EDF algorithms according to
Priority distributes to the D-task task instances that will be performed by " free time ", until the task terminates;
7) repeat step 6 is until whole task-set T terminates.
2. as claimed in claim 1 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, in the step 3, the schedulable sufficient and necessary condition of task-set is:
Wherein CiFor task TiWorst condition perform the time, PiFor task TiDuty cycle.
3. as claimed in claim 1 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, when dynamic EDF algorithms " free time " will distribute to task t in the step 6i, data structure Preempt-
Queue is non-NULL, and now dynamic EDF algorithms do not carry out redistributing for free time.
4. as claimed in claim 1 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, if the D-task is converted to P-task by free time enough in the step 6, then the task is put into into PCM
Middle execution, until the task terminates.
5. as claimed in claim 1 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, if free time is insufficient to for the D-task to be converted to P-task in the step 6, then according to maximum transport number
According to amount SiWith migration rate computation migration time of the data in different main memory, it is assumed that migration time is migTimei, when idle
Between be slack time, the task is put into dynamic EDF algorithms the time performed in PCM i.e. slack time-migTimei, such as
Fruit task within this time has been not carried out, then move in DRAM task from PCM, until the tasks carrying is complete.
6. as claimed in claim 1 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, when a D-task task is assigned the extra time in the step 6, but before its transition process occurs
Be preempted, when task from seize middle recovery when, dynamic EDF algorithms will carry out second time distribution to the task, i.e.,
When which is when middle recovery is seized, as seizing for task may produce new free time, the part-time can distribute to this
Being preempted for task.
7. as claimed in claim 4 a kind of towards the low energy consumption EDF real-time task scheduling methods for mixing main memory embedded system,
It is characterized in that, if task migrates back DRAM from PCM in the step 6, then the task can have been performed in DRAM always,
Extra time will not be allocated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410369550.0A CN104182180B (en) | 2014-07-30 | 2014-07-30 | Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410369550.0A CN104182180B (en) | 2014-07-30 | 2014-07-30 | Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104182180A CN104182180A (en) | 2014-12-03 |
CN104182180B true CN104182180B (en) | 2017-03-22 |
Family
ID=51963268
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410369550.0A Expired - Fee Related CN104182180B (en) | 2014-07-30 | 2014-07-30 | Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104182180B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104571462B (en) * | 2014-12-30 | 2017-05-03 | 深圳先进技术研究院 | Method and system for controlling battery power dissipation |
CN115858112B (en) * | 2022-11-18 | 2024-02-09 | 南京航空航天大学 | Constraint planning-based comprehensive avionics system task allocation and scheduling method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101226421A (en) * | 2008-01-16 | 2008-07-23 | 浙江大学 | MSR method for real time embedded system EDF low-power consumption scheduling |
CN103810026A (en) * | 2012-11-09 | 2014-05-21 | 中国科学院沈阳计算技术研究所有限公司 | Mixing scheduling method suitable for real-time system periodic tasks |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120079486A1 (en) * | 2010-09-23 | 2012-03-29 | International Business Machines Corporation | Integration of dissimilar job types into an earliest deadline first (edf) schedule |
KR101311305B1 (en) * | 2011-08-26 | 2013-09-25 | 국방과학연구소 | System and method for deadline based priority inheritance |
-
2014
- 2014-07-30 CN CN201410369550.0A patent/CN104182180B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101226421A (en) * | 2008-01-16 | 2008-07-23 | 浙江大学 | MSR method for real time embedded system EDF low-power consumption scheduling |
CN103810026A (en) * | 2012-11-09 | 2014-05-21 | 中国科学院沈阳计算技术研究所有限公司 | Mixing scheduling method suitable for real-time system periodic tasks |
Also Published As
Publication number | Publication date |
---|---|
CN104182180A (en) | 2014-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102193826B (en) | Method for high-efficiency task scheduling of heterogeneous multi-core processor | |
CN102043675B (en) | Thread pool management method based on task quantity of task processing request | |
CN104572106A (en) | Concurrent program developing method for processing of large-scale data based on small memory | |
CN102779072B (en) | Embedded system and dormancy and wake-up method of application process thereof | |
CN105117285B (en) | A kind of nonvolatile memory method for optimizing scheduling based on mobile virtual system | |
CN109324880A (en) | A kind of low-power consumption scheduling method suitable for real-time system periodic task model | |
CN102171627A (en) | Scheduling an application for performance on a heterogeneous computing system | |
WO2017080276A1 (en) | Resource management method and system, and computer storage medium | |
Li et al. | Energy-efficient scheduling in nonpreemptive systems with real-time constraints | |
US20090144228A1 (en) | Data parallel production and consumption | |
Legout et al. | Mixed-criticality multiprocessor real-time systems: Energy consumption vs deadline misses | |
Yoo et al. | Integrated scheduling of real-time and interactive tasks for configurable industrial systems | |
Lindberg et al. | Comparison and analysis of greedy energy-efficient scheduling algorithms for computational grids | |
Panda et al. | An efficient energy saving task consolidation algorithm for cloud computing systems | |
CN104182180B (en) | Low-energy EDF (earliest deadline first) real-time task scheduling method for mixed main memory embedded system | |
CN104182280B (en) | Low-energy RM real-time task scheduling method for hybrid main memory embedded system | |
CN109324891A (en) | A kind of periodic duty low-power consumption scheduling method of ratio free time distribution | |
CN108984298A (en) | A kind of resource regulating method and system of cloud computing platform | |
Chen et al. | Online dynamic power management with hard real-time guarantees | |
Ansari et al. | Power-aware scheduling of fixed priority tasks in soft real-time multicore systems | |
CN115756143A (en) | Energy-saving method and device for data packet processing, computer equipment and storage medium | |
He et al. | Enhanced schedulability analysis of hard real-time systems on power manageable multi-core platforms | |
CN106933325B (en) | A kind of fixed priority I/O device energy consumption management method | |
CN103440533B (en) | The confining method of the non-bottleneck ability of job shop under a kind of cloud manufacturing mode | |
CN108563497B (en) | Efficient multi-dimensional algorithm scheduling method and task 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 | ||
C14 | Grant of patent or utility model | ||
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: 20170322 Termination date: 20210730 |