CN109739332A - A kind of general energy consumption optimization method of multitask - Google Patents
A kind of general energy consumption optimization method of multitask Download PDFInfo
- Publication number
- CN109739332A CN109739332A CN201910073576.3A CN201910073576A CN109739332A CN 109739332 A CN109739332 A CN 109739332A CN 201910073576 A CN201910073576 A CN 201910073576A CN 109739332 A CN109739332 A CN 109739332A
- Authority
- CN
- China
- Prior art keywords
- server
- task
- state
- time
- processor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a kind of general energy consumption optimization methods of multitask, comprising the following steps: establishes n server model;Determine server state transformation rule;Determine that server parameter updates rule;According to earliest-deadline-first grade strategy dispatch server;The execution speed S of calculating task;Once processor free time is more than processor state handover overhead to, low power consumpting state is switched the processor into, until there is new task release.Method of the invention does not need any information for knowing task in advance, can dispatch any kind of task, by updating server utilization and conversion processor state, achievees the purpose that reduce system energy consumption.
Description
Technical field
The present invention relates to embedded system field energy optimization dispatching method, in particular to a kind of general energy consumption of multitask is excellent
Change method.
Background technique
Embedded system is widely used in industries such as aerospace, Industry Control, electric power, manufacturing industry, different applications,
Cause the task type of system different.But these tasks can substantially be divided into the periodic duty of deadline limitation and have response
The aperiodic task of time requirement;Aperiodic task, which can be further divided into two neighboring task instances release time interval, to be had
Limit and have the accidental task of deadline demand with there is no limit aperiodic task.Regardless of the task type of embedded system
How, real-time and low energy consumption are all the targets for designing embedded system.
Most of existing energy optimization algorithm is all only to be applicable in the embedded system of single task role type or more than one
The embedded system of task type needs to design the scheduling problem that polyalgorithm solves different type task.In addition, existing energy
Consumption optimization algorithm will often obtain in advance executes the information such as time under the type of task, period, worst case.
Summary of the invention
It is a primary object of the present invention to overcome drawbacks described above in the prior art, propose that a kind of general energy consumption of multitask is excellent
Change method, this method utilize server scheduling task, and execute speed according to what the arrival of task and server state determined task
Degree, to reduce system energy consumption.
The present invention adopts the following technical scheme:
A kind of general energy consumption optimization method of multitask characterized by comprising
Establish n server model;
Determine server state transformation rule;
Determine that server parameter updates rule;
According to earliest-deadline-first grade strategy dispatch server;
The execution speed S of calculating task;
Once processor free time is more than processor state handover overhead to, low power consumpting state is switched the processor into,
Until there is new task release;
The earliest-deadline-first grade strategy includes: that the deadline of server is smaller, and priority is higher, service
The deadline of device is bigger, and priority is lower;When the deadline of server is identical, time for being activated according to server
Determine priority, the time that is activated is closer, and priority is high, and the time that is activated is remoter, and priority is lower;When server quilt
When activationary time is identical, the small priority of server subscript is high, and server subscript is big, and priority is low;The high service of priority
Device is by priority scheduling.
It is described to establish n server model;Include:
System is made of n server, this n server SE1,SE2,…,SEnIt indicates;Any server SEi1≤i
≤ n, i are positive integer, including triple (Ui,Pi,Di), wherein UiIt is server S EiUtilization rate, PiIt is server S EiWeek
Phase, DiIt is server S EiDeadline;Each server S EiA generic task can be dispatched, this generic task can be the period times
Business, accidental task or aperiodic task.
Determine server state transformation rule, comprising:
Each server includes three states: active state, an inactive state or suspended state;Server is in when initial
An inactive state;In moment t, when there is task dispatching pending, server becomes suspended state from an inactive state;In addition,
All tasks before moment t are all completed to execute, and the processor budget for distributing to it does not exhaust, and server is still located at this time
In suspended state;In moment t, without waiting for the task of execution, and processor budget exhausts, and server enters an inactive state;
Once there is task to start to execute, server enters active state.
Determine that server parameter updates rule, comprising:
Server S EiPass through virtual time ViIts deadline is calculated with its period;V is set when beginningi=0 and Di=0;
As server S EiIn an inactive state, and task instancesAt the momentWhen arrival, V is updatediWith Di;It takes at this time
Be engaged in device SEiInto suspended state;
As server S EiIn active state, and complete task instancesExecution, at this time if there is new task
It reaches, server still keeps active state, updates ViWith Di;If not new task schedule, server S EiInto hang-up
State;
As server S EiVirtual time ViGreater than the current time t of systemcWhen, server S EiInto an inactive state;
As server S EiIn suspended state and task instancesIt reaches, updates its Di;Server, which enters, at this time enlivens shape
State;
When processor is in idle condition, all servers enter an inactive state.
The execution speed S of calculating task, comprising:
As server S EiIn an inactive state, and task instancesWhen arrival, the execution speed S=S+U of taski,
The initial value of middle S is set as 0;
As server S EiIn suspended state and virtual time ViEqual to the current time t of systemcOr processor budget
When exhausting, the execution speed S=S-U of taski。
Once processor free time is more than processor state handover overhead to, low power consumpting state is switched the processor into,
Until there is new task release, comprising:
Processor state handover overhead toIt is calculated by following formula:
to=max { To,Bo}
Wherein, ToIt is the time overhead of processor state conversion, BoIt is the time of processor energy-consuming balance.
By the above-mentioned description of this invention it is found that compared with prior art, the invention has the following beneficial effects:
(1) any information for knowing task in advance can not had to, be suitable for any embedded system;
(2) solving a kind of dispatching method, to be only applicable in a type of task insufficient, can dispatch a plurality of types of simultaneously
Business;
(3) energy consumption that about 19.43% is saved compared with the method for not using power-saving technology, can reduce the life of product
Produce cost.
Detailed description of the invention
Fig. 1 is the flow chart schematic diagram of the method for the present invention.
Specific embodiment
Below by way of specific embodiment, the invention will be further described.
Referring to Fig. 1, a kind of general energy consumption optimization method of multitask of the invention comprising following steps:
Step 101: establishing n server model;
System is made of n server, this n server SE1,SE2,…,SEnIt indicates;Any server SEi(1≤i
≤ n, i are positive integer) by triple (Ui,Pi,Di), wherein UiIt is server S EiUtilization rate, PiIt is server S EiPeriod,
DiIt is server S EiDeadline;Each server S EiA generic task can be dispatched, this generic task can be periodic duty,
Accidental task, aperiodic task.
Step 102: determining server state transformation rule;
Each server includes three states: active state, an inactive state, suspended state;So-called active state is
Refer to that server is carrying out task;So-called an inactive state is that finger processor is in idle condition or waits holding without task dispatching
Capable or processor budget exhausts;So-called suspended state, which refers to, has task dispatching pending or server has completed task
Execution but its processor budget do not exhaust;Server is in an inactive state when initial;In moment t, when there is task dispatching to wait for
When execution, server becomes suspended state from an inactive state;In addition, all tasks before moment t are all completed to execute, and
The processor budget for distributing to it does not exhaust, and server is still in suspended state at this time;In moment t, without waiting for execution
Task, and processor budget exhausts, and server enters an inactive state;Once there is task to start to execute, server enters work
Jump state.
Step 103: determining that server parameter updates rule;
Server S EiPass through virtual time ViIts deadline is calculated with its period;V is set when beginningi=0 and Di=0;
As server S EiIn an inactive state, and task instancesAt the momentWhen arrival, V is updatediWith Di, take at this time
Be engaged in device SEiInto suspended state;ViAnd DiIt is calculated respectively by following formula:
Wherein,It is task instancesArrival time, PiIt is server S EiPeriod;
As server S EiIn active state, task instances are completedExecution, at this time if there is new taskIt arrives
It reaches, server still keeps active state, updates Di;DiIt is calculated by following formula:
Di=Vi+Pi;
If not new task schedule, server S EiInto suspended state;
As server S EiVirtual time ViGreater than the current time t of systemcWhen, server S EiInto an inactive state;
As server S EiIn suspended state and task instancesWhen arrival, j is the positive integer greater than 1, updates its Di, Di
It is calculated by following formula:
Di=Vi+Pi;
Server enters active state at this time;
When processor is in idle condition, all servers enter an inactive state.
Step 104: according to earliest-deadline-first grade strategy dispatch server;
The priority of server is determined by its deadline, and the deadline of server updates rule by the parameter of server
Then determine;The deadline of server is smaller, and priority is higher;The deadline of server is bigger, and priority is lower;When
When the deadline of server is identical, priority is determined according to the time that server is activated;Time that is activated is closer, excellent
First grade is high;Being activated, the time is remoter, and priority is lower;It is active that the time that is activated of so-called server refers to that server enters
At the time of state;When server be activated the time it is identical when, the small priority of server subscript is high, and server subscript is big, excellent
First grade is low;The high server of priority is by priority scheduling.
Step 105: the execution speed S of calculating task;
As server S EiIn an inactive state, and task instancesWhen arrival, the execution speed S=S+U of taski,
The initial value of middle S is set as 0;
As server S EiIn suspended state and virtual time ViEqual to the current time t of systemcOr processor budget
When exhausting, the execution speed S=S-U of taski。
Step 106: once processor free time is more than processor state handover overhead to, switch the processor into low function
Consumption state, until there is new task release;
Once processor free time is more than processor state handover overhead to, low power consumpting state is switched the processor into,
Until there is new task release;Processor state handover overhead toIt is calculated by following formula:
to=max { To,Bo}
Wherein, ToIt is the time overhead of processor state conversion, BoIt is the time of processor energy-consuming balance, i.e., when idle
Between be not above BoWhen, low power consumpting state is switched the processor into, not only cannot be energy saving, increase energy consumption instead;Only processor
Free time is more than BoWhen, switching the processor into low power consumpting state could be energy saving, BoIt is calculated by following formula:
Wherein, EoIt is the energy consumption expense of processor state conversion, PaAnd PsIt is that processor is in active mode and suspend mode respectively
The power consumption of mode.
In the present embodiment, the system for there are 4 tasks is considered, wherein task T1With task T2It is accidental task, task T3With appoint
Be engaged in T4It is periodic duty.Accidental task T1With accidental task T2Minimum release interval be respectively 8 and 16.Accidental task T1Hold
The row time between 1~2, accidental task T2The execution time between 2~4.Accidental task T1First example release time
It is 0, executing the time is 1, and the release time of second example is 10, and executing the time is 2;Accidental task T2First reality
Example release time is 0, and executing the time is 2, and the release time of second example is 18, and executing the time is 4.Periodic duty T3
With periodic duty T4Period be respectively 4 and 32.Periodic duty T3And cycle T4The execution time be respectively 1 and 8.Periodic duty
T3With periodic duty T4First example also 0 moment discharge.
This 4 tasks are dispatched using the method for the present invention in section [0,32].It is corresponding with this four tasks, four clothes are set
Be engaged in device SE1,SE2,SE3,SE4Dispatch this 4 tasks.The utilization rate U of this four servers1,U2,U3,U40.25 is respectively set to,
0.25,0.25,0.25;The period P of this four servers1,P2,P3,P4It is respectively set to 8,16,4,32.At 0 moment, server
SE1,SE2,SE3,SE4Deadline be respectively 8,16,4,32;And servers all at this time all enters active state, at this time
Execution speed S=1.Therefore, periodic duty T3Start to execute speed S=1 execution, and complete to execute at the moment 1, server
SE3Into suspended state;Moment 1, accidental task T1Start to execute speed S=1 execution, completes to execute at the moment 2;It takes at this time
Be engaged in device SE1Into suspended state;Moment 2, accidental task T2Start to execute speed S=1 execution, completes to execute at the moment 4, clothes
Be engaged in device SE2Into suspended state;At the moment 4, SE1Into an inactive state, execution speed S=0.75 at this time, and at the moment 4,
Periodic duty T3Example reach, server S E3Deadline be 8 and enter active state;So at the moment 4, periodic duty
T3To execute speed S=0.75 execution, and complete to execute at the moment 5.33.At the moment 5.33, periodic duty T4To execute speed S
=0.75 executes, at the moment 8, periodic duty T3Example reach, server S E3Deadline be 12 and enter active state.
Server S E at this time2Into an inactive state, execution speed at this time is S=0.5.Periodic duty T3To execute speed S=0.5
Start to execute, completes to execute at the moment 10, server S E3Into suspended state.At the moment 10, accidental task T1Second reality
Example reaches, server S E1Into active state, and its deadline is 18, execution speed S=0.75 at this time.At the moment 12,
Periodic duty T3Example reach, server S E3Deadline be 16 and enter active state;Periodic duty T3Start with S=
0.75 executes, and completes to execute at the moment 13.33, server S E3Into suspended state.At the moment 13.33, accidental task T1After
It is continuous to be executed with S=0.75, and complete to execute at the moment 14, and server S E1Into suspended state.At the moment 14, periodic duty T4
To execute speed S=0.75 execution.At the moment 16, periodic duty T3Example reach, server S E3Deadline be 20 and
Into active state.Execution speed S=0.75 at this time.Periodic duty T3To execute speed S=0.75 execution, at the moment
17.33 complete to execute, server S E3Into suspended state.Moment 17.33, periodic duty T4To execute speed S=0.75 execution.
At the moment 18, accidental task T2Second example reach, server S E2Into active state, and its deadline is 34;This
When execution speed S=1.Therefore, in moment 18, periodic duty T4Continue to execute speed S=1 execution.In moment 20, period
Task T3Example reach, server S E3Deadline be 24 and enter active state.Periodic duty T3To execute speed S=
1 executes, and completes to execute at the moment 21, server S E3Into suspended state.At the moment 21, periodic duty T4Continue to execute speed
S=1 is executed, and it completes to execute at the moment 22.95, server S E4Into suspended state.At the moment 22.95, accidental task T2
To execute speed S=1 execution.At the moment 24, periodic duty T3Example reach, server S E3Deadline be 28 and enter
Active state.At the moment 25, periodic duty T3It completes to execute, and server S E3Into suspended state.At the moment 25, accidental task
T2It is continued to execute with executing speed S=1, and completes to execute at the moment 27.95, server S E2Into suspended state.At the moment 28,
Periodic duty T3Example reach, server S E3Deadline be 32 and enter active state.At the moment 29, periodic duty T3
It completes to execute, and server S E3Into suspended state.
Even if by it is found that the method for the present invention than not using the other methods of power-saving technology to save about 19.43%
Energy consumption.
The above is only a specific embodiment of the present invention, but the design concept of the present invention is not limited to this, all to utilize this
Design makes a non-material change to the present invention, and should all belong to behavior that violates the scope of protection of the present invention.
Claims (6)
1. a kind of general energy consumption optimization method of multitask characterized by comprising
Establish n server model;
Determine server state transformation rule;
Determine that server parameter updates rule;
According to earliest-deadline-first grade strategy dispatch server;
The execution speed S of calculating task;
Once processor free time is more than processor state handover overhead to, switch the processor into low power consumpting state, Zhi Daoyou
New task release;
The earliest-deadline-first grade strategy includes: that the deadline of server is smaller, and priority is higher, server
Deadline is bigger, and priority is lower;When the deadline of server is identical, determined according to the time that server is activated
Priority, the time that is activated is closer, and priority is high, and the time that is activated is remoter, and priority is lower;When server is activated
When time is identical, the small priority of server subscript is high, and server subscript is big, and priority is low;The high server quilt of priority
Priority scheduling.
2. the general energy consumption optimization method of multitask according to claim 1, which is characterized in that described to establish n server
Model;Include:
System is made of n server, this n server SE1,SE2,…,SEnIt indicates;Any server SEi1≤i≤n,i
For positive integer, including triple (Ui,Pi,Di), wherein UiIt is server S EiUtilization rate, PiIt is server S EiPeriod, DiIt is
Server S EiDeadline;Each server S EiA generic task can be dispatched, this generic task can be periodic duty, accidental
Task or aperiodic task.
3. the general energy consumption optimization method of multitask according to claim 2, which is characterized in that determine that server state is converted
Rule, comprising:
Each server includes three states: active state, an inactive state or suspended state;Server is in non-live when initial
Jump state;In moment t, when there is task dispatching pending, server becomes suspended state from an inactive state;In addition, in moment t
All tasks before are all completed to execute, and the processor budget for distributing to it does not exhaust, and server is still in extension at this time
The state of rising;In moment t, without waiting for the task of execution, and processor budget exhausts, and server enters an inactive state;Once
There is task to start to execute, server enters active state.
4. the general energy consumption optimization method of multitask according to claim 3, which is characterized in that determine that server parameter updates
Rule, comprising:
Server S EiPass through virtual time ViIts deadline is calculated with its period;V is set when beginningi=0 and Di=0;
As server S EiIn an inactive state, and task instances JiJ is at the momentWhen arrival, V is updatediWith DiJ is greater than 1
Positive integer;Server S E at this timeiInto suspended state;
As server S EiIn active state, and complete task instancesExecution, at this time if there is new taskIt reaches,
Server still keeps active state, updates ViWith Di;If not new task schedule, server S EiInto suspended state;
As server S EiVirtual time ViGreater than the current time t of systemcWhen, server S EiInto an inactive state;
As server S EiIn suspended state and task instancesIt reaches, updates its Di;Server enters active state at this time;
When processor is in idle condition, all servers enter an inactive state.
5. the general energy consumption optimization method of multitask according to claim 4, which is characterized in that the execution speed of calculating task
S, comprising:
As server S EiIn an inactive state, and task instancesWhen arrival, the execution speed S=S+U of taski, wherein S
Initial value is set as 0;
As server S EiIn suspended state and virtual time ViEqual to the current time t of systemcOr processor budget exhausts
When, the execution speed S=S-U of taski。
6. the general energy consumption optimization method of multitask according to claim 1, which is characterized in that once processor free time
More than processor state handover overhead to, low power consumpting state is switched the processor into, until there is new task release, comprising:
Processor state handover overhead toIt is calculated by following formula:
to=max { To,Bo}
Wherein, ToIt is the time overhead of processor state conversion, BoIt is the time of processor energy-consuming balance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910073576.3A CN109739332B (en) | 2019-01-25 | 2019-01-25 | Multi-task general energy consumption optimization method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910073576.3A CN109739332B (en) | 2019-01-25 | 2019-01-25 | Multi-task general energy consumption optimization method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109739332A true CN109739332A (en) | 2019-05-10 |
CN109739332B CN109739332B (en) | 2022-05-03 |
Family
ID=66366106
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910073576.3A Active CN109739332B (en) | 2019-01-25 | 2019-01-25 | Multi-task general energy consumption optimization method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109739332B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110825504A (en) * | 2019-10-24 | 2020-02-21 | 华侨大学 | Server-based hybrid key accidental task energy consumption optimization method |
CN112235859A (en) * | 2020-09-22 | 2021-01-15 | 国家卫星气象中心(国家空间天气监测预警中心) | Dynamic energy consumption control method based on multi-target constraint |
CN112633589A (en) * | 2020-12-30 | 2021-04-09 | 华侨大学 | Probability model-based hybrid key task energy consumption optimization scheduling method |
CN113821339A (en) * | 2021-08-20 | 2021-12-21 | 广州云硕科技发展有限公司 | Energy consumption monitoring method and device for IDC data center machine room |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7194385B2 (en) * | 2002-11-12 | 2007-03-20 | Arm Limited | Performance level setting of a data processing system |
US8112644B2 (en) * | 2006-12-15 | 2012-02-07 | Institute For Information Industry | Dynamic voltage scaling scheduling mechanism for sporadic, hard real-time tasks with resource sharing |
CN105630126A (en) * | 2014-11-05 | 2016-06-01 | 中国科学院沈阳计算技术研究所有限公司 | Low-power scheduling method for mixed task based on constant bandwidth server |
CN106970835A (en) * | 2017-03-20 | 2017-07-21 | 华侨大学 | Fixed priority resource limited system level energy consumption optimization method |
CN105893148B (en) * | 2016-03-30 | 2019-01-22 | 华侨大学 | A kind of accidental task low energy consumption dispatching method based on RM strategy |
-
2019
- 2019-01-25 CN CN201910073576.3A patent/CN109739332B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7194385B2 (en) * | 2002-11-12 | 2007-03-20 | Arm Limited | Performance level setting of a data processing system |
US8112644B2 (en) * | 2006-12-15 | 2012-02-07 | Institute For Information Industry | Dynamic voltage scaling scheduling mechanism for sporadic, hard real-time tasks with resource sharing |
CN105630126A (en) * | 2014-11-05 | 2016-06-01 | 中国科学院沈阳计算技术研究所有限公司 | Low-power scheduling method for mixed task based on constant bandwidth server |
CN105893148B (en) * | 2016-03-30 | 2019-01-22 | 华侨大学 | A kind of accidental task low energy consumption dispatching method based on RM strategy |
CN106970835A (en) * | 2017-03-20 | 2017-07-21 | 华侨大学 | Fixed priority resource limited system level energy consumption optimization method |
Non-Patent Citations (2)
Title |
---|
张忆文: "硬实时系统周期任务低功耗调度算法", 《西安交通大学学报》 * |
张忆文等: "面向硬实时系统零星任务低调度算法", 《小型微型计算机系统》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110825504A (en) * | 2019-10-24 | 2020-02-21 | 华侨大学 | Server-based hybrid key accidental task energy consumption optimization method |
CN110825504B (en) * | 2019-10-24 | 2023-03-07 | 华侨大学 | Server-based hybrid key accidental task energy consumption optimization method |
CN112235859A (en) * | 2020-09-22 | 2021-01-15 | 国家卫星气象中心(国家空间天气监测预警中心) | Dynamic energy consumption control method based on multi-target constraint |
CN112235859B (en) * | 2020-09-22 | 2022-08-05 | 国家卫星气象中心(国家空间天气监测预警中心) | Dynamic energy consumption control method based on multi-target constraint |
CN112633589A (en) * | 2020-12-30 | 2021-04-09 | 华侨大学 | Probability model-based hybrid key task energy consumption optimization scheduling method |
CN112633589B (en) * | 2020-12-30 | 2022-07-29 | 华侨大学 | Probability model-based hybrid key task energy consumption optimization scheduling method |
CN113821339A (en) * | 2021-08-20 | 2021-12-21 | 广州云硕科技发展有限公司 | Energy consumption monitoring method and device for IDC data center machine room |
Also Published As
Publication number | Publication date |
---|---|
CN109739332B (en) | 2022-05-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109739332A (en) | A kind of general energy consumption optimization method of multitask | |
CN101339521B (en) | Tasks priority dynamic dispatching algorithm | |
CN104536827B (en) | A kind of data dispatching method and device | |
CN102364447B (en) | Operation scheduling method for optimizing communication energy consumption among multiple tasks | |
CN101923487A (en) | Comprehensive embedded type real-time period task scheduling method | |
CN108984292A (en) | Mix critical system fixed priority periodic duty energy consumption optimization method | |
CN106445070B (en) | Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks | |
CN109324880A (en) | A kind of low-power consumption scheduling method suitable for real-time system periodic task model | |
CN105117284A (en) | Scheduling method for worker thread based on priority proportion queue | |
CN105117283A (en) | Task splitting method and system | |
CN109597378B (en) | Resource-limited hybrid task energy consumption sensing method | |
CN103810026A (en) | Mixing scheduling method suitable for real-time system periodic tasks | |
CN110308977A (en) | A kind of crucial accidental task low energy consumption method of dynamic utilization rate update mixing | |
CN103914346A (en) | Group-based dual-priority task scheduling and energy saving method for real-time operating system | |
CN107391244A (en) | A kind of Internet of Things operating system dispatching method based on mixed scheduling model | |
CN113535356B (en) | Energy-aware hierarchical task scheduling method and device | |
CN106802825B (en) | A kind of dynamic task scheduling method and system based on real-time system | |
CN109324891A (en) | A kind of periodic duty low-power consumption scheduling method of ratio free time distribution | |
CN105630126B (en) | One kind is based on normal bandwidth server hybrid task low-power consumption scheduling method | |
CN112130992A (en) | Low-power-consumption scheduling method based on high-performance open type numerical control system | |
CN102654843A (en) | Non-preemptive type fault-tolerant scheduling method in embedded processor and embedded processor | |
CN106648834B (en) | Virtual machine scheduling method based on batch packaging problem | |
CN108874517A (en) | The stand-by system availability of fixed priority divides energy consumption optimization method | |
CN109298917A (en) | A kind of self-adapting dispatching method suitable for real-time system hybrid task | |
CN105706022B (en) | A kind of method, processing unit and the terminal device of prediction processor utilization rate |
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 |