CN110427257A - Multi-task scheduling method, device and computer readable storage medium - Google Patents
Multi-task scheduling method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN110427257A CN110427257A CN201910699372.0A CN201910699372A CN110427257A CN 110427257 A CN110427257 A CN 110427257A CN 201910699372 A CN201910699372 A CN 201910699372A CN 110427257 A CN110427257 A CN 110427257A
- Authority
- CN
- China
- Prior art keywords
- task
- new
- queue
- time
- priority
- 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.)
- Pending
Links
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
- G06F9/4812—Task transfer initiation or dispatching by interrupt, e.g. masked
- G06F9/4831—Task transfer initiation or dispatching by interrupt, e.g. masked with variable priority
-
- 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
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Telephonic Communication Services (AREA)
Abstract
The present invention relates to a kind of artificial intelligence technologys, disclose a kind of multi-task scheduling method, comprising: receive new task tnew, according to access principle to the new task tnewAccess judgement is carried out, if the new task tnewIt is unsatisfactory for the access principle, then does not allow the new task tnewInto task queue, if the new task tnewMeet the access principle, is based on the new task tnewTask type operation of jumping the queue is carried out to the task queue, after the completion of the operation of jumping the queue, based on seizing principle to the new task tnewBe currently running that task seize operation and priority reverses operation to obtain updated task queue in the task queue, task schedule is carried out based on updated task queue.The present invention also proposes a kind of multi-task scheduling device and a kind of computer readable storage medium.Efficient multi-task scheduling function may be implemented in the present invention.
Description
Technical field
The present invention relates to field of artificial intelligence more particularly to a kind of multi-task scheduling method, device and computer can
Read storage medium.
Background technique
With the development of current science and technology, the quantity for servicing end subscriber is gradually increasing, and causes the task of server-side to bear and becomes
Weight, and then lead to server crash.And existing solution is mostly increase hardware configuration, i.e., by increasing server or right
Number of servers bears excessive problem to alleviate server.Although increase hardware configuration can solve speed of response of server slow
The problem of, but will cause the wasting of resources, therefore the multitask resource solution of high efficiency low cost is badly in need of solving always.
Summary of the invention
The present invention provides a kind of multi-task scheduling method, device and computer readable storage medium, main purpose and is
The link for easily occurring mistake in operating of insuring to user carries out intelligent prevention.
To achieve the above object, a kind of multi-task scheduling method provided by the invention, comprising:
Receive new task tnew, judge the new task tnewIt is made whether to meet task access principle;
When judging the new task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewInto appoint
Business queue;
When judging the new task tnewWhen meeting the task access principle, it is based on the new task tnewTask type
By the new task tnewIt is inserted into the task queue;
As the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenewAnd institute
It states the task execution resource being currently running in task queue and seizes operation and priority and operation is reversed to obtain updated
Business queue;
Task schedule is carried out based on updated task queue.
Optionally, the new task tnewIt is operated including one or more requests;
The task queue stores waiting task ti, the waiting task tiIt is divided into real-time task based on time factor
And un-real time job;
The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition,
Wherein, AiFor the waiting task tiSubmission time, TiFor the waiting task tiDuty cycle, EiTable
Show the waiting task tiAverage performance times, PiIndicate the waiting task tiTask priority, diDescribed in expression
Waiting task tiThe absolute cutoff time limit, idiRepresent the waiting task tiNumber.
Optionally, to the new task tnewCarry out task access judges;
Judge whether the task queue has been stored with the new task tnew;
When the task queue is stored with the new task tnewWhen, do not allow the new task tnewIt repeats to enter described appoint
Business queue;
As the not stored new task t of the task queuenewWhen, when judging that the task of the task queue averagely executes
Between with the new task tnewDuty cycle size relation;
When the task average performance times of task queue are greater than the new task tnewDuty cycle when, do not allow described
New task tnewInto the task queue;
When the task average performance times of task queue are less than or equal to the new task tnewDuty cycle when, allow
The new task tnewInto the task queue.
Optionally, described to be based on the new task tnewTask type by the new task tnewIt is inserted into the task team
Column, comprising:
Each un-real time job t in the task queue is setjParameter Cj, CjIndicate real-time task tjTime jumped the queue
Number, and initialize the parameter CjIt is 0;
As the new task tnewWhen not being real-time task, by the new task tnewInstitute is inserted into according to the task priority
State task queue;
As the new task tnewWhen for real-time task, by the new task tnewAccording to the time factor and the task
Priority is inserted into the task queue;
As the un-real time job tjIt is entered the new task t of the task queuenewWhen jumping the queue one time, the parameter CjFrom
It is dynamic to add 1;
As the parameter CjWhen greater than preset threshold C, by the un-real time job tjTime factor be set to 0, do not allow institute
State un-real time job tjIt is entered the new task t of the task queuenewIt jumps the queue.
Optionally, described that principle is seized to the new task t based on resourcenewWith currently transported in the task queue
Capable task execution resource seizes operation, comprising:
Judge the new task tnewTask priority PnewWith times for being currently running task in the task queue
Business priority PcurSize relation;
As the new task tnewTask priority and the task priority for being currently running task identical or less than institute
When stating the task priority for the task of being currently running, the new task t is not allowednewThe calculating for being currently running task is not seized
Resource;
As the new task tnewTask priority PnewHigher than the task priority P for being carrying out taskcurWhen,
Judge (Ecur-ecur)+EnewWith dnew-AnewSize relation, wherein AnewIndicate the new task tnewSubmission time, Ecur
The average performance times of task, E are currently running described in expressionnewIndicate the new task tnewAverage performance times, ecurIt indicates
The time for being currently running task and having run, dnewIndicate the new task tnewThe absolute cutoff time limit;
If (Ecur-ecur)+Enew< dnew-Anew, do not allow the new task tnewTask is currently running described in not seizing
Computing resource;
If (Ecur-ecur)≥Enew≥dnew-Anew, allow the new task tnewSeize the meter of being currently running for the task
Calculate resource.
In addition, to achieve the above object, the present invention also provides a kind of multi-task scheduling device, the device include memory and
Processor is stored with the multi-task scheduling program that can be run on the processor, the multi-task scheduling in the memory
Program realizes following steps when being executed by the processor:
Receive new task tnew, judge the new task tnewIt is made whether to meet task access principle;
When judging the new task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewInto appoint
Business queue;
When judging the new task tnewWhen meeting the task access principle, it is based on the new task tnewTask type
By the new task tnewIt is inserted into the task queue;
As the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenewAnd institute
It states the task execution resource being currently running in task queue and seizes operation and priority and operation is reversed to obtain updated
Business queue;
Task schedule is carried out based on updated task queue.
Optionally, the new task tnewIt is operated including one or more requests;
The task queue stores waiting task ti, the waiting task tiIt is divided into real-time task based on time factor
And un-real time job;
The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition,
Wherein, AiFor the waiting task tiSubmission time, TiFor the waiting task tiDuty cycle, EiTable
Show the waiting task tiAverage performance times, PiIndicate the waiting task tiTask priority, diDescribed in expression
Waiting task tiThe absolute cutoff time limit, idiRepresent the waiting task tiNumber.
Optionally, to the new task tnewCarry out task access judges;
Judge whether the task queue has been stored with the new task tnew;
When the task queue is stored with the new task tnewWhen, do not allow the new task tnewIt repeats to enter described appoint
Business queue;
As the not stored new task t of the task queuenewWhen, when judging that the task of the task queue averagely executes
Between with the new task tnewDuty cycle size relation;
When the task average performance times of task queue are greater than the new task tnewDuty cycle when, do not allow described
New task tnewInto the task queue;
When the task average performance times of task queue are less than or equal to the new task tnewDuty cycle when, allow
The new task tnewInto the task queue.
Optionally, described to be based on the new task tnewTask type by the new task tnewIt is inserted into the task team
Column, comprising:
Each un-real time job t in the task queue is setjParameter Cj, CjIndicate real-time task tjTime jumped the queue
Number, and initialize the parameter CjIt is 0;
As the new task tnewWhen not being real-time task, by the new task tnewInstitute is inserted into according to the task priority
State task queue;
As the new task tnewWhen for real-time task, by the new task tnewAccording to the time factor and the task
Priority is inserted into the task queue;
As the un-real time job tjIt is entered the new task t of the task queuenewWhen jumping the queue one time, the parameter CjFrom
It is dynamic to add 1;
As the parameter CjWhen greater than preset threshold C, by the un-real time job tjTime factor be set to 0, do not allow institute
State un-real time job tjIt is entered the new task t of the task queuenewIt jumps the queue.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer readable storage medium
Multi-task scheduling program is stored on storage medium, the multi-task scheduling program can be executed by one or more processor,
With the step of realizing multi-task scheduling method as described above.
The present invention carries out access judgement to task first by access principle, and the pressure of server can be reduced by access judgement
Power, while carrying out insertion judgement to task, can handle emergency task in time, the responsiveness to task can be improved, in addition into
The problem of row major grade reverses operation that can be ranked up to task, effectively prevent task excessive, some tasks are unable to get response.
Therefore multi-task scheduling method, device and computer readable storage medium proposed by the present invention, may be implemented the money of high efficiency low cost
Source solves.
Detailed description of the invention
Fig. 1 is the flow diagram for the multi-task scheduling method that one embodiment of the invention provides;
Fig. 2 is the schematic diagram of internal structure for the multi-task scheduling device that one embodiment of the invention provides;
The module diagram of multi-task scheduling program in the multi-task scheduling device that Fig. 3 provides for one embodiment of the invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The present invention provides a kind of multi-task scheduling method.It is the multitask that one embodiment of the invention provides shown in referring to Fig.1
The flow diagram of dispatching method.This method can be executed by a device, which can be by software and or hardware realization.
In the present embodiment, multi-task scheduling method includes:
S1, new task t is receivednew。
Preferably, the new task tnewIt is a series of request requirement, the request requires generally to be initiated by user, such as
Scientific algorithm task (such as addition subtraction multiplication and division, derivation), information handling task (such as data classification, storage, statistics), Computer Aided Design
With manufacture (such as building G- Design, art toning), network communication task (such as various soft and hardware resources shared).
Specifically, the new task tnewIt is being inserted into task queue for task, and in the task queue
Task be known as waiting task, use tiIt indicates.The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition, wherein
AiFor the task tiSubmission time;TiFor the tiDuty cycle;EiIndicate the tiAverage performance times;PiIt indicates
The tiQueue priority;diIndicate the tiThe absolute cutoff time limit;idiRepresent the tiNumber.Implement in the present invention
In example, 0≤idi≤254。
S2, according to task access principle to the new task tnewCarry out access judgement.
Preferably, the task access principle successively judges two conditions in sequence: (1) judging in the task queue
Whether the new task t is had existednew;When there is no the new task t in the task queuenew, then (2) judge the task
The task average performance times of queue and the new task tnewDuty cycle TnewSize relation.
Wherein, described to judge the task t whether is had existed in the task queuenewIt include: the traversal task team
All waiting task t in columniObtain waiting task set S={ t1, t2..., tn, wherein n is the task queue
The sum of middle waiting task;Judge the new task tnewWith each waiting task t in the waiting task set Si
Queue priority and number it is whether equal, i.e., whether Pi=Pnew, idi=idnewIf the queue priority and number all phases
Deng then refusing the new task tnewInto task queue, if the queue priority and numbering that at least one is unequal,
Into the Rule of judgment (2) of the task access principle.
Preferably, the queue priority includes three grades, respectively P1, P2, P3, wherein P1Indicate that high priority can
It seizes, P2Indicate that high priority can not be seized, P3Indicate Standard Task queue, queue priority is from P1To P3It successively reduces, P2
And P3Task in queue is scheduled according to first-serve scheduling policy (FCFS).
The principle of the first-serve scheduling policy refers to successively to be held according to the sequencing for entering the task queue
Row, i.e., task can not be seized mutually in the queue, once therefore task start to execute, to run to end, only work as task
When abandoning the CPU right to use, next task can be just continued to execute.Further, in practical applications by a task
More task handover operations will be will lead to by distributing multiple queue priorities, therefore be advised based on the first-serve scheduling policy
A fixed task can only be assigned in a task queue, guarantee a task only in a task queue, and only
One queue priority.
Preferably, the task average performance times and the new task t of the judgement task queuenewTask week
Phase TnewSize relation, comprising: traverse out in the waiting task set S all queue priorities and be higher than the new task
tnewThe set of tasks S of queue prioritys, to the set of tasks SsDescribed in task queue task average performance times summation
It obtains overall average and executes the time, judge that the overall average executes time and the new task tnewDuty cycle size relation,
If the overall average executes the time and is greater than the new task tnewDuty cycle, then refuse the new task tnewInto task team
Column, if the overall average executes the time and is less than the new task tnewDuty cycle, then allow the new task tnewInto appoint
Business queue.
S3, when judging the new task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewInto
Enter task queue.
According to step S2, the new task tnewBeing unsatisfactory for the task access principle, mainly there are two aspects: (1)
According to the first-serve scheduling policy, a task can only be assigned in a task queue, guarantee a task only
Have in a task queue, and only one queue priority, so when the task queue has had the new task tnew
When, the new task tnewIt is unsatisfactory for the task access principle;(2) when there is no the new task t in the task queuenew,
And average performance times are greater than the new task t in the task queuenewDuty cycle Tnew, therefore the new task tnew
It is unsatisfactory for the task access principle.
Preferably, as the new task tnewIt is unsatisfactory for the task access principle, the new task t can be ignorednewOr
The processing elements such as related CPU are called directly to the new task tnew。
S4, when judging the new task tnewWhen meeting the task access principle, it is based on the new task tnewTask
Type is by the new task tnewIt is inserted into the task queue.
Preferably, the task type includes real-time task and un-real time job.Wherein, to be processed with time restriction
Task is real-time task, and the waiting task without time restriction is un-real time job.New task tnewIt is quasi- to meet the task
Enter principle to put to when task queue, if there are the time factors of waiting task to be greater than the new task t in queuenewTime
The factor, can jump the queue phenomenon at this time, i.e., the big waiting task of time factor can be mobile to tail of the queue in queue.In the task
The waiting task having in queue has time restriction, and the time factor is to indicate the journey that is pressed for time of the real-time task
Degree, is generally described with the absolute cutoff time limit of the real-time task, i.e. the time factor of real-time task is smaller, in task queue
In queue priority it is higher, waiting task does not have time restriction to be known as un-real time job, then time factor is set to negative, with
Just it is distinguished with the waiting task with time restriction.
Since real-time task is described by the time factor, if the new task tnewIt is also real-time task, then when will be described
New task tnewBe put into when the task queue, the un-real time job in the task queue since time factor is negative, because
This described real-time task tnewIt can directly jump the queue, i.e., the described new task tnewPriority it is higher.However, if described jump the queue frequently,
It will cause un-real time jobs cannot handle always, it is therefore desirable to carry out operation of jumping the queue.
Preferably, the operation of jumping the queue is in advance by each un-real time job t in the task queuejSet a ginseng
Number Cj(parameter CjIt is initially 0), each new task tnewWhen meeting the task access principle, the new task t is judgednewWhether
For real-time task.If the new task tnewFor un-real time job, then according to the queue priority by the new task tnewIt puts
It sets in the task queue.If the new task tnewIt is for real-time task, then preferential according to the time factor and the queue
Grade, by the new task tnewIt is placed on the task queue, and updates the un-real time job tjBy number of jumping the queue.As described in
Un-real time job tjIt is jumped the queue primary, the parameter CjAdd 1, as the parameter CjIt, will be described non real-time when reaching preset threshold C
Task tjTime factor be set to 0.Time factor is smaller, and the queue priority in task queue is higher, therefore the non-reality
When task tjIt cannot be jumped the queue, and finally obtain execution.
S5, as the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenew
Operation is seized with the task execution resource being currently running in the task queue and priority reverse operates after obtaining update
Task queue.
It is found that the queue priority of waiting task is divided into three grades, respectively P as described in above-mentioned steps S21, P2,
P3, wherein P1Indicate that high priority can be seized, P2Indicate that high priority can not be seized, P3It indicates Standard Task queue, therefore ought have
New task tnewInto when task queue, if the new task tnewQueue priority be P1, and being currently running for task
Priority is P2, it is likely that task can occur and seize abnormal item, i.e., the described new task tnewReplace at once it is described it is current
The task of operation.
Present pre-ferred embodiments are based on the task and seize abnormal item, and principle is seized in building in advance.It is described to seize original
It then include following two aspects, as the new task tnewIn the case where meeting one of described two aspects, it is not possible into
Row task is seized.
Preferably, in terms of two for seizing principle successively are as follows: (1) as the new task tnewQueue priority and
The identical i.e. P of the queue priority of being currently running for tasknew=PcurNo matter the then new task tnewTime factor size, all
It not can be carried out task to seize.(2) as the new task tnewQueue priority be higher than the task queue priority that is carrying out i.e.
Pnew=P1, Pnew> PcurIf (Ecur-ecur)+Enew≥dnew-Anew, the new task tnewTimes being currently running can be seized
Business, otherwise can not seize.A in formulanewIndicate new task tnewSubmission time, EcurIndicate that being averaged for task being currently running is held
Row time, EnewIndicate new task tnewAverage performance times, ecurExpression is currently running the time that task has been run,
dnewIndicate new task tnewThe absolute cutoff time limit.
Preferably, the principle of seizing shows that task is seized only in the new task tnewQueue priority ratio
The priority of operation task is high, and meet can preemptive condition can just be seized, both ensure that important and urgent appointed in this way
While real-time of being engaged in, and reduce overhead.
If the new task tnewSatisfaction be when can seize the task that other queues are currently running, in this case system
Resource access mechanism is likely to result in priority reversal, i.e., after high-priority task waits low priority task to be finished
It could execute.It is reversed in order to avoid priority occurs, the present invention takes priority inheritance method, i.e., when high-priority task accesses
When the resource occupied by low priority task, the queue priority of low priority task is improved to the queue of high-priority task
Priority, and being set in queue is highest priority task i.e. head of the queue task, quickly can be preferentially finished, it will
High-priority task blocking time is limited in a smaller time interval, avoids the generation of priority reversal.
Based on the above, by the new task tnewIt is placed in the task queue, therefore team is completed in the task queue
Column, which update, obtains updated task queue.
S6, task schedule is carried out based on updated task queue.
In embodiments of the present invention, the task schedule is to read according to queue priority from the updated task queue
Take inter-related task.
Invention also provides a kind of multi-task scheduling device.It is the multitask that one embodiment of the invention provides referring to shown in Fig. 2
The schematic diagram of internal structure of dispatching device.
In the present embodiment, the multi-task scheduling device 1 can be PC (Personal Computer, PC),
The either terminal devices such as smart phone, tablet computer, portable computer are also possible to a kind of server etc..The multitask tune
It spends device 1 and includes at least memory 11, processor 12, communication bus 13 and network interface 14.
Wherein, memory 11 include at least a type of readable storage medium storing program for executing, the readable storage medium storing program for executing include flash memory,
Hard disk, multimedia card, card-type memory (for example, SD or DX memory etc.), magnetic storage, disk, CD etc..Memory 11
It can be the internal storage unit of multi-task scheduling device 1 in some embodiments, such as the multi-task scheduling device 1 is hard
Disk.Memory 11 is also possible to the External memory equipment of multi-task scheduling device 1, such as multitask tune in further embodiments
The plug-in type hard disk being equipped on degree device 1, intelligent memory card (Smart Media Card, SMC), secure digital (Secure
Digital, SD) card, flash card (Flash Card) etc..Further, memory 11 can also both include that multi-task scheduling fills
The internal storage unit for setting 1 also includes External memory equipment.Memory 11 can be not only used for storage and be installed on multi-task scheduling
Application software and Various types of data, such as the code of multi-task scheduling program 01 of device 1 etc. can be also used for temporarily storing
Data through exporting or will export.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), controller, microcontroller, microprocessor or other data processing chips, the program for being stored in run memory 11
Code or processing data, such as execute multi-task scheduling program 01 etc..
Communication bus 13 is for realizing the connection communication between these components.
Network interface 14 optionally may include standard wireline interface and wireless interface (such as WI-FI interface), be commonly used in
Communication connection is established between the device 1 and other electronic equipments.
Optionally, which can also include user interface, and user interface may include display (Display), input
Unit such as keyboard (Keyboard), optional user interface can also include standard wireline interface and wireless interface.It is optional
Ground, in some embodiments, display can be light-emitting diode display, liquid crystal display, touch-control liquid crystal display and OLED
(Organic Light-Emitting Diode, Organic Light Emitting Diode) touches device etc..Wherein, display can also be appropriate
Referred to as display screen or display unit, for being shown in the information handled in multi-task scheduling device 1 and for showing visualization
User interface.
Fig. 2 illustrates only the multi-task scheduling device 1 with component 11-14 and multi-task scheduling program 01, this field
Technical staff may include ratio it is understood that structure shown in fig. 1 does not constitute the restriction to multi-task scheduling device 1
It illustrates less perhaps more components and perhaps combines certain components or different component layouts.
In 1 embodiment of device shown in Fig. 2, multi-task scheduling program 01 is stored in memory 11;Processor 12 is held
Following steps are realized when the multi-task scheduling program 01 stored in line storage 11:
Step 1: receiving new task tnew。
Preferably, the new task tnewIt is a series of request requirement, the request requires generally to be initiated by user, such as
Scientific algorithm task (such as addition subtraction multiplication and division, derivation), information handling task (such as data classification, storage, statistics), Computer Aided Design
With manufacture (such as building G- Design, art toning), network communication task (such as various soft and hardware resources shared).
Specifically, the new task tnewIt is being inserted into task queue for task, and in the task queue
Task be known as waiting task, use tiIt indicates.The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition, wherein
AiFor the task tiSubmission time;TiFor the tiDuty cycle;EiIndicate the tiAverage performance times;PiIt indicates
The tiQueue priority;diIndicate the tiThe absolute cutoff time limit;idiRepresent the tiNumber.Implement in the present invention
In example, 0≤idi≤254。
Step 2: according to task access principle to the new task tnewCarry out access judgement.
Preferably, the task access principle successively judges two conditions in sequence: (1) judging in the task queue
Whether the new task t is had existednew;When there is no the new task t in the task queuenew, then (2) judge the task
The task average performance times of queue and the new task tnewDuty cycle TnewSize relation.
Wherein, described to judge the task t whether is had existed in the task queuenewIt include: the traversal task team
All waiting task t in columniObtain waiting task set S={ t1, t2..., tn, wherein n is the task queue
The sum of middle waiting task;Judge the new task tnewWith each waiting task t in the waiting task set Si
Queue priority and number it is whether equal, i.e., whether Pi=Pnew, idi=idnewIf the queue priority and number all phases
Deng then refusing the new task tnewInto task queue, if the queue priority and numbering that at least one is unequal,
Into the Rule of judgment (2) of the task access principle.
Preferably, the queue priority includes three grades, respectively P1, P2, P3, wherein P1Indicate that high priority can
It seizes, P2Indicate that high priority can not be seized, P3Indicate Standard Task queue, queue priority is from P1To P3It successively reduces, P2
And P3Task in queue is scheduled according to first-serve scheduling policy (FCFS).
The principle of the first-serve scheduling policy refers to successively to be held according to the sequencing for entering the task queue
Row, i.e., task can not be seized mutually in the queue, once therefore task start to execute, to run to end, only work as task
When abandoning the CPU right to use, next task can be just continued to execute.Further, in practical applications by a task
More task handover operations will be will lead to by distributing multiple queue priorities, therefore be advised based on the first-serve scheduling policy
A fixed task can only be assigned in a task queue, guarantee a task only in a task queue, and only
One queue priority.
Preferably, the task average performance times and the new task t of the judgement task queuenewTask week
Phase TnewSize relation, comprising: traverse out in the waiting task set S all queue priorities and be higher than the new task
tnewThe set of tasks S of queue prioritys, to the set of tasks SsDescribed in task queue task average performance times summation
It obtains overall average and executes the time, judge that the overall average executes time and the new task tnewDuty cycle size relation,
If the overall average executes the time and is greater than the new task tnewDuty cycle, then refuse the new task tnewInto task team
Column, if the overall average executes the time and is less than the new task tnewDuty cycle, then allow the new task tnewInto appoint
Business queue.
Step 3: when judging the new task tnewWhen being unsatisfactory for the task access principle, the new task is not allowed
tnewInto task queue.
According to step 2, the new task tnewBeing unsatisfactory for the task access principle, mainly there are two aspects: (1)
According to the first-serve scheduling policy, a task can only be assigned in a task queue, guarantee a task only
Have in a task queue, and only one queue priority, so when the task queue has had the new task tnew
When, the new task tnewIt is unsatisfactory for the task access principle;(2) when there is no the new task t in the task queuenew,
And average performance times are greater than the new task t in the task queuenewDuty cycle Tnew, therefore the new task tnew
It is unsatisfactory for the task access principle.
Preferably, as the new task tnewIt is unsatisfactory for the task access principle, the new task t can be ignorednewOr
The processing elements such as related CPU are called directly to the new task tnew。
Step 4: when judging the new task tnewWhen meeting the task access principle, it is based on the new task tnew's
Task type is by the new task tnewIt is inserted into the task queue.
Preferably, the task type includes real-time task and un-real time job.Wherein, to be processed with time restriction
Task is real-time task, and the waiting task without time restriction is un-real time job.New task tnewIt is quasi- to meet the task
Enter principle to put to when task queue, if there are the time factors of waiting task to be greater than the new task t in queuenewTime
The factor, can jump the queue phenomenon at this time, i.e., the big waiting task of time factor can be mobile to tail of the queue in queue.In the task
The waiting task having in queue has time restriction, and the time factor is to indicate the journey that is pressed for time of the real-time task
Degree, is generally described with the absolute cutoff time limit of the real-time task, i.e. the time factor of real-time task is smaller, in task queue
In queue priority it is higher, waiting task does not have time restriction to be known as un-real time job, then time factor is set to negative, with
Just it is distinguished with the waiting task with time restriction.
Since real-time task is described by the time factor, if the new task tnewIt is also real-time task, then when will be described
New task tnewBe put into when the task queue, the un-real time job in the task queue since time factor is negative, because
This described real-time task tnewIt can directly jump the queue, i.e., the described new task tnewPriority it is higher.However, if described jump the queue frequently,
It will cause un-real time jobs cannot handle always, it is therefore desirable to carry out operation of jumping the queue.
Preferably, the operation of jumping the queue is in advance by each un-real time job t in the task queuejSet a ginseng
Number Cj(parameter CjIt is initially 0), each new task tnewWhen meeting the task access principle, the new task t is judgednewWhether
For real-time task.If the new task tnewFor un-real time job, then according to the queue priority by the new task tnewIt puts
It sets in the task queue.If the new task tnewIt is for real-time task, then preferential according to the time factor and the queue
Grade, by the new task tnewIt is placed on the task queue, and updates the un-real time job tjBy number of jumping the queue.As described in
Un-real time job tjIt is jumped the queue primary, the parameter CjAdd 1, as the parameter CjIt, will be described non real-time when reaching preset threshold C
Task tjTime factor be set to 0.Time factor is smaller, and the queue priority in task queue is higher, therefore the non-reality
When task tjIt cannot be jumped the queue, and finally obtain execution.
Step 5: working as the new task tnewAfter being inserted into the task queue, principle is seized to the new task based on resource
tnewOperation is seized with the task execution resource being currently running in the task queue and priority reverses operation to be updated
Task queue afterwards.
It is found that the queue priority of waiting task is divided into three grades, respectively P as described in above-mentioned steps two1, P2,
P3, wherein P1Indicate that high priority can be seized, P2Indicate that high priority can not be seized, P3It indicates Standard Task queue, therefore ought have
New task tnewInto when task queue, if the new task tnewQueue priority be P1, and being currently running for task
Priority is P2, it is likely that task can occur and seize abnormal item, i.e., the described new task tnewReplace at once it is described it is current
The task of operation.
Present pre-ferred embodiments are based on the task and seize abnormal item, and principle is seized in building in advance.It is described to seize original
It then include following two aspects, as the new task tnewIn the case where meeting one of described two aspects, it is not possible into
Row task is seized.
Preferably, in terms of two for seizing principle successively are as follows: (1) as the new task tnewQueue priority and
The identical i.e. P of the queue priority of being currently running for tasknew=PcurNo matter the then new task tnewTime factor size, all
It not can be carried out task to seize.(2) as the new task tnewQueue priority be higher than the task queue priority that is carrying out i.e.
Pnew=P1, Pnew> PcurIf (Ecur-ecur)+Enew≥dnew-Anew, the new task tnewTimes being currently running can be seized
Business, otherwise can not seize.A in formulanewIndicate new task tnewSubmission time, ecurIndicate that being averaged for task being currently running is held
Row time, EnewIndicate new task tnewAverage performance times, ecurExpression is currently running the time that task has been run,
dnewIndicate new task tnewThe absolute cutoff time limit.
Preferably, the principle of seizing shows that task is seized only in the new task tnewQueue priority ratio
The priority of operation task is high, and meet can preemptive condition can just be seized, both ensure that important and urgent appointed in this way
While real-time of being engaged in, and reduce overhead.
If the new task tnewSatisfaction be when can seize the task that other queues are currently running, in this case system
Resource access mechanism is likely to result in priority reversal, i.e., after high-priority task waits low priority task to be finished
It could execute.It is reversed in order to avoid priority occurs, the present invention takes priority inheritance method, i.e., when high-priority task accesses
When the resource occupied by low priority task, the queue priority of low priority task is improved to the queue of high-priority task
Priority, and being set in queue is highest priority task i.e. head of the queue task, quickly can be preferentially finished, it will
High-priority task blocking time is limited in a smaller time interval, avoids the generation of priority reversal.
Based on the above, by the new task tnewIt is placed in the task queue, therefore team is completed in the task queue
Column, which update, obtains updated task queue.
Step 6: carrying out task schedule based on updated task queue.
In embodiments of the present invention, the task schedule is to read according to queue priority from the updated task queue
Take inter-related task.
Optionally, in other embodiments, multi-task scheduling program can also be divided into one or more module, and one
A or multiple modules are stored in memory 11, and are held by one or more processors (the present embodiment is by processor 12)
For row to complete the present invention, the so-called module of the present invention is the series of computation machine program instruction section for referring to complete specific function,
For describing implementation procedure of the multi-task scheduling program in multi-task scheduling device.
It is the journey of the multi-task scheduling program in one embodiment of multi-task scheduling device of the present invention for example, referring to shown in Fig. 3
Sequence module diagram, in the embodiment, the multi-task scheduling program can be divided into task access processing module 10, task
Seize processing module 20, scheduling of resource module 30 illustratively:
The task access processing module 10 is used for: receiving new task tnew, judge the new task tnewIt is made whether full
Sufficient task access principle, when judging the new task tnewWhen being unsatisfactory for the task access principle, the new task is not allowed
tnewInto task queue, when judging the new task tnewWhen meeting the task access principle, it is based on the new task tnew's
Task type is by the new task tnewIt is inserted into the task queue.
The task is seized processing module 20 and is used for: as the new task tnewAfter being inserted into the task queue, it is based on resource
Principle is seized to the new task tnewSeized with the task execution resource being currently running in the task queue operation and it is excellent
First grade reverses operation to obtain updated task queue.
The scheduling of resource module 30 is used for: carrying out task schedule based on updated task queue.
Above-mentioned task access processing module 10, task seize the program modules quilts such as processing module 20, scheduling of resource module 30
Functions or operations step and the above-described embodiment realized when execution are substantially the same, and details are not described herein.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
On be stored with multi-task scheduling program, the multi-task scheduling program can be executed by one or more processors, as follows to realize
Operation:
Receive new task tnew, judge the new task tnewIt is made whether to meet task access principle, it is described new when judging
Task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewIt is described new when judging into task queue
Task tnewWhen meeting the task access principle, it is based on the new task tnewTask type by the new task tnewInsertion
The task queue.
As the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenewAnd institute
It states the task execution resource being currently running in task queue and seizes operation and priority and operation is reversed to obtain updated
Business queue.
Task schedule is carried out based on updated task queue.
It should be noted that the serial number of the above embodiments of the invention is only for description, do not represent the advantages or disadvantages of the embodiments.And
The terms "include", "comprise" herein or any other variant thereof is intended to cover non-exclusive inclusion, so that packet
Process, device, article or the method for including a series of elements not only include those elements, but also including being not explicitly listed
Other element, or further include for this process, device, article or the intrinsic element of method.Do not limiting more
In the case where, the element that is limited by sentence " including one ... ", it is not excluded that including process, device, the object of the element
There is also other identical elements in product or method.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of multi-task scheduling method, which is characterized in that the described method includes:
Receive new task tnew, judge the new task tnewIt is made whether to meet task access principle;
When judging the new task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewInto task team
Column;
When judging the new task tnewWhen meeting the task access principle, it is based on the new task tnewTask type by institute
State new task tnewIt is inserted into the task queue;
As the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenewWith described
The task execution resource being currently running in business queue seizes operation and priority reverses operation to obtain updated task team
Column;
Task schedule is carried out based on updated task queue.
2. multi-task scheduling method as described in claim 1, which is characterized in that the new task tnewIncluding one or more
Request operation;
The task queue stores waiting task ti, the waiting task tiIt is divided into real-time task and non-based on time factor
Real-time task;
The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition,
Wherein, AiFor the waiting task tiSubmission time, TiFor the waiting task tiDuty cycle, EiIndicate institute
State waiting task tiAverage performance times, PiIndicate the waiting task tiTask priority, diIndicate described wait locate
Reason task tiThe absolute cutoff time limit, idiRepresent the waiting task tiNumber.
3. multi-task scheduling method as claimed in claim 2, which is characterized in that the new task tnewCarry out task access
Judgement includes;
Judge whether the task queue has been stored with the new task tnew;
When the task queue is stored with the new task tnewWhen, do not allow the new task tnewIt repeats to enter the task team
Column;
As the not stored new task t of the task queuenewWhen, judge the task average performance times of the task queue with
The new task tnewDuty cycle size relation;
When the task average performance times of task queue are greater than the new task tnewDuty cycle when, do not allow the new post
Be engaged in tnewInto the task queue;
When the task average performance times of task queue are less than or equal to the new task tnewDuty cycle when, described in permission
New task tnewInto the task queue.
4. multi-task scheduling method as claimed in claim 3, which is characterized in that described to be based on the new task tnewTask
Type is by the new task tnewIt is inserted into the task queue, comprising:
Each un-real time job t in the task queue is setjParameter Cj, CjIndicate real-time task tjThe number jumped the queue, and
Initialize the parameter CjIt is 0;
As the new task tnewWhen not being real-time task, by the new task tnewAppoint according to task priority insertion is described
Business queue;
As the new task tnewWhen for real-time task, by the new task tnewIt is preferential according to the time factor and the task
Grade is inserted into the task queue;
As the un-real time job tjIt is entered the new task t of the task queuenewWhen jumping the queue one time, the parameter CjIt is automatic to add
1;
As the parameter CjWhen greater than preset threshold C, by the un-real time job tjTime factor be set to 0, do not allow described non-
Real-time task tjIt is entered the new task t of the task queuenewIt jumps the queue.
5. multi-task scheduling method as claimed in claim 4, which is characterized in that described to seize principle to described new based on resource
Task tnewOperation is seized with the task execution resource being currently running in the task queue, comprising:
Judge the new task tnewTask priority PnewIt is excellent with being currently running task in the task queue of the task
First grade PcurSize relation;
As the new task tnewTask priority and the task priority for being currently running task identical or less than it is described just
In the task priority of the task of operation, the new task t is not allowednewThe computing resource for being currently running task is not seized;
As the new task tnewTask priority PnewHigher than the task priority P for being carrying out taskcurWhen, judgement
(Ecur-ecur)+EnewWith dnew-AnewSize relation, wherein AnewIndicate the new task tnewSubmission time, EcurIt indicates
The average performance times for being currently running task, EnewIndicate the new task tnewAverage performance times, ecurDescribed in expression
It is currently running the time that task has been run, dnewIndicate the new task tnewThe absolute cutoff time limit;
If (Ecur-ecur)+Enew< dnew-Anew, do not allow the new task tnewThe calculating for being currently running task is not seized
Resource;
If (Ecur-ecur)≥Enew≥dnew-Anew, allow the new task tnewSeize the calculating money of being currently running for the task
Source.
6. a kind of multi-task scheduling electronic device, which is characterized in that the electronic device includes memory and processor, described to deposit
The multi-task scheduling multi-task scheduling program that can be run on the processor, the multi-task scheduling program are stored on reservoir
Following steps are realized when being executed by the processor:
Receive new task tnew, judge the new task tnewIt is made whether to meet task access principle;
When judging the new task tnewWhen being unsatisfactory for the task access principle, the new task t is not allowednewInto task team
Column;
When judging the new task tnewWhen meeting the task access principle, it is based on the new task tnewTask type by institute
State new task tnewIt is inserted into the task queue;
As the new task tnewAfter being inserted into the task queue, principle is seized to the new task t based on resourcenewWith described
The task execution resource being currently running in business queue seizes operation and priority reverses operation to obtain updated task team
Column;
Task schedule is carried out based on updated task queue.
7. multi-task scheduling device as claimed in claim 6, which is characterized in that the new task tnewIncluding one or more
Request operation;
The task queue stores waiting task ti, the waiting task tiIt is divided into real-time task and non-based on time factor
Real-time task;
The waiting task tiBy (Pi, di, Ai, Ti, Ei, idi) composition,
Wherein, AiFor the waiting task tiSubmission time, TiFor the waiting task tiDuty cycle, EiIndicate institute
State waiting task tiAverage performance times, PiIndicate the waiting task tiTask priority, diIndicate described wait locate
Reason task tiThe absolute cutoff time limit, idiRepresent the waiting task tiNumber.
8. multi-task scheduling device as claimed in claim 7, which is characterized in that the new task tnewCarry out task access
Judgement includes;
Judge whether the task queue has been stored with the new task tnew;
When the task queue is stored with the new task tnewWhen, do not allow the new task tnewIt repeats to enter the task team
Column;
As the not stored new task t of the task queuenewWhen, judge the task average performance times of the task queue with
The new task tnewDuty cycle size relation;
When the task average performance times of task queue are greater than the new task tnewDuty cycle when, do not allow the new post
Be engaged in tnewInto the task queue;
When the task average performance times of task queue are less than or equal to the new task tnewDuty cycle when, described in permission
New task tnewInto the task queue.
9. multi-task scheduling device as described in claim 6, which is characterized in that described to be based on the new task tnewAppoint
Service type is by the new task tnewIt is inserted into the task queue, comprising:
Each un-real time job t in the task queue is setjParameter Cj, CjIndicate real-time task tjThe number jumped the queue, and
Initialize the parameter CjIt is 0;
As the new task tnewWhen not being real-time task, by the new task tnewAppoint according to task priority insertion is described
Business queue;
As the new task tnewWhen for real-time task, by the new task tnewIt is preferential according to the time factor and the task
Grade is inserted into the task queue;
As the un-real time job tjIt is entered the new task t of the task queuenewWhen jumping the queue one time, the parameter CjIt is automatic to add
1;
As the parameter CjWhen greater than preset threshold C, by the un-real time job tjTime factor be set to 0, do not allow described non-
Real-time task tjIt is entered the new task t of the task queuenewIt jumps the queue.
10. a kind of computer readable storage medium, which is characterized in that be stored with multitask on the computer readable storage medium
Scheduler program, the multi-task scheduling program can be executed by one or more processor, to realize as in claim 1 to 5
The step of described in any item multi-task scheduling methods.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910699372.0A CN110427257A (en) | 2019-07-29 | 2019-07-29 | Multi-task scheduling method, device and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910699372.0A CN110427257A (en) | 2019-07-29 | 2019-07-29 | Multi-task scheduling method, device and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110427257A true CN110427257A (en) | 2019-11-08 |
Family
ID=68413239
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910699372.0A Pending CN110427257A (en) | 2019-07-29 | 2019-07-29 | Multi-task scheduling method, device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110427257A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111143045A (en) * | 2019-12-11 | 2020-05-12 | 青岛海尔科技有限公司 | Task scheduling method and device of smart home operating system and storage medium |
CN111260254A (en) * | 2020-02-19 | 2020-06-09 | 中国人民解放军国防科技大学 | Online scheduling method for dynamically arriving tasks |
CN111913790A (en) * | 2020-07-03 | 2020-11-10 | 佛山市顺德区美的洗涤电器制造有限公司 | Task scheduling method and device, household appliance and computer readable storage medium |
CN111949424A (en) * | 2020-09-18 | 2020-11-17 | 成都精灵云科技有限公司 | Method for realizing queue for processing declarative events |
CN112463339A (en) * | 2020-12-11 | 2021-03-09 | 北京浪潮数据技术有限公司 | Multitask scheduling method, system, equipment and storage medium |
CN112529470A (en) * | 2020-12-25 | 2021-03-19 | 深圳前海微众银行股份有限公司 | Task execution method, device, equipment and storage medium of website robot |
CN113051059A (en) * | 2021-04-10 | 2021-06-29 | 作业帮教育科技(北京)有限公司 | Multi-GPU task real-time scheduling method and device |
WO2021259108A1 (en) * | 2020-06-23 | 2021-12-30 | 中兴通讯股份有限公司 | Resource allocation method and apparatus, server, and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102654843A (en) * | 2011-03-04 | 2012-09-05 | 北京国科环宇空间技术有限公司 | Non-preemptive type fault-tolerant scheduling method in embedded processor and embedded processor |
US20150121387A1 (en) * | 2013-10-30 | 2015-04-30 | Mediatek Inc. | Task scheduling method for dispatching tasks based on computing power of different processor cores in heterogeneous multi-core system and related non-transitory computer readable medium |
CN105391649A (en) * | 2015-10-16 | 2016-03-09 | 浪潮(北京)电子信息产业有限公司 | Queue scheduling method and queue scheduling device |
CN108037983A (en) * | 2017-11-22 | 2018-05-15 | 链家网(北京)科技有限公司 | Method for scheduling task and distributed scheduling system in distributed scheduling system |
CN109324891A (en) * | 2017-07-31 | 2019-02-12 | 沈阳高精数控智能技术股份有限公司 | A kind of periodic duty low-power consumption scheduling method of ratio free time distribution |
-
2019
- 2019-07-29 CN CN201910699372.0A patent/CN110427257A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102654843A (en) * | 2011-03-04 | 2012-09-05 | 北京国科环宇空间技术有限公司 | Non-preemptive type fault-tolerant scheduling method in embedded processor and embedded processor |
US20150121387A1 (en) * | 2013-10-30 | 2015-04-30 | Mediatek Inc. | Task scheduling method for dispatching tasks based on computing power of different processor cores in heterogeneous multi-core system and related non-transitory computer readable medium |
CN105391649A (en) * | 2015-10-16 | 2016-03-09 | 浪潮(北京)电子信息产业有限公司 | Queue scheduling method and queue scheduling device |
CN109324891A (en) * | 2017-07-31 | 2019-02-12 | 沈阳高精数控智能技术股份有限公司 | A kind of periodic duty low-power consumption scheduling method of ratio free time distribution |
CN108037983A (en) * | 2017-11-22 | 2018-05-15 | 链家网(北京)科技有限公司 | Method for scheduling task and distributed scheduling system in distributed scheduling system |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111143045A (en) * | 2019-12-11 | 2020-05-12 | 青岛海尔科技有限公司 | Task scheduling method and device of smart home operating system and storage medium |
CN111143045B (en) * | 2019-12-11 | 2024-03-22 | 青岛海尔科技有限公司 | Task scheduling method and device of intelligent home operating system and storage medium |
CN111260254A (en) * | 2020-02-19 | 2020-06-09 | 中国人民解放军国防科技大学 | Online scheduling method for dynamically arriving tasks |
CN111260254B (en) * | 2020-02-19 | 2022-07-26 | 中国人民解放军国防科技大学 | Online scheduling method for dynamically arriving tasks |
WO2021259108A1 (en) * | 2020-06-23 | 2021-12-30 | 中兴通讯股份有限公司 | Resource allocation method and apparatus, server, and storage medium |
CN111913790A (en) * | 2020-07-03 | 2020-11-10 | 佛山市顺德区美的洗涤电器制造有限公司 | Task scheduling method and device, household appliance and computer readable storage medium |
CN111949424A (en) * | 2020-09-18 | 2020-11-17 | 成都精灵云科技有限公司 | Method for realizing queue for processing declarative events |
CN112463339A (en) * | 2020-12-11 | 2021-03-09 | 北京浪潮数据技术有限公司 | Multitask scheduling method, system, equipment and storage medium |
CN112529470A (en) * | 2020-12-25 | 2021-03-19 | 深圳前海微众银行股份有限公司 | Task execution method, device, equipment and storage medium of website robot |
CN112529470B (en) * | 2020-12-25 | 2024-03-19 | 深圳前海微众银行股份有限公司 | Task execution method, device and equipment of website robot and storage medium |
CN113051059A (en) * | 2021-04-10 | 2021-06-29 | 作业帮教育科技(北京)有限公司 | Multi-GPU task real-time scheduling method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110427257A (en) | Multi-task scheduling method, device and computer readable storage medium | |
KR100509794B1 (en) | Method of scheduling jobs using database management system for real-time processing | |
US9277003B2 (en) | Automated cloud workload management in a map-reduce environment | |
US7441240B2 (en) | Process scheduling apparatus, process scheduling method, program for process scheduling, and storage medium recording a program for process scheduling | |
US8869159B2 (en) | Scheduling MapReduce jobs in the presence of priority classes | |
CN104199730B (en) | A kind of single thread multi-task processing method based on synchronous i/o multiplexing mechanism | |
US20070044102A1 (en) | A Method and System for Performing Fair-Share Preemption | |
CN102779075A (en) | Method, device and system for scheduling in multiprocessor nuclear system | |
CN110221901A (en) | Container asset creation method, apparatus, equipment and computer readable storage medium | |
CN111782355B (en) | Cloud computing task scheduling method and system based on mixed load | |
CN110427252A (en) | Method for scheduling task, device and the storage medium of task based access control dependence | |
TW200401529A (en) | System and method for the allocation of grid computing workload to network workstations | |
CN103873587B (en) | A kind of method and device that scheduling is realized based on cloud platform | |
CN112286671B (en) | Containerization batch processing job scheduling method and device and computer equipment | |
CN110362391B (en) | Resource scheduling method and device, electronic equipment and storage medium | |
CN111694675B (en) | Task scheduling method and device and storage medium | |
CN112559159A (en) | Task scheduling method based on distributed deployment | |
CN115168013A (en) | Task running system and method and computing device | |
CN105808346A (en) | Task scheduling method and device | |
CN111597044A (en) | Task scheduling method and device, storage medium and electronic equipment | |
CN114327894A (en) | Resource allocation method, device, electronic equipment and storage medium | |
CN112073532B (en) | Resource allocation method and device | |
Marau et al. | Performing flexible control on low-cost microcontrollers using a minimal real-time kernel | |
CN108255595A (en) | A kind of dispatching method of data task, device, equipment and readable storage medium storing program for executing | |
CN109426556A (en) | A kind of process scheduling method and device |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191108 |