CN109298917B - Self-adaptive scheduling method suitable for real-time system mixed task - Google Patents

Self-adaptive scheduling method suitable for real-time system mixed task Download PDF

Info

Publication number
CN109298917B
CN109298917B CN201710610065.1A CN201710610065A CN109298917B CN 109298917 B CN109298917 B CN 109298917B CN 201710610065 A CN201710610065 A CN 201710610065A CN 109298917 B CN109298917 B CN 109298917B
Authority
CN
China
Prior art keywords
task
time
periodic
deadline
subtask
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
CN201710610065.1A
Other languages
Chinese (zh)
Other versions
CN109298917A (en
Inventor
郭锐锋
彭阿珍
胡毅
吴昊天
邓昌义
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 Zhongke Cnc Technology Co ltd
Original Assignee
Shenyang Golding Nc & Intelligence Tech Co ltd
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 Golding Nc & Intelligence Tech Co ltd filed Critical Shenyang Golding Nc & Intelligence Tech Co ltd
Priority to CN201710610065.1A priority Critical patent/CN109298917B/en
Publication of CN109298917A publication Critical patent/CN109298917A/en
Application granted granted Critical
Publication of CN109298917B publication Critical patent/CN109298917B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to a self-adaptive scheduling method suitable for a mixed task of a real-time system, which is characterized in that the total utilization rate of periodic tasks and the total utilization rate of non-periodic tasks are calculated before the mixed task set is scheduled; sequencing a periodic task set and inserting the periodic task set into a ready queue; when the non-periodic task arrives, calculating the predicted execution time of the non-periodic task, and dividing the non-periodic task into two subtasks; allocating deadline for the main subtask, inserting the deadline into a ready queue, and updating the deadline of the main subtask after the main subtask is executed; and updating the deadline of the aperiodic task if the aperiodic task is completed before the deadline of the main subtask. On the premise of ensuring the schedulability of periodic tasks in the system, the invention allocates the deadline for the non-periodic task by adopting the predicted execution time, thereby shortening the response time of the non-periodic task; and allocating the calculation bandwidth of the processor for the periodic task and the non-periodic task, thereby improving the overall performance of the system.

Description

Self-adaptive scheduling method suitable for real-time system mixed task
Technical Field
The invention relates to the field of real-time system hybrid scheduling, in particular to an adaptive scheduling method suitable for a real-time system hybrid task.
Background
With the increasing diversity and complexity of real-time embedded systems, it is more and more common for multiple types of hard real-time, soft real-time, and non-real-time tasks to coexist in the same system. This puts new requirements on the scheduling algorithm: the response time of soft real-time and non-real-time tasks is reduced as much as possible while ensuring that hard real-time tasks are completed before their deadlines.
Currently, the scheduling for the mixed task set mainly adopts a server-based method, and the essence of the method is to create and execute a periodic task to process an aperiodic task. The Total Bandwidth Server (TBS) algorithm is a server approach that can be used to efficiently schedule a mixed task set based on EDFs. The TBS algorithm aims to shorten the response time of the non-periodic task as much as possible on the premise of ensuring the schedulability of the periodic task.
The existing TBS-based algorithm for a mixed task set schedules according to the WCET of an aperiodic task, and obtains shorter response time of the aperiodic task at the cost of sacrificing the execution time of the periodic task. In practice, however, the actual execution time of the aperiodic task is usually much smaller than its WCET, and scheduling according to its WCET will result in too long response time of the aperiodic task.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an adaptive scheduling method suitable for a mixed task of a real-time system, which adopts the predicted execution time of an aperiodic task to replace the WCET of the aperiodic task for scheduling and shortens the response time of the aperiodic task on the premise of ensuring the scheduling integrity of the periodic task.
The technical scheme adopted by the invention for realizing the purpose is as follows:
a self-adaptive scheduling method suitable for a real-time system mixing task comprises the following steps:
step 1: before the mixed task set is scheduled, the total utilization rate of periodic tasks is calculated, and the total utilization rate of non-periodic tasks is calculated according to the total utilization rate;
step 2: sequencing the periodic task sets according to an earliest deadline first principle and inserting the periodic task sets into a ready queue;
and step 3: when the non-periodic task arrives, calculating the predicted execution time of the non-periodic task, and dividing the non-periodic task into two subtasks, namely a main subtask and a standby subtask;
and 4, step 4: allocating deadline for the main and sub tasks, inserting the main and sub tasks into a ready queue according to an earliest deadline priority principle, and updating the deadline of the main and sub tasks after the main and sub tasks are executed;
and 5: if the non-periodic task is executed and completed before the deadline of the main sub-task, respectively recording the actual execution time and the actual completion time of the non-periodic task, and updating the deadline of the non-periodic task; otherwise, allocating a deadline for the standby subtask, inserting the standby subtask into the ready queue according to the earliest deadline priority principle, updating the deadline of the standby subtask after the standby subtask is executed, completing the execution of the non-periodic task, recording the actual execution time and the actual completion time of the non-periodic task, and updating the deadline of the non-periodic task.
The total utilization rate of the periodic tasks is as follows:
Figure BDA0001359304070000021
the total utilization rate of the aperiodic tasks is as follows:
Figure BDA0001359304070000022
wherein, UpFor the total utilization of periodic tasks, UsFor the total utilization of non-periodic tasks, CiFor periodic tasks TiIs performed in the worst case, PiFor periodic tasks TiThe period of (c).
The predicted execution time of the aperiodic task is as follows:
Figure BDA0001359304070000023
wherein,
Figure BDA0001359304070000024
for non-periodic task JkThe predicted execution time of (a) is,
Figure BDA0001359304070000025
for non-periodic task JkIs executed in the worst case of the execution time,
Figure BDA0001359304070000026
for non-periodic task Jsα is a weight coefficient.
The non-periodic task is divided into two subtasks:
the worst execution time is
Figure BDA0001359304070000031
Time of arrival rkIs non-periodic task JkThe method is divided into two subtasks:
Figure BDA0001359304070000032
and
Figure BDA0001359304070000033
wherein the main subtask
Figure BDA0001359304070000034
Is the worst execution time of
Figure BDA0001359304070000035
Prepare subtask
Figure BDA0001359304070000036
The worst execution time of (c) is:
Figure BDA0001359304070000037
main and sub tasks
Figure BDA0001359304070000038
And prepare subtasks
Figure BDA0001359304070000039
All the arrival times of (are r)k。
The distribution deadline for the main subtask is as follows:
Figure BDA00013593040700000310
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure BDA00013593040700000324
for a preceding non-periodic task Jk-1Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure BDA00013593040700000311
is a main subtask
Figure BDA00013593040700000312
Worst execution time of UsIs the total utilization of the aperiodic task.
The deadline for updating the main subtask is as follows:
Figure BDA00013593040700000313
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure BDA00013593040700000314
for a previous non-periodic task
Figure BDA00013593040700000315
Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure BDA00013593040700000316
is a main subtask
Figure BDA00013593040700000317
Actual execution time of, UsIs the total utilization of the aperiodic task.
The distribution deadline for the standby subtasks is as follows:
Figure BDA00013593040700000318
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure BDA00013593040700000319
the deadline for updating after the execution of the main subtask is finished,
Figure BDA00013593040700000320
being the actual completion time of the main sub-task,
Figure BDA00013593040700000321
for the worst execution time of the standby subtask, UsIs the total utilization of the aperiodic task.
The deadline of the updating backup subtask is as follows:
Figure BDA00013593040700000322
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure BDA00013593040700000323
the deadline for updating after the execution of the main subtask is finished,
Figure BDA0001359304070000041
being the actual completion time of the main sub-task,
Figure BDA0001359304070000042
for preparing the actual execution time of the subtask, UsIs the total utilization of the aperiodic task.
The deadline of the updating non-periodic task is as follows:
Figure BDA0001359304070000043
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure BDA0001359304070000044
for a preceding non-periodic task Jk-1Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure BDA0001359304070000045
for non-periodic task JkActual execution time of, UsIs the total utilization of the aperiodic task.
The invention has the following beneficial effects and advantages:
1. on the premise of ensuring the schedulability of periodic tasks in the system, the invention allocates the deadline for the non-periodic task by adopting the predicted execution time, thereby shortening the response time of the non-periodic task;
2. the invention can allocate the calculation bandwidth of the processor for the periodic task and the non-periodic task as reasonably as possible, thereby improving the overall performance of the system.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
fig. 2 is a graph of a simulation experiment comparing the TBS algorithm with the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Fig. 1 shows a flow chart of the method of the present invention.
The invention comprises the following steps:
calculating the total utilization rate U of the periodic tasks before the mixed task set starts to be scheduledpAnd calculating the total utilization rate U of the non-periodic tasks＝1-Up；
Sequencing the periodic task sets according to an earliest deadline first principle and inserting the periodic task sets into a ready queue;
the kth aperiodic task JkOn arrival, calculate JkPredicted execution time of
Figure BDA0001359304070000046
And will beIt is divided into two subtasks:
Figure BDA0001359304070000047
and
Figure BDA0001359304070000048
taking a resource recovery strategy and a total bandwidth server algorithm as main subtasks
Figure BDA0001359304070000049
Distribution deadline
Figure BDA00013593040700000410
And the main and sub tasks are processed according to the earliest deadline priority algorithm
Figure BDA0001359304070000051
Inserting into a ready queue;
main and sub tasks
Figure BDA0001359304070000052
After the execution is finished, recording
Figure BDA0001359304070000053
Actual execution time of
Figure BDA0001359304070000054
And actual completion time
Figure BDA0001359304070000055
And is updated according to the result
Figure BDA0001359304070000056
The updated deadline is recorded as
Figure BDA0001359304070000057
If the kth aperiodic task JkIn that
Figure BDA0001359304070000058
When the previous execution is finished, the non-periodic task JkEnd of execution, JkActual execution time of
Figure BDA0001359304070000059
And actual completion time fkIs a main subtask
Figure BDA00013593040700000510
Actual execution time of
Figure BDA00013593040700000511
And actual completion time
Figure BDA00013593040700000512
And update J accordinglykThe updated deadline is recorded as
Figure BDA00013593040700000513
Otherwise, the sub-task is prepared according to the resource recovery strategy and the total bandwidth server algorithm
Figure BDA00013593040700000514
Distribution deadline
Figure BDA00013593040700000515
And preparing the subtasks according to the earliest deadline priority algorithm
Figure BDA00013593040700000516
Inserting into a ready queue;
prepare subtask
Figure BDA00013593040700000517
After the execution is finished, recording
Figure BDA00013593040700000518
Actual execution time of
Figure BDA00013593040700000519
And actual completion time
Figure BDA00013593040700000520
And is updated according to the result
Figure BDA00013593040700000521
The deadline is recorded as the updated deadline
Figure BDA00013593040700000522
Now aperiodic task JkAfter execution, JkActual execution time of
Figure BDA00013593040700000523
Is the sum of the actual execution time of the two subtasks and the actual completion time fkFor preparing subtasks
Figure BDA00013593040700000524
Is/are as follows
Figure BDA00013593040700000525
Update JkThe updated deadline is recorded as
Figure BDA00013593040700000526
Further explanation is as follows:
before the task set starts to be scheduled, the total utilization rate U of the periodic tasks is calculated by the formula (1)p：
Figure BDA00013593040700000527
In the formula (1), Ci、PiI are periodic tasks T respectivelyiAnd in the worst case, executing time, period and task identifiers, wherein n is the number of periodic tasks.
Calculating the total utilization rate U of the non-periodic task by the formula (2)s：
Figure BDA00013593040700000528
And inserting all periodic tasks into the ready queue according to an earliest deadline first principle.
Earliest deadline first principle: the shorter the task deadline is, the higher the priority is, and when the deadline of the two tasks is the same, the task with the shorter arrival time has the higher priority; when the deadlines and arrival times of two tasks are the same, the task with the smaller series index has a higher priority.
When the kth aperiodic task JkWhen arriving, calculate J from equation (3)kThe predicted execution time of (c):
Figure BDA0001359304070000061
in the formula (3)
Figure BDA0001359304070000062
Is JkIs executed in the worst case of the execution time,
Figure BDA0001359304070000063
the predicted execution time for the k-1 st aperiodic task,
Figure BDA0001359304070000064
and alpha is a weight coefficient, and is the actual execution time of the s-th aperiodic task.
Total bandwidth server algorithm: and allocating a temporary absolute deadline to the non-periodic task, and then scheduling the non-periodic task and the periodic real-time task which are allocated with the deadline by adopting an earliest deadline priority algorithm. Deadline d of kth aperiodic taskkCalculated from equation (4):
Figure BDA0001359304070000065
wherein r iskFor the arrival time of the kth aperiodic task, dk-1Is the deadline of the k-1 st aperiodic task, CkFor the worst execution time (WCET) of the kth aperiodic tasksProcessor utilization of a server for scheduling aperiodic tasks.
Resource recovery strategy: at the end of the task execution, the deadline is recalculated using the actual execution time, and then the deadline for the next aperiodic task is calculated using the new deadline.
In the resource recovery strategy, the k-th aperiodic task J is given by formula (5)kAssign a new deadline:
Figure BDA0001359304070000066
wherein r iskIs JkThe arrival time of,
Figure BDA0001359304070000067
For a preceding non-periodic task Jk-1Deadline and f for recalculation after execution endsk-1Is Jk-1Completion time of (due to J)k-1Is d 'as a deadline'k-1Is performed so that Jk-1Deadline for recalculation after execution ends
Figure BDA0001359304070000068
May be less than fk-1)、CkWorst execution time (WCET), U for kth aperiodic tasksProcessor utilization of a server for scheduling aperiodic tasks. While
Figure BDA0001359304070000069
Is calculated according to equation (6):
Figure BDA00013593040700000610
wherein r isk-1Is Jk-1The arrival time of,
Figure BDA00013593040700000611
For non-periodic task Jk-2Deadline and f for recalculation after execution endsk-2Is Jk-2The completion time of,
Figure BDA00013593040700000612
For the k-1 st aperiodic task Jk-1Actual execution time, UsProcessor utilization of a server for scheduling aperiodic tasks.
The worst execution time is
Figure BDA0001359304070000071
Time of arrival rkTask J ofkAre divided into arrival times rkTwo subtasks of (2):
Figure BDA0001359304070000072
and
Figure BDA0001359304070000073
task
Figure BDA0001359304070000074
Is the worst execution time of
Figure BDA0001359304070000075
Task
Figure BDA0001359304070000076
The worst execution time of (c) is:
Figure BDA0001359304070000077
is an aperiodic task J according to equation (7)kMain and sub tasks
Figure BDA0001359304070000078
Distribution deadline:
Figure BDA0001359304070000079
wherein r iskFor non-periodic task JkTime of arrival of fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure BDA00013593040700000710
for a preceding non-periodic task Jk-1The deadline for updating after the execution is finished.
Will main subtask
Figure BDA00013593040700000711
And adding the data into the ready queue according to an earliest deadline first algorithm. In that
Figure BDA00013593040700000712
After completion, the actual execution time is recorded as
Figure BDA00013593040700000713
The actual completion time is
Figure BDA00013593040700000714
And updated according to equation (8)
Figure BDA00013593040700000715
The deadline of (2):
Figure BDA00013593040700000716
if JkAt main subtask deadline
Figure BDA00013593040700000717
When the previous execution is finished, the non-periodic task JkThe actual execution time is
Figure BDA00013593040700000718
The actual completion time is
Figure BDA00013593040700000719
Updating J by equation (9)kThe deadline of (2):
Figure BDA00013593040700000720
otherwise, the formula (10) is used as a standby subtask
Figure BDA00013593040700000721
Distribution deadline:
Figure BDA00013593040700000722
will be prepared for subtask
Figure BDA00013593040700000723
And adding the data into the ready queue according to an earliest deadline first algorithm. In that
Figure BDA00013593040700000724
After completion, the actual execution time is recorded as
Figure BDA00013593040700000725
The actual completion time is
Figure BDA00013593040700000726
And updated by the formula (11)
Figure BDA00013593040700000727
The deadline of (2):
Figure BDA00013593040700000728
hence the aperiodic task JkThe actual execution time is
Figure BDA00013593040700000729
The actual completion time is
Figure BDA00013593040700000730
Updating J according to equation (12)kThe deadline of (2):
Figure BDA00013593040700000731
example (b):
in order to compare the performance of the method and the TBS algorithm in the aspect of scheduling the mixed task set, the average response time of the non-periodic task is selected as an index for measuring the performance of the algorithm.
In a simulation experiment, 10 periodic task sets and 10 sporadic task sets are selected to form 100 mixed task sets, and the final result is the average value of the 100 mixed task sets. For a periodic task, the period follows an exponential distribution with the average value of 100, and the worst case execution time is equal to the actual execution time and follows an exponential distribution with the average value of 10. One aperiodic task set comprises 4 aperiodic tasks, and the arrival time of the aperiodic tasks obeys Poisson distribution with the average value of 2. The worst execution time of the non-periodic task follows an exponential distribution with a mean value of 8, and the upper limit of the actual execution time is the corresponding worst execution time and follows an exponential distribution with a mean value of 4. The difference in average response time of the inventive method and TBS algorithm was observed when the periodic task total resource utilization varied from 60% to 90%.
FIG. 2 is a diagram of a simulation experiment comparing the TBS algorithm with the present invention, wherein the abscissa is the total utilization rate U of the periodic taskpAnd the ordinate is the average response time of the aperiodic task. As can be seen from FIG. 2, UpAt 65%, the difference between the TBS and the average response time of the method of the invention is small. This is due to the fact that there is now sufficient bandwidth (1-U)p35%) process requests for non-periodic tasks. When U is turnedpWhen the average response time of the non-periodic tasks exceeds 70 percent, the average response time of the non-periodic tasks begins to be different when U is equal topAt 90%, the difference is maximal. Simulation experiments show that compared with TBS, the method of the invention can effectively shorten the response time of the non-periodic task, thereby improving the overall performance of the system.

Claims (9)

1. A self-adaptive scheduling method suitable for a real-time system mixing task is characterized by comprising the following steps: the method comprises the following steps:
step 1: before the mixed task set is scheduled, the total utilization rate of periodic tasks is calculated, and the total utilization rate of non-periodic tasks is calculated according to the total utilization rate;
step 2: sequencing the periodic task sets according to an earliest deadline first principle and inserting the periodic task sets into a ready queue;
and step 3: when the non-periodic task arrives, calculating the predicted execution time of the non-periodic task, and dividing the non-periodic task into two subtasks, namely a main subtask and a standby subtask;
and 4, step 4: allocating deadline for the main and sub tasks, inserting the main and sub tasks into a ready queue according to an earliest deadline priority principle, and updating the deadline of the main and sub tasks after the main and sub tasks are executed;
and 5: if the non-periodic task is executed and completed before the deadline of the main sub-task, respectively recording the actual execution time and the actual completion time of the non-periodic task, and updating the deadline of the non-periodic task; otherwise, allocating a deadline for the standby subtask, inserting the standby subtask into the ready queue according to the earliest deadline priority principle, updating the deadline of the standby subtask after the standby subtask is executed, completing the execution of the non-periodic task, recording the actual execution time and the actual completion time of the non-periodic task, and updating the deadline of the non-periodic task.
2. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the total utilization rate of the periodic tasks is as follows:
Figure FDA0002552062860000011
the total utilization rate of the aperiodic tasks is as follows:
Figure FDA0002552062860000012
wherein, UpFor the total utilization of periodic tasks, UsFor the total utilization of non-periodic tasks, CiFor periodic tasks TiIs performed in the worst case, PiFor periodic tasks TiN is periodicThe number of transactions.
3. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the predicted execution time of the aperiodic task is as follows:
Figure FDA0002552062860000021
wherein,
Figure FDA0002552062860000022
for non-periodic task JkThe predicted execution time of (a) is,
Figure FDA0002552062860000023
for non-periodic task JkIs executed in the worst case of the execution time,
Figure FDA0002552062860000024
(s-1, 2, … k-1) is an aperiodic task Jsα is a weight coefficient.
4. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the non-periodic task is divided into two subtasks:
the worst execution time is
Figure FDA0002552062860000025
Time of arrival rkIs non-periodic task JkThe method is divided into two subtasks:
Figure FDA0002552062860000026
and
Figure FDA0002552062860000027
wherein the main subtask
Figure FDA0002552062860000028
Is the worst execution time of
Figure FDA0002552062860000029
Prepare subtask
Figure FDA00025520628600000210
The worst execution time of (c) is:
Figure FDA00025520628600000211
main and sub tasks
Figure FDA00025520628600000212
And prepare subtasks
Figure FDA00025520628600000213
All the arrival times of (are r)k。
5. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the distribution deadline for the main subtask is as follows:
Figure FDA00025520628600000214
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure FDA00025520628600000215
for a preceding non-periodic task Jk-1Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure FDA00025520628600000216
is a main subtask
Figure FDA00025520628600000217
Worst execution time of UsIs the total utilization of the aperiodic task.
6. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the deadline for updating the main subtask is as follows:
Figure FDA00025520628600000218
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure FDA00025520628600000219
for a preceding non-periodic task Jk-1Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure FDA00025520628600000220
is a main subtask
Figure FDA00025520628600000221
Actual execution time of, UsIs the total utilization of the aperiodic task.
7. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the distribution deadline for the standby subtasks is as follows:
Figure FDA0002552062860000031
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure FDA0002552062860000032
the deadline for updating after the execution of the main subtask is finished,
Figure FDA0002552062860000033
being the actual completion time of the main sub-task,
Figure FDA0002552062860000034
for the worst execution time of the standby subtask, UsIs the total utilization of the aperiodic task.
8. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the deadline of the updating backup subtask is as follows:
Figure FDA0002552062860000035
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure FDA0002552062860000036
the deadline for updating after the execution of the main subtask is finished,
Figure FDA0002552062860000037
being the actual completion time of the main sub-task,
Figure FDA0002552062860000038
for preparing the actual execution time of the subtask, UsIs the total utilization of the aperiodic task.
9. The adaptive scheduling method for the hybrid task of the real-time system according to claim 1, wherein: the deadline of the updating non-periodic task is as follows:
Figure FDA0002552062860000039
wherein r iskFor non-periodic task JkThe time of arrival of the time-of-arrival,
Figure FDA00025520628600000310
for a preceding non-periodic task Jk-1Deadline for update after execution, fk-1For a preceding non-periodic task Jk-1The actual time of completion of the process,
Figure FDA00025520628600000311
for non-periodic task JkActual execution time of, UsIs the total utilization of the aperiodic task.
CN201710610065.1A 2017-07-25 2017-07-25 Self-adaptive scheduling method suitable for real-time system mixed task Active CN109298917B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710610065.1A CN109298917B (en) 2017-07-25 2017-07-25 Self-adaptive scheduling method suitable for real-time system mixed task

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710610065.1A CN109298917B (en) 2017-07-25 2017-07-25 Self-adaptive scheduling method suitable for real-time system mixed task

Publications (2)

Publication Number Publication Date
CN109298917A CN109298917A (en) 2019-02-01
CN109298917B true CN109298917B (en) 2020-10-30

Family

ID=65167753

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710610065.1A Active CN109298917B (en) 2017-07-25 2017-07-25 Self-adaptive scheduling method suitable for real-time system mixed task

Country Status (1)

Country Link
CN (1) CN109298917B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110532103A (en) * 2019-09-09 2019-12-03 北京西山居互动娱乐科技有限公司 A kind of method and device of multitasking
US11579959B2 (en) 2021-05-26 2023-02-14 Honeywell International Inc. Systems and methods for margin based diagnostic tools for priority preemptive schedulers

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5349656A (en) * 1990-11-28 1994-09-20 Hitachi, Ltd. Task scheduling method in a multiprocessor system where task selection is determined by processor identification and evaluation information
EP0753813A1 (en) * 1995-07-11 1997-01-15 Cegelec Method for scheduling aperiodic tasks
US6567840B1 (en) * 1999-05-14 2003-05-20 Honeywell Inc. Task scheduling and message passing
CN102999385A (en) * 2012-11-06 2013-03-27 苏州懿源宏达知识产权代理有限公司 Multiprocessor co-processing method in computation equipment
CN103365710A (en) * 2012-03-31 2013-10-23 索尼公司 Real-time task scheduling device and method and computer system
CN103384206A (en) * 2012-05-02 2013-11-06 中国科学院计算机网络信息中心 Concurrent processing method and system for mass data
CN105260818A (en) * 2015-09-17 2016-01-20 福州大学 Online optimized scheduling method for workflow groups with deadline constraint in mixed cloud environment
CN106155781A (en) * 2015-04-08 2016-11-23 中国科学院软件研究所 A kind of real-time task scheduling method in multiple agent platform
WO2017028696A1 (en) * 2015-08-17 2017-02-23 阿里巴巴集团控股有限公司 Method and device for monitoring load of distributed storage system

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7302685B2 (en) * 2000-06-02 2007-11-27 Honeywell International Inc. Methods and apparatus for sharing slack in a time-partitioned system
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
CN103491174A (en) * 2013-09-26 2014-01-01 中国船舶重工集团公司第七一六研究所 Periodic/aperiodic mixed real-time task scheduling method based on delay servers
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
CN106878132B (en) * 2017-02-16 2020-01-07 北京中航通用科技有限公司 Bus scheduling method based on period and priority

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5349656A (en) * 1990-11-28 1994-09-20 Hitachi, Ltd. Task scheduling method in a multiprocessor system where task selection is determined by processor identification and evaluation information
EP0753813A1 (en) * 1995-07-11 1997-01-15 Cegelec Method for scheduling aperiodic tasks
US6567840B1 (en) * 1999-05-14 2003-05-20 Honeywell Inc. Task scheduling and message passing
CN103365710A (en) * 2012-03-31 2013-10-23 索尼公司 Real-time task scheduling device and method and computer system
CN103384206A (en) * 2012-05-02 2013-11-06 中国科学院计算机网络信息中心 Concurrent processing method and system for mass data
CN102999385A (en) * 2012-11-06 2013-03-27 苏州懿源宏达知识产权代理有限公司 Multiprocessor co-processing method in computation equipment
CN106155781A (en) * 2015-04-08 2016-11-23 中国科学院软件研究所 A kind of real-time task scheduling method in multiple agent platform
WO2017028696A1 (en) * 2015-08-17 2017-02-23 阿里巴巴集团控股有限公司 Method and device for monitoring load of distributed storage system
CN105260818A (en) * 2015-09-17 2016-01-20 福州大学 Online optimized scheduling method for workflow groups with deadline constraint in mixed cloud environment

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Adaptive Total Bandwidth Server: Using Predictive Execution Time;Kiyofumi Tanaka ET AL;《International Embedded Systems Symposium 2013》;20131231;第250-256页 *
一种混合实时任务***的公平调度算法;张惠娟;《西安电子科技大学学报（自然科学版）》;20040430;第31卷(第2期);全文 *

Also Published As

Publication number Publication date
CN109298917A (en) 2019-02-01

Similar Documents

Publication Publication Date Title
JP6799947B2 (en) Systems and methods for scheduling tasks using the sliding time window
US8713571B2 (en) Asynchronous task execution
US9990229B2 (en) Scheduling method and apparatus for applying laxity correction based on task completion proportion and preset time
CN104951372B (en) A kind of Map/Reduce data processing platform (DPP) memory source dynamic allocation methods based on prediction
US8607240B2 (en) Integration of dissimilar job types into an earliest deadline first (EDF) schedule
CN103324525B (en) Method for scheduling task under a kind of cloud computing environment
US8954975B2 (en) Task scheduling method for real time operating system
CN104765640B (en) A kind of intelligent Service dispatching method
US8875146B2 (en) Systems and methods for bounding processing times on multiple processing units
CN111104211A (en) Task dependency based computation offload method, system, device and medium
CN109298920B (en) Mixed key task scheduling method based on quasi-partition thought
CN105389204B (en) A kind of more resource partial order dispatching methods
CN109298917B (en) Self-adaptive scheduling method suitable for real-time system mixed task
JP2011180894A (en) Job scheduling program, device, and method
CN111176637B (en) Schedulability analysis method of AADL model based on cache preemption delay constraint
US20050182747A1 (en) Method and system for executing multiple tasks at adaptively controlled resource utilization rates to achieve equal QoS levels
CN110175073B (en) Scheduling method, sending method, device and related equipment of data exchange job
KR101335038B1 (en) Periodic and aperiodic task scheduling algorithm based on topological sort and residual time
CN108874517B (en) Method for optimizing utilization rate division energy consumption of standby system with fixed priority
CN108845870B (en) Probabilistic real-time task scheduling method based on pWCET shaping
KR20120047352A (en) Method of updating software in real-time system and apparatus for the same
US9152451B2 (en) Method of distributing processor loading between real-time processor threads
Tanaka Virtual release advancing for earlier deadlines
Azhen et al. Adaptive real-time scheduling for mixed task sets based on total bandwidth server
CN108268310B (en) Method and device for determining minimum scheduling granularity

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: No.16, Nanping East Road, Dongling District, Shenyang City, Liaoning Province, 110168

Patentee after: Shenyang Zhongke CNC Technology Co.,Ltd.

Address before: No.16, Nanping East Road, Dongling District, Shenyang City, Liaoning Province, 110168

Patentee before: Shenyang Golding Nc Intelligence Tech.co.,ltd.