CN109917705A - A kind of multi-task scheduling method - Google Patents
A kind of multi-task scheduling method Download PDFInfo
- Publication number
- CN109917705A CN109917705A CN201910138564.4A CN201910138564A CN109917705A CN 109917705 A CN109917705 A CN 109917705A CN 201910138564 A CN201910138564 A CN 201910138564A CN 109917705 A CN109917705 A CN 109917705A
- Authority
- CN
- China
- Prior art keywords
- task
- scheduling
- resource
- conflict
- arbitration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Manipulator (AREA)
- General Factory Administration (AREA)
Abstract
The present invention relates to technical field of robot control more particularly to a kind of multi-task scheduling methods, are applied to a robot, include multiple resources in robot；Include: step S1, successively obtains the resource scheduling request of several tasks in order；Step S2 carries out conflict arbitration to the resource scheduling request currently obtained according to a conflict arbitration strategy, so that current resource scheduling request obtains a scheduling result；Step S3 executes the corresponding scheduling result of current resource scheduling request；Step S4 repeats step S2~S3, the corresponding scheduling result of resource scheduling request until obtaining and executing the last one task；Step S5 terminates；The multi-task scheduling under the complicated resource environment of robot can be completed, realizes to the optimum use of the resource of robot, is advantageously implemented the optimized control of robot, ensure that robot product forms the user experience of high quality.
Description
Technical field
The present invention relates to technical field of robot control more particularly to a kind of multi-task scheduling methods.
Background technique
Robot is the software and hardware system of a set of complexity, to allow its behavior expression obtain it is more intelligent, personalize, need
It is wanted to have a variety of base powers, for example locomitivity, visual capacity, listening and speaking ability, screen demonstrate one's ability, networked capabilities etc.,
And these base powers are effectively combined and form increasingly complex changeable robot application, and allow robot simultaneously
When completing multinomial application function, it is necessary to the scheduling system between a set of robot application and base power is developed, to solve
Conflict and sharing problem when multitask accesses resource simultaneously.
Multiple tasks dispatching system similar at present includes four classes:
The first kind is traditional computer operating system, such as windows, linux etc. and embedded real-time operation system
System, such as uCos, RTOS etc., this kind are traditional figure spirit framework computer operating systems, they to solve the problems, such as in it is most main
One wanted is the shared scheduling used for completing multiple application programs to the kernel resources of limited central processing unit, center
The principle of the heart is time-sharing multiplex, i.e. the rotation of the timeslice of central processing unit；Second class is the shape that robot operating system provides
State machine, it provides a kind of application development framework of state machine mechanism, is directed to the exploitation for solving complex machines people application, it is needed
Will clearly robot state and state switching definition；Third class is similar car entertainment interactive system this kind system
System, they provide it is limited apply between scheduling, such as navigation with music two application play sound simultaneously when, sound
The happy sound played can turn down volume temporarily to make the sound of prior navigation more prominent；4th class is to be widely used in interconnecting
Dispatching method in cluster resource scheduling in network server, when mainly solving concurrently to access, is scheduled limited resource
Problem.
But the multi-task scheduling ability of first kind traditional computer operating system is directed under intrinsic figure spirit framework
The scheduling of this resource of central processing unit, its core concept are the time-sharing multiplex of central processing unit, without handling multiple
Be engaged in the conflict access problems of multiple resources, thus it and do not have the scheduling energy being coupled by force between multitask and more resources
Power, and its dispatching method is not suitable for the variability of the resource of robot yet；What the second class robot operating system provided
State machine application and development mechanism, there is provided a set of more easily based on the application and development of state machine mechanism for the essence of this scheme
Tool develops the state definition and conversion for realizing robot behavior by this tool, therefore substantially it is not more than one
Business scheduling system and does not have Conflict solving ability when multitask accesses multiple resources simultaneously yet；Third class car entertainment is handed over
The volume automatic adjustment that mutual system provides, the multitask application scenarios and the scene of robot of this kind of system are the most similar, still
It is relatively easy due to application scenarios, what this function was usually the method by communicating between application to realize, such scheme
It is only suitable for that there is low coupling degree between task, for example only realizes the collision problem to volume between navigation and music,
When application task becomes more, when conflicting resource becomes more, this method is just not suitable for, because the high coupling degree between, is answered
With exploitation other than paying close attention to the service logic using itself, it is also necessary to the resource contention relationship between processing and other application,
The scalability of application is greatly reduced, maintenance cost is increased；In addition under the scene of some complexity, this task communication
Mode can not even solve wherein complicated conflict relationship；There is also the 4th class server cluster resource regulating method, such sides
The mode of method use priority carries out prioritizing to solve limited resource contention access problem, still to multiple tasks
Also due to the diversity of the resource of robot and the complexity of robot application, the single priority by application this
Dimension is no longer satisfied the demand of robot application exploitation.
Summary of the invention
In view of the above-mentioned problems, the invention proposes a kind of multi-task scheduling methods, wherein it is applied to a robot, it is described
It include multiple resources in robot；Include:
Step S1 successively obtains the resource scheduling request of several tasks in order；
Step S2 carries out conflict arbitration to the resource scheduling request currently obtained according to a conflict arbitration strategy, with
So that the current resource scheduling request obtains a scheduling result；
Step S3 executes the corresponding scheduling result of the current resource scheduling request；
Step S4, repeating said steps S2~S3, the resource tune until obtaining and executing the last one task
Degree requests the corresponding scheduling result；
Step S5 terminates.
Above-mentioned multi-task scheduling method, wherein it is excellent that the task each currently obtained all has a preset execution
First grade；
In the step S2, the conflict arbitration strategy includes that an execution priority judges substrategy；
The execution priority judges the judgment mechanism of substrategy specifically, judging the described of the task currently obtained
Whether execution priority is higher than other the described tasks being carrying out；
If so, the scheduling result is to be scheduled；If it is not, then the scheduling result is without scheduling.
Above-mentioned multi-task scheduling method, wherein in the step S2, if the execution for the task that judgement currently obtains
Priority be higher than be carrying out other described tasks when, then according to the sequence from big to small of the execution priority again according to
The secondary task that is carrying out to other carries out the conflict arbitration, holds described in the task that other are carrying out
Row major grade is the smallest to complete the conflict arbitration, forms a stable state.
Above-mentioned multi-task scheduling method, wherein in the step S1, each task is respectively from one and answers
With；
In the step S3, while executing the current resource scheduling request corresponding scheduling result, will also
The scheduling result is fed back in the corresponding application.
Above-mentioned multi-task scheduling method, wherein in the step S3, the scheduling result is without generating when scheduling
Have and the relevant conflicting information of the task of conflict；
When the scheduling result is without scheduling, also the conflicting information is fed back in the corresponding application.
Above-mentioned multi-task scheduling method, wherein it include at least one shared resource in the resource, it is each described shared
Resource has a control focus；
Each task corresponds to each shared resource with a preset Focus priority grade；
In the step S2, the conflict arbitration strategy includes that a Focus priority grade judges substrategy；
Wherein, the Focus priority grade judges the judgment mechanism of substrategy specifically, the task that judgement currently obtains
The Focus priority grade that has of the correspondence shared resource whether be higher than other the described tasks being carrying out；
If so, the scheduling result is to be scheduled；If it is not, then the scheduling result is without scheduling.
Above-mentioned multi-task scheduling method, wherein each shared resource is preset with a shared number of tasks amount upper limit.
Above-mentioned multi-task scheduling method, wherein in the step S4~S5, at the conflict of the task currently obtained
It manages in implementation procedure, when multiple tasks being currently running are affected, then the affected task according to institute
The sequence of the height of execution priority from big to small is stated to be put into Conflict Tasks queue, again successively to these by Conflict Tasks into
The row conflict arbitration, until being empty, one stable state of formation in the conflict queue.
Above-mentioned multi-task scheduling method, wherein in the step S2, it is secondary that the conflict is completed using a task dispatcher
It cuts out.
Above-mentioned multi-task scheduling method, wherein in the step S1, the resource scheduling request of several tasks
Equal serial arrangement is in a messaging list.
Above-mentioned multi-task scheduling method, wherein in the step S1, each task all has a new task category
Property；
In the step S3, when conflict generates, the conflict arbitration strategy includes a new task priority arbitration substrategy；
The arbitration mechanism of the new task priority arbitration substrategy are as follows:
Judge whether the execution priority of the task currently obtained is equal to other the described tasks being carrying out,
And whether the task currently obtained, which is that user is newest, is assigned needing to be implemented for task；
If being, the scheduling result is to be scheduled；If it is any be it is no, the scheduling result be without adjust
Degree.
Above-mentioned multi-task scheduling method, wherein the task each currently obtained has pre-set needs and accounted for
Whether resource is the core resource of the corresponding task, and locates the non-core resource other than the core resource；
The conflict arbitration strategy includes a core resource conflict arbitration substrategy；
The arbitration mechanism of the core resource arbitration substrategy are as follows:
Judge whether institute's core resource in need of the task currently obtained has got the right to use；
If so, no matter whether the non-core resource that other tasks need gets the right to use, all will
The scheduling result of presently described task is arranged for dispatching.
Above-mentioned multi-task scheduling method, wherein each described resource all has a preset shared attribute
Value, for indicating whether the corresponding resource can be by the shared use of multiple tasks.
The utility model has the advantages that a kind of multi-task scheduling method proposed by the present invention, can complete the complicated resource environment of robot
Under multi-task scheduling, realize to the optimum use of the resource of robot, be advantageously implemented the optimized control of robot, protect
The user experience that robot product forms high quality is demonstrate,proved.
Detailed description of the invention
Fig. 1 is the step flow chart of multi-task scheduling method in one embodiment of the invention.
Specific embodiment
Invention is further explained with reference to the accompanying drawings and examples.
In a preferred embodiment, as shown in Figure 1, proposing a kind of multi-task scheduling method, wherein can apply
It include multiple resources in a robot, robot；The multi-task scheduling method may include:
Step S1 successively obtains the resource scheduling request of several tasks in order；
Step S2 carries out conflict arbitration to the resource scheduling request currently obtained according to a conflict arbitration strategy, so that
Current resource scheduling request obtains a scheduling result；
Step S3 executes the corresponding scheduling result of current resource scheduling request；
Step S4 repeats step S2~S3, until the resource scheduling request for obtaining and executing the last one task is corresponding
Scheduling result；
Step S5 terminates.
In above-mentioned technical proposal, since the resource environment to be dispatched of robot is increasingly complex, robot is carrying out
When multiple tasks are dispatched, it is necessary to conflict arbitration is carried out to the resource scheduling request of each task, thus in multiple tasks
Middle the determining to need preferentially to execute of the task, or shared task can be carried out to resource, thus under guaranteeing multitask environment
The stability and reliability of robot.
In a preferred embodiment, each currently obtaining for task all has a preset execution priority；
In step S2, conflict arbitration strategy includes that an execution priority judges substrategy；
Execution priority judges the judgment mechanism of substrategy specifically, the execution priority for the task that judgement currently obtains is
It is no higher than other tasks being carrying out；
If so, scheduling result is to be scheduled；If it is not, then scheduling result is without scheduling.
In above-mentioned technical proposal, for example, assuming that currently obtain is task A, that be carrying out is task B, if
The value of the execution priority of task A is 50, and the value of the execution priority of task B is 35, then may determine that the execution of task A is preferential
Grade is higher than task B, and scheduling result at this time, which should be, to be scheduled, and after executing the scheduling result, task A can interrupt task B
Preferential to execute, task B then delays execution；Scheduling result should have at least two states, that is, be scheduled state and without adjusting
Degree state.
In above-described embodiment, it is preferable that in step S2, if the execution priority for the task that judgement currently obtains is higher than
When other tasks executed, then according to the sequence from big to small of execution priority again successively to the task that other are carrying out
Conflict arbitration is carried out, until the smallest completion conflict arbitration of execution priority in the task that other are carrying out, forms a stable state.
In above-mentioned technical proposal, after completing above-mentioned conflict arbitration circulation, each task is no longer influenced by influence, at this time may be used
To think to form stable state or equilibrium, the process of this circulation can be considered as the process of recursion cycle.
In above-described embodiment, it is preferable that in step S1, each task is respectively from an application；
In step S3, while executing current resource scheduling request corresponding scheduling result, also scheduling result is fed back
To in corresponding application.
In above-mentioned technical proposal, using this may include: background application, automatically recharge application, dancing control application, voice
Chat application, security protection alarm application etc..
In above-described embodiment, it is preferable that in step S3, scheduling result is without generating having with conflict for task when scheduling
Relevant conflicting information；
When scheduling result is without scheduling, also conflicting information is fed back in corresponding application.
In above-mentioned technical proposal, the application for receiving feedback can be analyzed according to conflicting information, and selection, which re-initiates, appoints
The opportunity of business.
It in a preferred embodiment, include at least one shared resource in resource, each shared resource has a control
System power focus；
Each task corresponds to each shared resource with a preset Focus priority grade；
In step S2, conflict arbitration strategy includes that a Focus priority grade judges substrategy；
Wherein, Focus priority grade judges the judgment mechanism of substrategy specifically, the correspondence for the task that judgement currently obtains is total
Whether the Focus priority grade that enjoying resource has is higher than other tasks being carrying out；
If so, scheduling result is to be scheduled；If it is not, then scheduling result is without scheduling.
In above-mentioned technical proposal, due to being directed to shared resource, it is assumed that the Focus priority grade of the task C currently obtained is high
In the task D being carrying out, then scheduling result, which should be, is scheduled, and task D is not off after execution, but shared with task C
At least one same resource, but task C obtains the control focus of the resource at this time；For example, the control of radio
Focus is the volume of radio, and Focus priority, which weighs big task, can occupy the larger volume of this resource of radio, but not
The relatively small volume that other tasks being carrying out occupy radio can be interrupted.
In above-described embodiment, it is preferable that each shared resource is preset with a shared number of tasks amount upper limit, to avoid excessive
Task occupy same resource.
In above-described embodiment, it is preferable that in step S4~S5, in the clash handle implementation procedure of currently obtaining for task,
When multiple being currently running for tasks are affected, then affected task according to execution priority height from big to small
Sequence be put into Conflict Tasks queue, again successively to these by Conflict Tasks carry out conflict arbitration, until conflict queue in
For sky, a stable state is formed.
In a preferred embodiment, in step S2, conflict arbitration can be completed using a task dispatcher, it is independent
Task dispatcher can be convenient for that conflict resolving strategy is adjusted or is replaced.
In a preferred embodiment, in step S1, the resource scheduling requests of several tasks can serial arrangement exist
In one messaging list.
In a preferred embodiment, in step S1, each task all has a new task attribute；
In step S3, when conflict generates, conflict arbitration strategy includes a new task priority arbitration substrategy；
The arbitration mechanism of new task priority arbitration substrategy are as follows:
Judge whether the execution priority of currently obtaining for task is equal to other tasks being carrying out, and currently obtains
Task whether be that user is newest assign the needing to be implemented of the task；
If being, scheduling result is to be scheduled；If it is any be it is no, scheduling result be without scheduling.
In a preferred embodiment, each currently obtaining for task has pre-set the resource that needs occupy
Non-core resource other than the no core resource for corresponding task, and place's core resource；
Conflict arbitration strategy includes a core resource conflict arbitration substrategy；
The arbitration mechanism of core resource arbitration substrategy are as follows:
Judge whether institute's core resource in need of currently obtaining for task has got the right to use；
If so, no matter whether the non-core resource that other tasks need gets the right to use, all by the tune of current task
Degree result is arranged for dispatching.
In a preferred embodiment, each resource all has a preset shared attribute value, for referring to
Whether bright corresponding resource can be by the shared use of multiple tasks.
Specifically, above-mentioned several embodiments can also be combined with each other, and form occupation condition as shown in the table:
In upper table, left side one is classified as resource, one behavior task of upside/application.
In conclusion a kind of multi-task scheduling method proposed by the present invention, is applied to a robot, it include more in robot
A resource；Include: step S1, successively obtains the resource scheduling request of several tasks in order；Step S2, it is secondary according to a conflict
It cuts out strategy and conflict arbitration is carried out to the resource scheduling request currently obtained, so that current resource scheduling request obtains a scheduling
As a result；Step S3 executes the corresponding scheduling result of current resource scheduling request；Step S4 repeats step S2~S3, until
To and execute the corresponding scheduling result of resource scheduling request of the last one task；Step S5 terminates；It can complete robot
Multi-task scheduling under complicated resource environment is realized to the optimum use of the resource of robot, is advantageously implemented robot
Optimized control ensure that robot product forms the user experience of high quality.
By description and accompanying drawings, the exemplary embodiments of the specific structure of specific embodiment are given, based on present invention essence
Mind can also make other conversions.Although foregoing invention proposes existing preferred embodiment, however, these contents are not intended as
Limitation.
For a person skilled in the art, after reading above description, various changes and modifications undoubtedly be will be evident.
Therefore, appended claims should regard the whole variations and modifications for covering true intention and range of the invention as.It is weighing
The range and content of any and all equivalences, are all considered as still belonging to the intent and scope of the invention within the scope of sharp claim.
Claims (13)
1. a kind of multi-task scheduling method, which is characterized in that be applied to a robot, include multiple resources in the robot；
Include:
Step S1 successively obtains the resource scheduling request of several tasks in order；
Step S2 carries out conflict arbitration to the resource scheduling request currently obtained according to a conflict arbitration strategy, so that
The current resource scheduling request obtains a scheduling result；
Step S3 executes the corresponding scheduling result of the current resource scheduling request；
Step S4, repeating said steps S2~S3, the scheduling of resource until obtaining and executing the last one task are asked
Seek the corresponding scheduling result；
Step S5 terminates.
2. multi-task scheduling method according to claim 1, which is characterized in that the task each currently obtained has
There is a preset execution priority；
In the step S2, the conflict arbitration strategy includes that an execution priority judges substrategy；
The execution priority judges the judgment mechanism of substrategy specifically, judging the execution of the task currently obtained
Whether priority is higher than other the described tasks being carrying out；
If so, the scheduling result is to be scheduled；If it is not, then the scheduling result is without scheduling.
3. multi-task scheduling method according to claim 2, which is characterized in that in the step S2, if judgement currently obtains
When the execution priority of the task taken is higher than other the described tasks being carrying out, then according to the execution priority from
Small sequence is arrived greatly, the conflict arbitration successively is carried out to the task that other are carrying out again, until other are carrying out
The task described in execution priority is the smallest completes the conflict arbitration, form a stable state.
4. multi-task scheduling method according to claim 2, which is characterized in that in the step S1, each task
It is respectively from an application；
It, will also be described while executing the current resource scheduling request corresponding scheduling result in the step S3
Scheduling result is fed back in the corresponding application.
5. multi-task scheduling method according to claim 4, which is characterized in that in the step S3, the scheduling result
To have and the relevant conflicting information of the task of conflict without being generated when scheduling；
When the scheduling result is without scheduling, also the conflicting information is fed back in the corresponding application.
6. multi-task scheduling method according to claim 1, which is characterized in that shared including at least one in the resource
Resource, each shared resource have a control focus；
Each task corresponds to each shared resource with a preset Focus priority grade；
In the step S2, the conflict arbitration strategy includes that a Focus priority grade judges substrategy；
Wherein, the Focus priority grade judges the judgment mechanism of substrategy specifically, judging pair of the task currently obtained
Whether the Focus priority grade for answering the shared resource to have is higher than other the described tasks being carrying out；
If so, the scheduling result is to be scheduled；If it is not, then the scheduling result is without scheduling.
7. multi-task scheduling method according to claim 6, which is characterized in that each shared resource is preset with altogether
Enjoy task the upper limit of the number.
8. multi-task scheduling method according to claim 2, which is characterized in that current to obtain in the step S4~S5
The task clash handle implementation procedure in, when multiple tasks being currently running are affected, then by shadow
The loud task is put into Conflict Tasks queue according to the height sequence from big to small of the execution priority, again successively
The conflict arbitration is carried out by Conflict Tasks to these, until being empty, one stable state of formation in the conflict queue.
9. multi-task scheduling method according to claim 1, which is characterized in that in the step S2, using a task tune
It spends device and completes the conflict arbitration.
10. multi-task scheduling method according to claim 1, which is characterized in that in the step S1, several tasks
The equal serial arrangement of the resource scheduling request in a messaging list.
11. multi-task scheduling method according to claim 1, which is characterized in that in the step S1, each task
All have a new task attribute；
In the step S3, when conflict generates, the conflict arbitration strategy includes a new task priority arbitration substrategy；
The arbitration mechanism of the new task priority arbitration substrategy are as follows:
Judge whether the execution priority of the task currently obtained is equal to other the described tasks being carrying out, and
Whether the task currently obtained, which is that user is newest, is assigned needing to be implemented for task；
If being, the scheduling result is to be scheduled；If it is any be it is no, the scheduling result be without scheduling.
12. multi-task scheduling method according to claim 1, which is characterized in that the task each currently obtained is equal
Pre-set need the resource that occupies whether be the corresponding task core resource, and place's core resource with
Outer non-core resource；
The conflict arbitration strategy includes a core resource conflict arbitration substrategy；
The arbitration mechanism of the core resource arbitration substrategy are as follows:
Judge whether institute's core resource in need of the task currently obtained has got the right to use；
If so, no matter whether the non-core resource that other tasks need gets the right to use, it all will be current
The scheduling result of the task is arranged for dispatching.
13. multi-task scheduling method according to claim 1, which is characterized in that it is pre- that each described resource all has one
The shared attribute value first set, for indicating whether the corresponding resource can be by the shared use of multiple tasks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910138564.4A CN109917705B (en) | 2019-02-25 | 2019-02-25 | Multi-task scheduling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910138564.4A CN109917705B (en) | 2019-02-25 | 2019-02-25 | Multi-task scheduling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109917705A true CN109917705A (en) | 2019-06-21 |
| CN109917705B CN109917705B (en) | 2021-10-22 |
Family
ID=66962268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910138564.4A Active CN109917705B (en) | 2019-02-25 | 2019-02-25 | Multi-task scheduling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109917705B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111176819A (en) * | 2019-12-31 | 2020-05-19 | 苏州长光华医生物医学工程有限公司 | Sample test pipeline task scheduling method, system and computer storage medium |
| CN111399451A (en) * | 2020-03-17 | 2020-07-10 | 广东博智林机器人有限公司 | Conflict resolution method, device and storage medium based on scheduling system |
| CN111390905A (en) * | 2020-03-18 | 2020-07-10 | 深圳市人工智能与机器人研究院 | Robot multitask control method and device and terminal equipment |
| CN111857995A (en) * | 2020-06-30 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Process scheduling method and device, storage medium and electronic device |
| CN112053066A (en) * | 2020-09-09 | 2020-12-08 | 上海有个机器人有限公司 | Multi-task scheduling method and device for robot and robot |
| CN111857995B (en) * | 2020-06-30 | 2024-05-24 | 海尔优家智能科技（北京）有限公司 | Process scheduling method and device, storage medium and electronic device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102298539A (en) * | 2011-06-07 | 2011-12-28 | 华东师范大学 | Method and system for scheduling shared resources subjected to distributed parallel treatment |
| CN104503948A (en) * | 2015-01-19 | 2015-04-08 | 中国人民解放军国防科学技术大学 | Tightly coupled self-adaptive co-processing system supporting multi-core network processing framework |
| CN104822163A (en) * | 2014-02-03 | 2015-08-05 | 马维尔国际有限公司 | General scheduling system and method used for single-antenna multi-mode multi-card terminal |
| CN105873233A (en) * | 2016-03-29 | 2016-08-17 | 西安电子科技大学 | IEEE802.11ax access enhancement method based on hierarchical scheduling |
| CN106341483A (en) * | 2016-09-28 | 2017-01-18 | 深圳市普渡科技有限公司 | Local area network multicast group-based distributed dispatching system communication method |
| CN107977268A (en) * | 2017-10-13 | 2018-05-01 | 北京百度网讯科技有限公司 | Method for scheduling task, device and the computer-readable recording medium of the isomerization hardware of artificial intelligence |
| CN108123987A (en) * | 2016-11-30 | 2018-06-05 | 华为技术有限公司 | The method and device of master scheduler is determined from cloud computing system |
| CN108446852A (en) * | 2018-03-23 | 2018-08-24 | 西北工业大学 | A kind of multiagent schedule fragments formula coordination optimizing method |
| WO2018206994A1 (en) * | 2017-05-11 | 2018-11-15 | Bull Sas | Method of managing resource providing in a computers cluster running jobs |
-
2019
- 2019-02-25 CN CN201910138564.4A patent/CN109917705B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102298539A (en) * | 2011-06-07 | 2011-12-28 | 华东师范大学 | Method and system for scheduling shared resources subjected to distributed parallel treatment |
| CN104822163A (en) * | 2014-02-03 | 2015-08-05 | 马维尔国际有限公司 | General scheduling system and method used for single-antenna multi-mode multi-card terminal |
| CN104503948A (en) * | 2015-01-19 | 2015-04-08 | 中国人民解放军国防科学技术大学 | Tightly coupled self-adaptive co-processing system supporting multi-core network processing framework |
| CN105873233A (en) * | 2016-03-29 | 2016-08-17 | 西安电子科技大学 | IEEE802.11ax access enhancement method based on hierarchical scheduling |
| CN106341483A (en) * | 2016-09-28 | 2017-01-18 | 深圳市普渡科技有限公司 | Local area network multicast group-based distributed dispatching system communication method |
| CN108123987A (en) * | 2016-11-30 | 2018-06-05 | 华为技术有限公司 | The method and device of master scheduler is determined from cloud computing system |
| WO2018206994A1 (en) * | 2017-05-11 | 2018-11-15 | Bull Sas | Method of managing resource providing in a computers cluster running jobs |
| CN107977268A (en) * | 2017-10-13 | 2018-05-01 | 北京百度网讯科技有限公司 | Method for scheduling task, device and the computer-readable recording medium of the isomerization hardware of artificial intelligence |
| CN108446852A (en) * | 2018-03-23 | 2018-08-24 | 西北工业大学 | A kind of multiagent schedule fragments formula coordination optimizing method |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111176819A (en) * | 2019-12-31 | 2020-05-19 | 苏州长光华医生物医学工程有限公司 | Sample test pipeline task scheduling method, system and computer storage medium |
| CN111176819B (en) * | 2019-12-31 | 2021-07-09 | 苏州长光华医生物医学工程有限公司 | Sample test pipeline task scheduling method, system and computer storage medium |
| CN111399451A (en) * | 2020-03-17 | 2020-07-10 | 广东博智林机器人有限公司 | Conflict resolution method, device and storage medium based on scheduling system |
| CN111399451B (en) * | 2020-03-17 | 2021-06-22 | 广东博智林机器人有限公司 | Conflict resolution method, device and storage medium based on scheduling system |
| CN111390905A (en) * | 2020-03-18 | 2020-07-10 | 深圳市人工智能与机器人研究院 | Robot multitask control method and device and terminal equipment |
| CN111857995A (en) * | 2020-06-30 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Process scheduling method and device, storage medium and electronic device |
| CN111857995B (en) * | 2020-06-30 | 2024-05-24 | 海尔优家智能科技（北京）有限公司 | Process scheduling method and device, storage medium and electronic device |
| CN112053066A (en) * | 2020-09-09 | 2020-12-08 | 上海有个机器人有限公司 | Multi-task scheduling method and device for robot and robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109917705B (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109917705A (en) | A kind of multi-task scheduling method | |
| US11669372B2 (en) | Flexible allocation of compute resources | |
| CN108563500A (en) | Method for scheduling task, cloud platform based on cloud platform and computer storage media | |
| US6237058B1 (en) | Interrupt load distribution system for shared bus type multiprocessor system and interrupt load distribution method | |
| CN103189843B (en) | For the method and apparatus managing software application | |
| CN102469126B (en) | Application scheduling system, method thereof and related device | |
| CN105320570B (en) | Method for managing resource and system | |
| CN101996099A (en) | Method and system for processing information | |
| CN110187956B (en) | Layered real-time task scheduling method and system for multi-agent platform | |
| CN108762687A (en) | I/O service quality control method, device, equipment and storage medium | |
| CN112217676A (en) | Kubernetes container cluster node selection method based on mixed element heuristic algorithm | |
| WO2021180092A1 (en) | Task dispatching method and apparatus | |
| CN111258735A (en) | Deep learning task scheduling method supporting QoS (quality of service) perception of user | |
| Smys et al. | Performance evaluation of game theory based efficient task scheduling for edge computing | |
| CN112162835A (en) | Scheduling optimization method for real-time tasks in heterogeneous cloud environment | |
| CN109522090A (en) | Resource regulating method and device | |
| CN109857535A (en) | The implementation method and device of task priority control towards Spark JDBC | |
| CN103548324A (en) | Method for virtual machine assignment and device for virtual machine assignment | |
| CN106293917A (en) | The optimization method of a kind of I O scheduling cfq algorithm and system | |
| CN106900007B (en) | Overlapping clustering-oriented complex task allocation method for wireless sensor network | |
| CN111949398A (en) | Resource scheduling method and device | |
| CN111400021B (en) | Deep learning method, device and system | |
| CN115129458B (en) | Container-based process scheduling method, device, equipment and storage medium | |
| Ramamritham et al. | Meta-Level Control in Distributed Real-Time Systems. | |
| CN100576175C (en) | The parallel executing method and the system that are used for a plurality of kernels |
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 |