CN109739332B - Multi-task general energy consumption optimization method - Google Patents

Multi-task general energy consumption optimization method Download PDF

Info

Publication number
CN109739332B
CN109739332B CN201910073576.3A CN201910073576A CN109739332B CN 109739332 B CN109739332 B CN 109739332B CN 201910073576 A CN201910073576 A CN 201910073576A CN 109739332 B CN109739332 B CN 109739332B
Authority
CN
China
Prior art keywords
server
task
time
state
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.)
Active
Application number
CN201910073576.3A
Other languages
Chinese (zh)
Other versions
CN109739332A (en
Inventor
张忆文
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.)
Huaqiao University
Original Assignee
Huaqiao University
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 Huaqiao University filed Critical Huaqiao University
Priority to CN201910073576.3A priority Critical patent/CN109739332B/en
Publication of CN109739332A publication Critical patent/CN109739332A/en
Application granted granted Critical
Publication of CN109739332B publication Critical patent/CN109739332B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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

Landscapes

  • Power Sources (AREA)

Abstract

The invention discloses a multitask general energy consumption optimization method, which comprises the following steps of establishing n server models; determining a server state transition rule; determining a server parameter updating rule; scheduling the server according to the earliest deadline priority policy; calculating the execution speed S of the task; processor state switching overhead t once processor idle time exceedsoAnd switching the processor to a low power consumption state until a new task is released. The method of the invention can schedule any type of tasks without knowing any information of the tasks in advance, and achieves the purpose of reducing the energy consumption of the system by updating the utilization rate of the server and converting the state of the processor.

Description

Multi-task general energy consumption optimization method
Technical Field
The invention relates to an energy consumption optimization scheduling method in the field of embedded systems, in particular to a multi-task general energy consumption optimization method.
Background
The embedded system is widely used in the industries of aerospace, industrial control, electric power, manufacturing industry and the like, and different application causes different task types of the system. But these tasks can be roughly divided into periodic tasks with deadline limits and non-periodic tasks with response time requirements; the non-periodic task can be further divided into an occasional task with limitation on the release time interval of two adjacent task instances and deadline requirements and a non-limited non-periodic task. Regardless of the task type of the embedded system, real-time and low power consumption are goals for designing the embedded system.
Most of the existing energy consumption optimization algorithms are only suitable for embedded systems of a single task type, or one embedded system of a multi-task type needs to design a plurality of algorithms to solve the scheduling problem of tasks of different types. In addition, the existing energy consumption optimization algorithm needs to acquire information such as the type, the period, the worst execution time and the like of a task in advance.
Disclosure of Invention
The main purpose of the present invention is to overcome the above mentioned drawbacks in the prior art, and to provide a method for optimizing general energy consumption for multiple tasks, which uses a server to schedule tasks and determines the execution speed of the tasks according to the arrival of the tasks and the state of the server, so as to reduce the energy consumption of the system.
The invention adopts the following technical scheme:
a multitask general energy consumption optimization method is characterized by comprising the following steps:
establishing n server models;
determining a server state transition rule;
determining a server parameter updating rule;
scheduling the server according to the earliest deadline priority policy;
calculating the execution speed S of the task;
processor state switching overhead t once processor idle time exceedsoSwitching the processor to a low power consumption state until a new task is released;
the earliest deadline priority policy comprises: the smaller the deadline of the server is, the higher the priority of the server is, and the larger the deadline of the server is, the lower the priority of the server is; when the deadline of the server is the same, determining the priority according to the activated time of the server, wherein the closer the activated time is, the higher the priority is, and the farther the activated time is, the lower the priority is; when the activated time of the server is the same, the priority of the server with small subscript is high, the priority of the server with large subscript is low; the server with the higher priority is scheduled preferentially.
The n server models are established; the method comprises the following steps:
the system consists of n servers, which use SE1,SE2,…,SEnRepresents; any server SEiI is more than or equal to 1 and less than or equal to n, i is a positive integer and comprises a triplet (U)i,Pi,Di) Wherein U isiIs a server SEiUtilization ratio of (P)iIs a server SEiPeriod of (D)iIs a server SEiThe deadline of (2); each server SEiA type of task may be scheduled, which may be a periodic task, an occasional task, or an aperiodic task.
Determining a server state transition rule, comprising:
each server contains three states: an active state, an inactive state, or a suspended state; initially, the server is in an inactive state; at time t, when a task is waiting to be executed, the server changes from an inactive state to a suspended state; furthermore, all tasks before time t are finished executing and the processor budget allocated to it is not exhausted, while the server is still in a suspended state; at time t, no task waiting to be executed and the processor budget is exhausted, the server enters an inactive state; once a task begins execution, the server enters an active state.
Determining a server parameter update rule, comprising:
server SEiBy virtual time ViCalculating the deadline of the period of the time; set V at the beginningi0 and Di＝0；
When server SEiIn an inactive state, and task instance
Figure BDA0001958029690000021
At the moment of time
Figure BDA0001958029690000022
When it arrives, update ViAnd Di(ii) a Server SE at this timeiEntering a suspended state;
when server SEiIs in active state and completes task instance
Figure BDA0001958029690000023
When there is a new task
Figure BDA0001958029690000024
When the server is in the active state, updating ViAnd Di(ii) a If there is no new task schedule, server SEiEntering a suspended state;
when server SEiVirtual time V ofiGreater than the current time t of the systemcTime, server SEiEntering an inactive state;
when server SEiIn suspended state and task instance
Figure BDA0001958029690000025
Arrives and updates its Di(ii) a At this time, the server enters an active state;
when the processor is in the idle state, all servers enter an inactive state.
Calculating the execution speed S of the task, comprising:
when server SEiIn an inactive state, and task instance
Figure BDA0001958029690000026
When the task arrives, the execution speed of the task is S + UiWherein the initial value of S is set to 0;
when server SEiIn suspended state and virtual time ViEqual to the current time t of the systemcOr when the processor budget is exhausted, the execution speed S of the task is S-Ui。
Processor state switching overhead t once processor idle time exceedsoSwitching the processor to a low power consumption state until a new task is released, comprising:
processor state switching overhead toCalculated from the following formula:
to＝max{To,Bo}
wherein, ToIs the time overhead of the processor state transition, BoIs the time for the processor to power balance.
As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages:
(1) any information of the task is not needed to be known in advance, and the method is suitable for any embedded system;
(2) the problem that one scheduling method is only suitable for one type of tasks is solved, and various types of tasks can be scheduled at the same time;
(3) compared with the method without the energy-saving technology, the method saves the energy consumption by about 19.43 percent and can reduce the production cost of the product.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
Detailed Description
The invention is further described below by means of specific embodiments.
Referring to fig. 1, the method for optimizing multitask general energy consumption of the present invention includes the following steps:
step 101: establishing n server models;
the system consists of n servers, which use SE1,SE2,…,SEnRepresents; any server SEi(1. ltoreq. i. ltoreq. n, i is a positive integer) from a triplet (U)i,Pi,Di) Wherein U isiIs a server SEiUtilization ratio of (P)iIs a server SEiPeriod of (D)iIs a server SEiThe deadline of (2); each server SEiA class of tasks can be scheduled, which can be periodic tasks, sporadic tasks, or aperiodic tasks.
Step 102: determining a server state transition rule;
each server contains three states: active state, inactive state, suspended state; the active state refers to the server performing a task; by inactive state is meant that the processor is in an idle state or no task is waiting to execute or the budget of the processor is exhausted; the suspended state refers to that a task is waiting to execute or a server has finished executing the task but the processor budget of the server is not exhausted; initially, the server is in an inactive state; at time t, when a task is waiting to be executed, the server changes from an inactive state to a suspended state; furthermore, all tasks before time t are finished executing and the processor budget allocated to it is not exhausted, while the server is still in a suspended state; at time t, no task waiting to be executed and the processor budget is exhausted, the server enters an inactive state; once a task begins execution, the server enters an active state.
Step 103: determining a server parameter updating rule;
server SEiBy virtual time ViCalculating the deadline of the period of the time; set at the beginningVi0 and Di＝0；
When server SEiIn an inactive state, and task instance
Figure BDA0001958029690000041
At the moment of time
Figure BDA0001958029690000042
When it arrives, update ViAnd DiAt this time, the server SEiEntering a suspended state; viAnd DiAre calculated by the following formulas, respectively:
Figure BDA0001958029690000043
wherein the content of the first and second substances,
Figure BDA0001958029690000044
is a task instance
Figure BDA0001958029690000045
Time of arrival of, PiIs a server SEiA period of (a);
when server SEiIn active state, completing task instance
Figure BDA0001958029690000046
When there is a new task
Figure BDA0001958029690000047
When the server is in the active state, updating Di；DiCalculated from the following formula:
Di＝Vi+Pi；
if there is no new task schedule, server SEiEntering a suspended state;
when server SEiVirtual time V ofiGreater than the current time t of the systemcTime, server SEiEntering an inactive state;
when server SEiIn suspended state and task instance
Figure BDA0001958029690000048
When the value reaches, j is a positive integer larger than 1, and D of the value is updatedi，DiCalculated from the following formula:
Di＝Vi+Pi；
at this time, the server enters an active state;
when the processor is in the idle state, all servers enter an inactive state.
Step 104: scheduling the server according to the earliest deadline priority policy;
the priority of the server is determined by the deadline of the server, and the deadline of the server is determined by the parameter updating rule of the server; the smaller the deadline of the server is, the higher the priority thereof is; the larger the deadline of the server is, the lower the priority thereof is; when the deadline of the server is the same, determining the priority according to the activated time of the server; the closer the activated time is, the higher the priority is; the farther the activated time, the lower its priority; the activated time of the server refers to the moment when the server enters an active state; when the activated time of the server is the same, the priority of the server with small subscript is high, the priority of the server with large subscript is low; the server with the higher priority is scheduled preferentially.
Step 105: calculating the execution speed S of the task;
when server SEiIn an inactive state, and task instance
Figure BDA0001958029690000049
When the task arrives, the execution speed of the task is S + UiWherein the initial value of S is set to 0;
when server SEiIn suspended state and virtual time ViEqual to the current time t of the systemcOr when the processor budget is exhausted, the execution speed S of the task is S-Ui。
Step 106: once processor idle time exceedsProcessor state switching overhead toSwitching the processor to a low power consumption state until a new task is released;
processor state switching overhead t once processor idle time exceedsoSwitching the processor to a low power consumption state until a new task is released; processor state switching overhead toCalculated from the following formula:
to＝max{To,Bo}
wherein, ToIs the time overhead of the processor state transition, BoIs the time that the processor energy consumption balances, i.e. when the idle time does not exceed BoWhen the processor is switched to a low power consumption state, energy cannot be saved, and energy consumption is increased; only processors with idle times exceeding BoWhen it is time to switch the processor to a low power consumption state, BoCalculated from the following formula:
Figure BDA0001958029690000051
wherein E isoIs the energy consumption overhead of the processor state transition, PaAnd PsPower consumption of the processor in active mode and sleep mode, respectively.
In this embodiment, consider a system with 4 tasks, where task T1And task T2Is a sporadic task, task T3And task T4Is a periodic task. Sporadic task T1And sporadic task T2Are 8 and 16, respectively. Sporadic task T1The execution time of (1) to (2) and an occasional task T2The execution time of the method is between 2 and 4. Sporadic task T1The release time of the first instance of (1) is 0, the execution time thereof is 1, the release time of the second instance of (10) is 2; sporadic task T2The first instance of (2) has a release time of 0 and an execution time of 2, and the second instance has a release time of 18 and an execution time of 4. Periodic task T3And a periodic task T4Are 4 and 32, respectively. Periodic task T3And period T4Are 1 and 8, respectively. Periodic task T3And a periodic task T4Also released at time 0.
In the interval [0,32 ]]The 4 tasks are scheduled using the method of the present invention. Corresponding to these four tasks, four servers SE are provided1,SE2,SE3,SE4These 4 tasks are scheduled. Utilization U of these four servers1,U2,U3,U4Set to 0.25,0.25, respectively; period P of these four servers1,P2,P3,P4Set to 8,16,4,32, respectively. At time 0, server SE1,SE2,SE3,SE4The deadlines of (a) are 8,16,4, 32; and all servers enter an active state at this time, and the execution speed S at this time is 1. Thus, the periodic task T3Start execution with execution speed S ═ 1 and finish execution at time 1, server SE3Entering a suspended state; time 1, sporadic task T1Starting execution at an execution speed S of 1 and completing execution at time 2; server SE at this time1Entering a suspended state; time 2, sporadic task T2Start execution at execution speed S-1, finish execution at time 4, server SE2Entering a suspended state; at time 4, SE1The inactive state is entered, the execution speed S at this time is 0.75, and at time 4, the periodic task T3Instance of (SE) to the Server SE3Is 8 and enters the active state; so at time 4, the periodic task T3Execution is completed at execution speed S of 0.75 and at time 5.33. At time 5.33, the periodic task T4At the execution speed S equal to 0.75, at the time 8, the periodic task T3Instance of (SE) to the Server SE3Is 12 and enters the active state. Server SE at this time2And entering an inactive state, wherein the execution speed is S-0.5. Periodic task T3The execution is started at an execution speed S of 0.5, and completed at time 10, and server SE3A suspend state is entered. At time 10, sporadic task T1The second instance of (2) arrives at the server SE1Enter into active stateAnd the state is 18, and the execution speed S at this time is 0.75. At time 12, the periodic task T3Instance of (SE) to the Server SE3Is 16 and enters the active state; periodic task T3Starting execution with S-0.75 and completing execution at time 13.33, server SE3A suspend state is entered. At time 13.33, contingent task T1Execution continues with S-0.75 and completes at time 14, and server SE1A suspend state is entered. At the time 14, the periodic task T4The execution is performed at an execution speed S of 0.75. At the time 16, the periodic task T3Instance of (SE) to the Server SE3Is 20 and enters the active state. The execution speed S at this time is 0.75. Periodic task T3Execution is completed at time 17.33 with execution speed S equal to 0.75, server SE3A suspend state is entered. At time 17.33, periodic task T4The execution is performed at an execution speed S of 0.75. At time 18, sporadic task T2The second instance of (2) arrives at the server SE2Enter the active state and have a deadline of 34; the execution speed S at this time is 1. Thus, at time 18, the periodic task T4The execution is continued at the execution speed S ═ 1. At time 20, the periodic task T3Instance of (SE) to the Server SE3Is 24 and enters the active state. Periodic task T3Execution is completed at time 21 with execution speed S equal to 1, server SE3A suspend state is entered. At the time 21, the periodic task T4Continuing with execution speed S ═ 1 and which completes execution at time 22.95, server SE4A suspend state is entered. At time 22.95, sporadic task T2The execution is performed at an execution speed S of 1. At the time 24, the periodic task T3Instance of (SE) to the Server SE328 and enters the active state. At time 25, the periodic task T3Execution is completed and server SE3A suspend state is entered. At time 25, sporadic task T2Execution continues with execution speed S ═ 1, and execution completes at time 27.95, server SE2A suspend state is entered. At the time 28, the periodic task T3Instance of (2) to a serverSE3Is 32 and enters the active state. At the time 29, the periodic task T3Execution is completed and server SE3A suspend state is entered.
It is clear that the process of the invention saves about 19.43% of the energy consumption compared to other processes which do not use energy saving techniques.
The above description is only an embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the present invention.

Claims (4)

1. A multitask general energy consumption optimization method is characterized in that a system is composed of n servers, and the n servers use SE1,SE2,…,SEnRepresents; any server SEiI is more than or equal to 1 and less than or equal to n, i is a positive integer and comprises a triplet (U)i,Pi,Di) Wherein U isiIs a server SEiUtilization ratio of (P)iIs a server SEiPeriod of (D)iIs a server SEiThe deadline of (2); each server SEiA type of task may be scheduled, which may be a periodic task, an occasional task, or an aperiodic task including:
establishing n server models;
determining a server state transition rule;
determining a server parameter update rule comprising: server SEiBy virtual time ViCalculating the deadline of the period of the time; set V at the beginningi0 and Di＝0；
When server SEiIn an inactive state, and task instance
Figure FDA0003456504810000011
At the moment of time
Figure FDA0003456504810000012
When it arrives, update ViAnd DiJ is a positive integer greater than 1Counting; server SE at this timeiEntering a suspended state;
when server SEiIs in active state and completes task instance
Figure FDA0003456504810000013
When there is a new task
Figure FDA0003456504810000014
When the server is in the active state, updating ViAnd Di(ii) a If there is no new task schedule, server SEiEntering a suspended state;
when server SEiVirtual time V ofiGreater than the current time t of the systemcTime, server SEiEntering an inactive state;
when server SEiIn suspended state and task instance
Figure FDA0003456504810000015
Arrives and updates its Di(ii) a At this time, the server enters an active state;
when the processor is in an idle state, all the servers enter an inactive state;
scheduling the server according to the earliest deadline priority policy;
calculating the execution speed S of the task;
processor state switching overhead t once processor idle time exceedsoSwitching the processor to a low power consumption state until a new task is released;
the earliest deadline priority policy comprises: the smaller the deadline of the server is, the higher the priority of the server is, and the larger the deadline of the server is, the lower the priority of the server is; when the deadline of the server is the same, determining the priority according to the activated time of the server, wherein the closer the activated time is, the higher the priority is, and the farther the activated time is, the lower the priority is; when the activated time of the server is the same, the priority of the server with small subscript is high, the priority of the server with large subscript is low; the server with the higher priority is scheduled preferentially.
2. The method of claim 1, wherein determining the server state transition rule comprises:
each server contains three states: an active state, an inactive state, or a suspended state; initially, the server is in an inactive state; at time t, when a task is waiting to be executed, the server changes from an inactive state to a suspended state; furthermore, all tasks before time t are finished executing and the processor budget allocated to it is not exhausted, while the server is still in a suspended state; at time t, no task waiting to be executed and the processor budget is exhausted, the server enters an inactive state; once a task begins execution, the server enters an active state.
3. The method of claim 1, wherein calculating the execution speed S of the task comprises:
when server SEiIn an inactive state, and task instance
Figure FDA0003456504810000021
When the task arrives, the execution speed of the task is S + UiWherein the initial value of S is set to 0;
when server SEiIn suspended state and virtual time ViEqual to the current time t of the systemcOr when the processor budget is exhausted, the execution speed S of the task is S-Ui。
4. The method of claim 1, wherein the processor state switching overhead t is exceeded once the processor idle time exceeds the processor state switching overhead toSwitching the processor to a low power consumption state until a new task is released, comprising:
processor state switching overhead toCalculated from the following formula:
to＝max{To,Bo}
wherein, ToIs the time overhead of the processor state transition, BoIs the time for the processor to power balance.
CN201910073576.3A 2019-01-25 2019-01-25 Multi-task general energy consumption optimization method Active CN109739332B (en)

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 CN109739332A (en) 2019-05-10
CN109739332B true 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)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110825504B (en) * 2019-10-24 2023-03-07 华侨大学 Server-based hybrid key accidental task energy consumption optimization method
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
CN112633589B (en) * 2020-12-30 2022-07-29 华侨大学 Probability model-based hybrid key task energy consumption optimization scheduling method
CN113821339B (en) * 2021-08-20 2022-11-08 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)

* Cited by examiner, † Cited by third party
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

Patent Citations (5)

* Cited by examiner, † Cited by third party
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)

* Cited by examiner, † Cited by third party
Title
硬实时系统周期任务低功耗调度算法;张忆文;《西安交通大学学报》;20140731;全文 *
面向硬实时系统零星任务低调度算法;张忆文等;《小型微型计算机系统》;20160131;全文 *

Also Published As

Publication number Publication date
CN109739332A (en) 2019-05-10

Similar Documents

Publication Publication Date Title
CN109739332B (en) Multi-task general energy consumption optimization method
CN101339521B (en) Tasks priority dynamic dispatching algorithm
CN109324880A (en) A kind of low-power consumption scheduling method suitable for real-time system periodic task model
US9104500B1 (en) Lock-free job scheduler for multi-processor systems
CN106970835B (en) Hierarchical energy consumption optimization method for fixed priority resource-limited system
CN110187956B (en) Layered real-time task scheduling method and system for multi-agent platform
CN110308977B (en) Low-energy-consumption method for updating mixed key accidental tasks by dynamic utilization rate
CN101923487A (en) Comprehensive embedded type real-time period task scheduling method
EP3278220B1 (en) Power aware scheduling and power manager
CN110221907B (en) Real-time task scheduling method based on EDF algorithm and fuzzy set
CN112269632B (en) Scheduling method and system for optimizing cloud data center
CN109597378B (en) Resource-limited hybrid task energy consumption sensing method
CN106445070B (en) Energy consumption optimization scheduling method for hard real-time system resource-limited sporadic tasks
CN107391244A (en) A kind of Internet of Things operating system dispatching method based on mixed scheduling model
CN109324891A (en) A kind of periodic duty low-power consumption scheduling method of ratio free time distribution
JP2011501309A (en) Managing preemption in real-time operating systems
CN113535356A (en) Energy-aware hierarchical task scheduling method and device
CN107329817B (en) Standby standby system mixed partitioning reliability perception energy consumption optimization method
CN112130992A (en) Low-power-consumption scheduling method based on high-performance open type numerical control system
KR20130022976A (en) System and method for deadline based priority inheritance
CN108563497B (en) Efficient multi-dimensional algorithm scheduling method and task server
CN110825504B (en) Server-based hybrid key accidental task energy consumption optimization method
JP2014225137A (en) Task scheduler, microprocessor, and task scheduling method
CN105677449B (en) A kind of low-power consumption scheduling method suitable for digital control system
CN111813553B (en) Task dynamic priority low-energy consumption method based on selectable factor period

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