CN102915254B - task management method and device - Google Patents
task management method and device Download PDFInfo
- Publication number
- CN102915254B CN102915254B CN201110219354.1A CN201110219354A CN102915254B CN 102915254 B CN102915254 B CN 102915254B CN 201110219354 A CN201110219354 A CN 201110219354A CN 102915254 B CN102915254 B CN 102915254B
- Authority
- CN
- China
- Prior art keywords
- task
- message queue
- priority
- distributed message
- system resource
- 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.)
- Expired - Fee Related
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
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
Abstract
The invention discloses a kind of task management method and device.This method includes：Server end obtains the precedence information of task, is stored task into the Distributed Message Queue of respective priority according to precedence information；Corresponding task is recalled and distributes corresponding system resource by server end according to system resource free message from Distributed Message Queue.By means of technical scheme, server end is set to preserve and run largely being performed by conventional client management for task, enabling to need the long period to pay close attention to and take, the tasks of a large amount of client resources stablizes operates in server end, makes all tasks from be created to can be by efficient, rational execution in the life cycle that scheduling completes.
Description
Technical field
The present invention relates to computer realm, more particularly to a kind of task management method and device.
Background technology
In the prior art, traditional task management mode is that software client is installed on local computing, and task is opened
Begin, perform and terminate artificially to be controlled management by local.If task execution time is long, for example, to Internet resources
Download, occupancy local central processing unit (CentralProcessing Unit, referred to as CPU) that the task may be lasting,
The resource such as internal memory and network bandwidth.
Traditional Portable Batch System relatively focuses on task scheduling algorithm, by more optimized algorithm come high efficiency to a large amount of
Concurrent tasks are handled in real time.Above-mentioned processing mode make it that the quantity size of task is relatively small, and the operation of task is general
It is scheduled by operating system, limits the scope in task source in form.
With the development of the high performance supercomputer such as the Milky Way, nebula, server hardware disposal ability is led progressively to strengthen,
The computing capability that the research and development of cloud computing technology simultaneously, large-scale data center or server can provide is more and more stronger.
Therefore, it is badly in need of a kind of mission bit stream dispatching method at present, server end is preserved and run largely by traditional client end pipe
Performing for task is managed, what the enabling to need the long period to pay close attention to and take a large amount of client resources of the task was stablized operates in service
Device end.
The content of the invention
The present invention provides a kind of task management method and device, server end is preserved and run largely by traditional visitor
The task that family end pipe reason performs, the operation for enabling task of needing the long period to pay close attention to and take a large amount of client resources to stablize
In server end.
The present invention provides a kind of task management method, including：
Server end obtains the precedence information of task, and task storage is arrived to point of respective priority according to precedence information
In cloth message queue；
Corresponding task is recalled and distributes phase by server end according to system resource free message from Distributed Message Queue
The system resource answered.
Present invention also offers a kind of task management device, positioned at server end, including：
Task collection module, for obtaining the precedence information of task, task is sent to accordingly according to precedence information
In the Distributed Message Queue of priority；
Distributed Message Queue, for storing appointing for respective priority in the Distributed Message Queue of different priorities
Business；
Task scheduling modules, for being recalled from Distributed Message Queue corresponding task according to system resource free message
And distribute corresponding system resource.
The present invention has the beneficial effect that：
By the way that task storage is scheduled and system resource allocation into the Distributed Message Queue of respective priority, make
Server end can preserve and run largely being performed by conventional client management for task, make to need the long period to pay close attention to and take
What the task of a large amount of client resources can be stablized operates in server end, makes all tasks from the life for being created to scheduling completion
Can be by efficient, rational execution in cycle.
Brief description of the drawings
Fig. 1 is the flow chart of the task management method of the embodiment of the present invention；
Fig. 2 is the schematic diagram of the optimum decision system composition of the task management method of the embodiment of the present invention；
Fig. 3 is the preferred process flow chart of the task management method of the embodiment of the present invention；
Fig. 4 is the flow chart of the detailed processing of the task management method of the embodiment of the present invention；
Fig. 5 is the structural representation of the task management device of the embodiment of the present invention.
Embodiment
The invention provides a kind of task management method and device, server end is preserved and is run largely by tradition
The task that client-side management performs, the fortune for enabling task of needing the long period to pay close attention to and take a large amount of client resources to stablize
Go in server end, to realize that the safety of server end task is stored and dispatched.The technical scheme of the embodiment of the present invention includes will
The flow of task deposit Distributed Message Queue and the flow that task is removed to queue by scheduling strategy.Below in conjunction with accompanying drawing with
And embodiment, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only solving
The present invention is released, does not limit the present invention.
Embodiment of the method
According to an embodiment of the invention, there is provided a kind of task management method, Fig. 1 are the task managements of the embodiment of the present invention
The flow chart of method, as shown in figure 1, task management method according to embodiments of the present invention includes following processing：
Step 101, server end obtains the precedence information of task, is stored task to corresponding excellent according to precedence information
In the Distributed Message Queue of first level；It should be noted that above-mentioned Distributed Message Queue is configurable to persistence energy
The storage message queue of power, non-persistentization storage message queue can also be configured to, if Distributed Message Queue be configured to
Storage message queue with persistence ability, it can prevent the task in queue from losing in exception or in the case of being anticipated that
Lose.In addition, storage container of the above-mentioned distributed message team as mission bit stream object, is distributed multiple preferential with specifying
The fifo queue (that is, above-mentioned Distributed Message Queue can be the Distributed Message Queue of first in first out) of level；
Also, above-mentioned distributed message team can realize active push function, when receiving scheduling signals, by the head of the queue of scheduled queue
Information extrudes queue, or the head of the queue task of high-priority queue is actively sent into server end.
Specifically, in a step 101, when server end obtains the precedence information of task, following processing is specifically included：1、
In the case where task does not carry precedence information, server end distributes corresponding according to the characteristic information of task for task
Priority, and obtain corresponding precedence information；2nd, in the case where task carries precedence information, server end directly obtains
Take the precedence information carried in task.
That is, in a step 101, task is added respective priority by server end by a kind of priority policy
Queue.
Step 102, server end recalls from Distributed Message Queue corresponding task according to system resource free message
And distribute corresponding system resource.
Step 102 specifically includes following processing：1st, determining that system resource is able to carry out according to system resource free message
In the case of task, server end notifies corresponding Distributed Message Queue, receives being located at for Distributed Message Queue push
The first task of queue, and distribute corresponding system resource and perform the task；Or server end is actively from distributed message team
The queue first place acquisition task of the high priority Distributed Message Queue of row, and distribute corresponding system resource and perform the task；
2nd, determining that system resource can not be in the case of execution task according to system resource free message, server end isolation is distributed
Message queue, forbid receiving the task of Distributed Message Queue push.
That is, when required by task resource can obtain, server end distribution of notifications formula message queue, distribution disappears
Breath queue according to priority pushes task, and server end distribution respective resources perform the task；When required by task resource is all occupied
Used time, no longer server end isolation message queue, receiving queue task.When resource space idle, server end can also actively from
Acquisition task in a certain high-priority queue.
In actual applications, in the case where needing to perform high-priority task, server end is low excellent by what is currently performed
The task status of first level task is wait state from running state conversion, is again stored low priority task according to task status
Into the Distributed Message Queue of respective priority, and perform high-priority task.
Preferably, before step 102 is performed, server end can be by the Distributed Message Queue of limit priority
Task is stored into other Distributed Message Queues, and the high-priority task storage for being currently needed for performing is arrived into limit priority
Distributed Message Queue Head-of-line, then by store in other Distributed Message Queues task storage to highest it is preferential
In the Distributed Message Queue of level behind high-priority task.That is, the above-mentioned processing of the embodiment of the present invention passes through one
Kind replacement policy, can preferentially ensure that high-priority task seizes execution.
After step 102, server end can recalculate current execution according to the current running status for performing task
The priority of task, the current priority for performing task is improved or reduced according to result of calculation, and perform task according to current
Priority the current task status for performing task is set；In the case where task status is set to wait for state, server end
Again task storage is performed into the Distributed Message Queue of respective priority by current, running status is arranged in task status
In the case of, server end continues executing with current execution task.
The above-mentioned technical proposal of the embodiment of the present invention is described in detail below in conjunction with accompanying drawing.
Fig. 2 is the schematic diagram of the optimum decision system composition of the task management method of the embodiment of the present invention, as shown in Fig. 2 at this
In example, five resident Distributed Message Queues with different priorities can be created, i.e. Distributed Message Queue -1, divide
Cloth message queue -2, Distributed Message Queue -3, Distributed Message Queue -4, Distributed Message Queue -5；In practical application
In, naturally it is also possible to more Distributed Message Queues for being adapted to application are created, can also be instant according to the characteristic attribute of task
Establishment respective priority Distributed Message Queue, Distributed Message Queue of the destruction without task that can also be at any time make clothes
The scheduling search at business device end maintains a less scope, improves the efficiency of response.As shown in Fig. 2 task source -1, task
Source -2 ..., task source-N caused task is sent to server end, server end carries out task collection, then, according to
The precedence information of task, task is respectively stored in corresponding Distributed Message Queue, finally, server end is according to system
The idle condition of resource, task scheduling is carried out from Distributed Message Queue.
Fig. 3 is the preferred process flow chart of the task management method of the embodiment of the present invention, as shown in figure 3, server end enters
After row task is collected, according to the high, medium and low precedence information of task, task is respectively stored in high priority distributed message
In queue, middle priority Distributed Message Queue and low priority Distributed Message Queue, then, server end is according to being
The idle condition for resource of uniting, carries out task scheduling from Distributed Message Queue.It should be noted that this example entering task
The processing of Distributed Message Queue includes two kinds of situations：Situation one, server end are stored in queue after task priority is confirmed
Task；2nd, task scheduling modules allow high-priority task to swap out the task of low priority, by the task status of task from operation
State Transferring is wait state and is reentered into Distributed Message Queue (that is, the flow shown in dotted arrow).In addition, server
The preferential of task can also be improved or be reduced in end according to the running status of task by weighting, calculating the methods of run time
Level, change task status.
Fig. 4 is the flow chart of the detailed processing of the task management method of the embodiment of the present invention, as shown in figure 4, including as follows
Processing：
Step 1, task source creates task, and task call server end carries out the interface of task collection；
Step 2, the task that received server-side task source is sent, and the priority of task is analyzed, obtain task
Precedence information；
Step 3, server end stores task into different Distributed Message Queues according to the precedence information of task；
Step 4, Distributed Message Queue active push task is to server end；
Step 5, the task of low priority can be stored distribution again and disappeared by server end according to the priority of task
Cease in queue, and the task scheduling of high priority is come out；
Step 6, task distributing system resource of the server end for scheduling out；
Step 7, server end performs the task using system resource.
In summary, the task that task source creates is stored in phase by the technical scheme of the embodiment of the present invention by assigned priority
In the Distributed Message Queue for answering priority, task can be pushed dequeue and be scheduled for performing by Distributed Message Queue, also may be used
To wait task to be scheduled dequeue.Server end is enabled to respond the task storage of the big quantity of task source initiation and adjust
Degree, make all tasks from be created to scheduling complete life cycle in can be by efficient, rational execution.In addition, the present invention is real
The push function that example also flexibly uses Distributed Message Queue is applied, enables task actively to attempt to apply for that resource performs, makes task
Real-time be further enhanced；The embodiment of the present invention is by the way that the flow of task to be summarized as being stored in queue and recalling queue holding
OK, become apparent from the management process of task, there is more preferable autgmentability and exploitativeness so that be existing and following be likely to occur
Task storage and management realization can be more flexible.
Device embodiment
According to an embodiment of the invention, there is provided a kind of task management device, Fig. 5 are the task managements of the embodiment of the present invention
The structural representation of device, as shown in figure 5, task management device according to embodiments of the present invention includes：Task collection module 50,
The modules of the embodiment of the present invention are described in detail by Distributed Message Queue 52, task scheduling modules 54 below.
Task collection module 50, for obtaining the precedence information of task, task is sent to by phase according to precedence information
In the Distributed Message Queue 52 for answering priority；
Task collection module 50 is specifically used for：In the case where task does not carry precedence information, according to the spy of task
Reference ceases, and distributes corresponding priority for task, and obtain corresponding precedence information；Precedence information is carried in task
In the case of, the precedence information that is carried in direct acquisition task；That is, task collection module 50 passes through a kind of priority plan
Slightly, task is added into respective priority queue.
That is, task collection module 50 is as the hinge mould between task creation source and Distributed Message Queue 52
Block, it is responsible for distributing priority or the priority carried according to task after the task analysis received its characteristic information, by task
It is sent to the distributed queue of respective priority；Assessment of tasks priority can be stored in Distributed Message Queue 52.
Distributed Message Queue 52, for storing appointing for respective priority in the Distributed Message Queue of different priorities
Business；
It should be noted that above-mentioned Distributed Message Queue 52 is configurable to the storage message team with persistence ability
Row, non-persistentization storage message queue can also be configured to, if Distributed Message Queue 52 is configured to persistence energy
The storage message queue of power, it can prevent the task in queue from losing in exception or in the case of being anticipated that.In addition, above-mentioned point
Storage container of the cloth message queue 52 as mission bit stream object, it is distributed multiple advanced elder generations with assigned priority
Dequeue (that is, above-mentioned Distributed Message Queue can be the Distributed Message Queue of first in first out)；Also, above-mentioned point
Cloth message queue 52 can realize active push function, and when receiving scheduling signals, the head of the queue information of scheduled queue is squeezed
Dequeue, or the head of the queue task of high-priority queue is actively sent to server end.
Distributed Message Queue 52 is further used for：Task storage in the Distributed Message Queue of limit priority is arrived
In other Distributed Message Queues, and the high-priority task storage for being currently needed for performing is disappeared to the distribution of limit priority
The Head-of-line of queue is ceased, then the distribution for arriving limit priority will be stored to the task storage in other Distributed Message Queues
In message queue behind high-priority task.The above-mentioned processing of the embodiment of the present invention, can be preferential by a kind of replacement policy
Ensure that high-priority task seizes execution.
Task scheduling modules 54, for being adjusted from Distributed Message Queue corresponding task according to system resource free message
Go out and distribute corresponding system resource.
Task scheduling modules 54 are specifically used for：Determining that system resource is able to carry out task according to system resource free message
In the case of, corresponding Distributed Message Queue 52 is notified, receives the first positioned at queue of the Distributed Message Queue 52 push
Task, and distribute corresponding system resource and perform the task；Or actively from the high priority point of Distributed Message Queue 52
The queue first place acquisition task of cloth message queue, and distribute corresponding system resource and perform the task；According to system resource
Free message determines that system resource can not isolate Distributed Message Queue in the case of execution task, forbids receiving distributed
The task that message queue 52 pushes.
That is, when required by task resource can obtain, the distribution of notifications formula message queue 52 of task scheduling modules 54,
Distributed Message Queue 52 according to priority pushes task, and task scheduling modules 54 distribute respective resources and perform the task；Work as task
When required resource is all occupied, task scheduling modules 54 isolate message queue 52, no longer receiving queue task.When resource is idle
When, task scheduling modules 54 can also be actively from the acquisition task in a certain high-priority queue.
Task scheduling modules 54 are further used for：In the case where needing to perform high-priority task, by what is currently performed
The task status of low priority task is wait state from running state conversion, and low priority task is sent into task collects mould
Block 50, and perform high-priority task；According to the current running status for performing task, the preferential of current execution task is recalculated
Level, the current priority for performing task is improved or reduced according to result of calculation, and set according to the current priority for performing task
Put the current task status for performing task；In the case where task status is set to wait for state, task transmission will be currently performed
To task collection module 50, in the case where task status is arranged to running status, current execution task is continued executing with；
In summary, task scheduling modules 54 are responsible for the task that scheduling is stored in Distributed Message Queue 52, and manage
Its own Resource Guarantee tasks carrying.
The above-mentioned technical proposal of the embodiment of the present invention is described in detail below in conjunction with accompanying drawing.
As shown in Fig. 2 in this example, five resident Distributed Message Queues with different priorities can be created,
That is, Distributed Message Queue -1, Distributed Message Queue -2, Distributed Message Queue -3, Distributed Message Queue -4, distribution
Message queue -5；In actual applications, naturally it is also possible to more Distributed Message Queues for being adapted to application are created, can also root
According to the Distributed Message Queue of the instant establishment respective priority of the characteristic attribute of task, destruction that can also be at any time is without task
Distributed Message Queue, search for the scheduling of task scheduling modules 54 and maintain a less scope, improve the efficiency of response.
As shown in Fig. 2 task source -1, task source -2 ..., task source-N caused task is sent to task collection module 50,
Task collection module 50 carries out task collection, then, according to the precedence information of task, task is respectively stored in into corresponding point
In cloth message queue 52, finally, task scheduling modules 54 are according to the idle condition of system resource, from Distributed Message Queue
Carry out task scheduling.
As shown in figure 3, after task collection module 50 carries out task collection, according to the high, medium and low precedence information of task,
Task is respectively stored in high priority Distributed Message Queue, middle priority Distributed Message Queue and low priority point
In cloth message queue, then, task scheduling modules 54 are according to the idle condition of system resource, from Distributed Message Queue 52
Carry out task scheduling.It should be noted that the processing for entering Distributed Message Queue 52 of task is included two kinds of situations by this example：
Situation one, task collection module 50 are stored in the task of queue after task priority is confirmed；2nd, task scheduling modules 54 allow height
Priority tasks swap out the task of low priority, are wait state from running state conversion by the task status of task and relay
Enter Distributed Message Queue 52 (that is, the flow shown in dotted arrow).In addition, task scheduling modules 54 can also be according to task
Running status improves or reduced the priority of task by weighting, calculating the methods of run time, changes task status.
Fig. 4 is the flow chart of the detailed processing of the task management method of the embodiment of the present invention, as shown in figure 4, including as follows
Processing：
Step 1, task source creates task, and task call task collection module 50 carries out the interface of task collection；
Step 2, task collection module 50 receives the task that task source is sent, and the priority of task is analyzed, and obtains
Take the precedence information of task；
Step 3, task storage is arrived different distributed messages by task collection module 50 according to the precedence information of task
In queue 52；
Step 4, the active push task of Distributed Message Queue 52 is to task scheduling modules 54；
Step 5, task scheduling modules 54 can store the task of low priority point again according to the priority of task
In cloth message queue, and the task scheduling of high priority is come out；
Step 6, task distributing system resource of the task scheduling modules 54 for scheduling out；
Step 7, server end performs the task using system resource.
In summary, the task that task source creates is stored in phase by the technical scheme of the embodiment of the present invention by assigned priority
In the Distributed Message Queue for answering priority, task can be pushed dequeue and be scheduled for performing by Distributed Message Queue, also may be used
To wait task to be scheduled dequeue.Server end is enabled to respond the task storage of the big quantity of task source initiation and adjust
Degree, make all tasks from be created to scheduling complete life cycle in can be by efficient, rational execution.In addition, the present invention is real
The push function that example also flexibly uses Distributed Message Queue is applied, enables task actively to attempt to apply for that resource performs, makes task
Real-time be further enhanced；The embodiment of the present invention is by the way that the flow of task to be summarized as being stored in queue and recalling queue holding
OK, become apparent from the management process of task, there is more preferable autgmentability and exploitativeness so that be existing and following be likely to occur
Task storage and management realization can be more flexible.
Although being example purpose, the preferred embodiments of the present invention are had been disclosed for, those skilled in the art will recognize
Various improvement, increase and substitution are also possible, and therefore, the scope of the present invention should be not limited to above-described embodiment.
Claims (9)
- A kind of 1. task management method, it is characterised in that including：Server end obtains the precedence information of task, and task storage is arrived into respective priority according to the precedence information Distributed Message Queue in；The server end corresponding task is recalled from the Distributed Message Queue according to system resource free message and point With corresponding system resource；The server end corresponding task is recalled from the Distributed Message Queue according to system resource free message and point Specifically included with corresponding system resource：In the case where determining that system resource is able to carry out task according to the system resource free message, the server end leads to Know corresponding Distributed Message Queue, receive the task the first positioned at queue of Distributed Message Queue push, and distribute phase The system resource answered performs the task；Or the server end is actively from the high priority point of the Distributed Message Queue The queue first place acquisition task of cloth message queue, and distribute corresponding system resource and perform the task；Determining that system resource can not be in the case of execution task according to the system resource free message, the server end Isolate the Distributed Message Queue, forbid receiving the task of the Distributed Message Queue push.
- 2. the method as described in claim 1, it is characterised in that the precedence information that the server end obtains task specifically wraps Include：In the case where the task does not carry precedence information, the server end according to the characteristic information of the task, Corresponding priority is distributed for the task, and obtains corresponding precedence information；In the case where the task carries the precedence information, the server end directly obtains to be carried in the task Precedence information.
- 3. the method as described in claim 1, it is characterised in that the Distributed Message Queue is：The distribution of first in first out Message queue.
- 4. the method as described in claim 1, it is characterised in that methods described also includes：In the case where needing to perform high-priority task, the server end is by the task of the low priority task currently performed State is wait state from running state conversion, is again arrived low priority task storage according to the task status corresponding In the Distributed Message Queue of priority, and perform the high-priority task.
- 5. the method as described in claim 1, it is characterised in that the server end will be corresponding according to system resource free message Task is recalled from the Distributed Message Queue and distributed after corresponding system resource, and methods described also includes：The server end recalculates the current priority for performing task according to the current running status for performing task, The current priority for performing task is improved or reduced according to result of calculation, and according to the preferential of the current execution task Level sets the current task status for performing task；In the case where the task status is set to wait for state, the server end again deposits the current execution task In the Distributed Message Queue for storing up respective priority, in the case where the task status is arranged to running status, the clothes Business device end continues executing with the current execution task.
- 6. the method as described in claim 1, it is characterised in that the server end will be corresponding according to system resource free message Task is recalled from the Distributed Message Queue and distributed before corresponding system resource, and methods described also includes：Task storage in the Distributed Message Queue of limit priority is arrived other Distributed Message Queues by the server end In, and the high-priority task for being currently needed for performing is stored to the head of the queue position of the Distributed Message Queue of the limit priority Put, then the distributed message team for arriving the limit priority will be stored to the task storage in other Distributed Message Queues Described in row behind high-priority task.
- 7. a kind of task management device, it is characterised in that positioned at server end, described device includes：Task collection module, for obtaining the precedence information of task, the task is sent to according to the precedence information In the Distributed Message Queue of respective priority；The Distributed Message Queue, for storing appointing for respective priority in the Distributed Message Queue of different priorities Business；Task scheduling modules, for being recalled from the Distributed Message Queue corresponding task according to system resource free message And distribute corresponding system resource；The task scheduling modules are specifically used for：Determining that system resource is able to carry out appointing according to the system resource free message In the case of business, corresponding Distributed Message Queue is notified, receives the first positioned at queue of Distributed Message Queue push Task, and distribute corresponding system resource and perform the task；Or actively from the high priority point of the Distributed Message Queue The queue first place acquisition task of cloth message queue, and distribute corresponding system resource and perform the task；According to the system Resource free time information determining system resource can not isolate the Distributed Message Queue, forbid connecing in the case of execution task Receive the task of the Distributed Message Queue push.
- 8. device as claimed in claim 7, it is characterised in thatThe task collection module is specifically used for：In the case where the task does not carry precedence information, according to described The characteristic information of business, corresponding priority is distributed for the task, and obtain corresponding precedence information；Carried in the task In the case of having the precedence information, the precedence information carried in the task is directly obtained.
- 9. device as claimed in claim 7, it is characterised in thatThe task scheduling modules are further used for：It is low by what is currently performed in the case where needing to perform high-priority task The task status of priority tasks is wait state from running state conversion, and the low priority task is sent into the task Collection module, and perform the high-priority task；According to the current running status for performing task, recalculate and described currently hold The priority of row task, the current priority for performing task is improved or reduced according to result of calculation, and worked as according to described The priority of preceding execution task sets the current task status for performing task；State is set to wait in the task status In the case of, the current execution task is sent to the task collection module, is arranged to run shape in the task status In the case of state, the current execution task is continued executing with；The Distributed Message Queue is further used for：Task storage in the Distributed Message Queue of limit priority is arrived it In his Distributed Message Queue, and the high-priority task for being currently needed for performing is stored to the distribution for arriving the limit priority The Head-of-line of message queue, then it is preferential to the highest to the task storage in other Distributed Message Queues by storing Described in the Distributed Message Queue of level behind high-priority task.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110219354.1A CN102915254B (en) | 2011-08-02 | 2011-08-02 | task management method and device |
| PCT/CN2012/072340 WO2012155628A1 (en) | 2011-08-02 | 2012-03-14 | Method and apparatus for task management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110219354.1A CN102915254B (en) | 2011-08-02 | 2011-08-02 | task management method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102915254A CN102915254A (en) | 2013-02-06 |
| CN102915254B true CN102915254B (en) | 2018-04-06 |
Family
ID=47176237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110219354.1A Expired - Fee Related CN102915254B (en) | 2011-08-02 | 2011-08-02 | task management method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102915254B (en) |
| WO (1) | WO2012155628A1 (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179009B (en) * | 2013-04-22 | 2016-01-13 | 烽火通信科技股份有限公司 | A kind of dynamic self-adapting call method of distributed management system |
| CN104252390B (en) * | 2013-06-28 | 2018-08-14 | 华为技术有限公司 | Resource regulating method, device and system |
| CN103678488B (en) * | 2013-11-12 | 2015-03-04 | 德比软件(上海)有限公司 | Distributed mass dynamic task engine and method for processing data with same |
| EP3097545A4 (en) * | 2014-01-22 | 2017-07-19 | Speakeasy, Inc. | Systems and methods of socially-driven product offerings |
| CN104978232A (en) * | 2014-04-09 | 2015-10-14 | 阿里巴巴集团控股有限公司 | Computation resource capacity expansion method for real-time stream-oriented computation, computation resource release method for real-time stream-oriented computation, computation resource capacity expansion device for real-time stream-oriented computation and computation resource release device for real-time stream-oriented computation |
| CN104102740A (en) * | 2014-07-30 | 2014-10-15 | 精硕世纪科技（北京）有限公司 | Distribution type information acquisition system and method |
| CN104915256B (en) * | 2015-06-05 | 2019-03-26 | 惠州Tcl移动通信有限公司 | A kind of the Real-Time Scheduling implementation method and its system of task |
| CN105094972A (en) * | 2015-07-28 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and device for processing database operation tasks |
| CN106201676A (en) * | 2016-06-28 | 2016-12-07 | 浪潮软件集团有限公司 | Task allocation method and device |
| CN107665163B (en) * | 2016-07-29 | 2021-03-30 | 北京京东尚科信息技术有限公司 | Method and system for automatic data backtracking |
| CN108205470A (en) * | 2016-12-20 | 2018-06-26 | 北京奇虎科技有限公司 | A kind of distribution ad data calculating task management system and method |
| CN108345501B (en) * | 2017-01-24 | 2021-10-29 | 全球能源互联网研究院有限公司 | Distributed resource scheduling method and system |
| CN107423120B (en) * | 2017-04-13 | 2020-06-30 | 阿里巴巴集团控股有限公司 | Task scheduling method and device |
| CN107733967B (en) * | 2017-07-24 | 2020-01-07 | 深圳壹账通智能科技有限公司 | Processing method and device for push information, computer equipment and storage medium |
| CN109408220A (en) * | 2017-08-17 | 2019-03-01 | 北京国双科技有限公司 | A kind of task processing method and device |
| CN108563495A (en) * | 2018-04-08 | 2018-09-21 | 郑州云海信息技术有限公司 | The cloud resource queue graded dispatching system and method for data center's total management system |
| CN110569252B (en) * | 2018-05-16 | 2023-04-07 | 杭州海康威视数字技术股份有限公司 | Data processing system and method |
| CN109388504A (en) * | 2018-09-26 | 2019-02-26 | 平安科技（深圳）有限公司 | Messageization docks processing method, device, computer equipment and storage medium |
| CN109582445A (en) * | 2018-09-29 | 2019-04-05 | 阿里巴巴集团控股有限公司 | Message treatment method, device, electronic equipment and computer readable storage medium |
| CN109582446A (en) * | 2018-10-11 | 2019-04-05 | 中国建设银行股份有限公司 | Quasi real time asynchronous batch processing system, method, apparatus and storage medium |
| CN109756565A (en) * | 2018-12-26 | 2019-05-14 | 成都科来软件有限公司 | A kind of Multitask Data method for pushing based on statistical form |
| CN109814997B (en) * | 2019-01-18 | 2021-08-06 | 创新奇智(广州)科技有限公司 | Distributed autonomous balanced artificial intelligence task scheduling method and system |
| CN110225102B (en) * | 2019-05-23 | 2022-06-03 | 北京小米移动软件有限公司 | Method, device and system for pushing information |
| CN110162418B (en) * | 2019-05-30 | 2023-06-27 | 努比亚技术有限公司 | Application cold start method, mobile terminal and computer readable storage medium |
| CN110210789A (en) * | 2019-06-17 | 2019-09-06 | 广东电网有限责任公司 | Resource distribution dispatching method, the device, equipment of power grid test business |
| CN110888726A (en) * | 2019-11-25 | 2020-03-17 | 中冶赛迪重庆信息技术有限公司 | Multitask concurrent processing method and system |
| CN111240865A (en) * | 2020-01-14 | 2020-06-05 | 山东汇贸电子口岸有限公司 | Elastic flexible activity priority management system and method based on message queue |
| CN113411365A (en) * | 2020-03-17 | 2021-09-17 | 中国移动通信集团山东有限公司 | Data processing method and device |
| CN111404838A (en) * | 2020-03-17 | 2020-07-10 | 上海云励科技有限公司 | Message processing method, device and equipment |
| CN111399989B (en) * | 2020-04-10 | 2022-11-18 | 中国人民解放军国防科技大学 | Container cloud-oriented task preemption and scheduling method and system |
| CN111784288A (en) * | 2020-06-29 | 2020-10-16 | 广州华多网络科技有限公司 | Task management method, device, system, storage medium and equipment |
| CN111709723B (en) * | 2020-07-20 | 2023-11-28 | 港胜技术服务(深圳)有限公司 | RPA business process intelligent processing method, device, computer equipment and storage medium |
| CN112269999A (en) * | 2020-11-04 | 2021-01-26 | 杭州迪普科技股份有限公司 | Vulnerability scanning task scheduling method, device, equipment and medium |
| CN113162990B (en) * | 2021-03-30 | 2022-08-16 | 杭州趣链科技有限公司 | Message sending method, device, equipment and storage medium |
| CN113590299B (en) * | 2021-09-28 | 2022-03-01 | 南京国睿信维软件有限公司 | Conversion scheduling framework algorithm of high-concurrency high-availability heterogeneous system |
| CN115766613B (en) * | 2022-10-28 | 2023-05-23 | 华能信息技术有限公司 | Data transmission management method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3893136B2 (en) * | 2004-05-13 | 2007-03-14 | ファームウェアシステム株式会社 | Embedded computer control program, recording medium recording the program, and embedded system |
| CN101084680A (en) * | 2004-10-28 | 2007-12-05 | 意大利电信股份公司 | Method for managing resources in a platform for telecommunication service and/or network management, corresponding platform and computer program product therefor |
| CN101719931A (en) * | 2009-11-27 | 2010-06-02 | 南京邮电大学 | Multi-intelligent body-based hierarchical cloud computing model construction method |
| CN101743534A (en) * | 2007-07-16 | 2010-06-16 | 微软公司 | By increasing and shrinking resources allocation and dispatch |
| KR20100065572A (en) * | 2008-12-08 | 2010-06-17 | 재단법인대구경북과학기술원 | Method for re-defining priority and managing ready queue in system having static task define function and system for executing the method |
-
2011
- 2011-08-02 CN CN201110219354.1A patent/CN102915254B/en not_active Expired - Fee Related
-
2012
- 2012-03-14 WO PCT/CN2012/072340 patent/WO2012155628A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3893136B2 (en) * | 2004-05-13 | 2007-03-14 | ファームウェアシステム株式会社 | Embedded computer control program, recording medium recording the program, and embedded system |
| CN101084680A (en) * | 2004-10-28 | 2007-12-05 | 意大利电信股份公司 | Method for managing resources in a platform for telecommunication service and/or network management, corresponding platform and computer program product therefor |
| CN101743534A (en) * | 2007-07-16 | 2010-06-16 | 微软公司 | By increasing and shrinking resources allocation and dispatch |
| KR20100065572A (en) * | 2008-12-08 | 2010-06-17 | 재단법인대구경북과학기술원 | Method for re-defining priority and managing ready queue in system having static task define function and system for executing the method |
| CN101719931A (en) * | 2009-11-27 | 2010-06-02 | 南京邮电大学 | Multi-intelligent body-based hierarchical cloud computing model construction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102915254A (en) | 2013-02-06 |
| WO2012155628A1 (en) | 2012-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102915254B (en) | task management method and device | |
| US20210311781A1 (en) | Method and system for scalable job processing | |
| CN109451072A (en) | A kind of message caching system and method based on Kafka | |
| CN103516807B (en) | A kind of cloud computing platform server load balancing system and method | |
| CN103927225B (en) | A kind of internet information processing optimization method of multi-core framework | |
| CN104123182B (en) | Based on the MapReduce task of client/server across data center scheduling system and method | |
| CN109697122A (en) | Task processing method, equipment and computer storage medium | |
| CN108762896A (en) | One kind being based on Hadoop cluster tasks dispatching method and computer equipment | |
| CN101242360B (en) | A network address conversion method and system based on priority queue | |
| CN104112049B (en) | Based on the MapReduce task of P2P framework across data center scheduling system and method | |
| WO2021057500A1 (en) | Message sending management method and device | |
| CN109445921A (en) | A kind of distributed data task processing method and device | |
| CN114610474A (en) | Multi-strategy job scheduling method and system in heterogeneous supercomputing environment | |
| CN106095581B (en) | Network storage virtualization scheduling method under private cloud condition | |
| CN114710571B (en) | Data packet processing system | |
| CN102609307A (en) | Multi-core multi-thread dual-operating system network equipment and control method thereof | |
| CN114938372A (en) | Federal learning-based micro-grid group request dynamic migration scheduling method and device | |
| Serdaroglu et al. | An efficient multipriority data packet traffic scheduling approach for fog of things | |
| CN110365786A (en) | Job processing system, asynchronous operation dispatching method and computer equipment | |
| CN104580209A (en) | Device and method for implementing multi-platform message processing | |
| Dong et al. | Task-aware flow scheduling with heterogeneous utility characteristics for data center networks | |
| CN108667920B (en) | Service flow acceleration system and method for fog computing environment | |
| CN107665145B (en) | Browser stand-alone resource management method based on container | |
| CN112769788B (en) | Charging service data processing method and device, electronic equipment and storage medium | |
| CN107797870A (en) | A kind of cloud computing data resource dispatching method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180406 Termination date: 20190802 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |