CN104636200B - A kind of method of computer multi-process equity dispatching - Google Patents
A kind of method of computer multi-process equity dispatching Download PDFInfo
- Publication number
- CN104636200B CN104636200B CN201310567625.1A CN201310567625A CN104636200B CN 104636200 B CN104636200 B CN 104636200B CN 201310567625 A CN201310567625 A CN 201310567625A CN 104636200 B CN104636200 B CN 104636200B
- Authority
- CN
- China
- Prior art keywords
- scheduling
- queue
- collection
- standby
- dispatching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of methods of computer multi-process equity dispatching, comprising the following steps: step 1: the process composition scheduling collection that can carry out concurrent scheduling is taken out from process sets, by scheduled process does not form standby scheduling queue in process sets；Step 2: judging whether process scheduling request terminates, and is such as not finished and performs the next step suddenly, otherwise finishing scheduling；Step 3: scheduling collection is scheduled；Step 4: the process that scheduling is completed is entered to the tail of the queue of standby scheduling queue；Step 5: scheduling collection is emptied；Step 6: choosing the new scheduling collection of process composition for carrying out concurrent scheduling from standby scheduling queue, and gos to step two.The present invention is operated by set operation, constructs the scheduling collection and standby scheduling queue of process, realizes computer processor to the equity dispatching of multi-process, and deadlock when avoiding process concurrent scheduling occurs, and improves computer processor to the dispatching of concurrent process.
Description
Technical field
The present invention relates in concurrent computer architecture multi-process resource allocation techniques field more particularly to a kind of meter
The method of calculation machine multi-process equity dispatching.
Background technique
Multiprocess scheduling problem in operating system is an important research contents of computer science, process tune
The quality of degree algorithm directly influences the utilization rate of computer resource and the speed of service of operating system.Multiprocess scheduling algorithm solution
Key problem certainly is Process Synchronization (Synchronization) problem.The problem refers to number of processes and available resources number
It mismatches, solves the problems, such as that this needs to consider three aspect of concurrency, deadlock and fairness.Solution master for this problem at present
It concentrates on: 1) avoiding process deadlock method when process concurrent scheduling；2) effective workaround when process deadlock occurs.Tool
The method of body includes: waiter's solution, resource hierarchical solution, Chandy/Mistra solution and Petri network simulation multi-process tune
The solution etc. of degree.But these methods it is difficult to ensure that computer to the fairness of multiprocess scheduling.
Waiter's solution is the scheduling of a processor monitoring process in designated computer, and other processors carry out process
Scheduling, it is all to be scheduled the license for requiring to obtain par-ticular processor.It is scheduled, is avoided by this specified processor
Deadlock when process scheduling, but this process scheduling method is not directed to fairness.Resource hierarchical solution is to resource
The relationship of a partial order or classification is distributed, and arranges all resources and is obtained all in accordance with this sequence, is discharged in reverse order, and
And guarantee required for being worked simultaneously by same item there are no two unrelated resource.But when number of processes increases in operating system,
Process scheduling sequence predetermined will not change, and cause the long-term waiting of multiple processes, therefore resource hierarchical solution is not suitable for
Dynamic dispatching of the computer to multi-process.Chandy/Mistra solution allows process to compete any resource, is suitable for actual conditions
Scheduling of the lower computer to process, but whithin a period of time, there are certain processes repeatedly to be dispatched, and certain processes seldom quilt
The case where scheduling, therefore this method cannot be guaranteed the fairness of process scheduling.When Petri network simulates multiprocess scheduling problem, increase
Two clearance cards are added, have guaranteed that the process being not carried out possesses resource access authority using clearance card, has avoided process and die of hunger shape
The generation of state also ensures the fairness of process scheduling to a certain extent.But this method will cause shape in Petri net model
Sharply increasing for state, leads to State-explosion problem, is unfavorable for computer processor to the Real-Time Scheduling of multi-process.
In an operating system, concurrent process number is dynamic change, and demand of each process to resource is also that dynamic becomes
Change, it is difficult to ensure that the fairness of process scheduling, reduces behaviour when multiprocess scheduling algorithm is such issues that solve in the prior art
Make system to the managerial ability of process scheduling.
Therefore, a kind of multiprocess scheduling method for designing justice has become urgently to be solved in computer science ask
Topic.
Summary of the invention
The purpose of the present invention is the unjustnesses for solution multiprocess scheduling algorithm in the prior art of determining, and propose a kind of computer
The method of multi-process equity dispatching.By set operation, the scheduling collection and standby scheduling queue of process are constructed, computer is realized
Fairness dispatch of the processor to multi-process.The algorithm not only can effectively solve the scheduling problem and process deadlock of concurrent process
Problem, while computer processor is also achieved to the fairness of multiprocess scheduling, operating system is improved to concurrent process tune
The managerial ability of degree.
The method of computer multi-process equity dispatching of the present invention needs to be assumed as follows and defined:
It is assumed that process sets A={ a1..., an, wherein aiIndicate a process, resource set B={ b1..., bm, wherein biIt indicates
For a resource in computer.
It defines 1 (impact factor): after a certain process a is called, k resource in resource set B can be occupied, and then influence
Other j processes obtain the resource in resource set B in process sets A, then claim this j process ai+l..., ai+jIt is process a in resource
Collect the impact factor in B.
Define 2 (impact factor functions): if process ai+l..., ai+jIt is impact factor of the process a in resource set B, then letter
Number f (a)={ ai+l..., ai+jIt is referred to as impact factor function of the process a in resource set B.
Define 3 (scheduling collection): scheduling collection D is a proper subclass of process sets A, is had for any d1, d2 ∈ D, and for any a ∈ A-D, there is always d ∈ D to make a ∈ f (d).
Define 4 (standby scheduling queues): the process in standby scheduling queue is the process in set A-D.
The invention proposes a kind of methods of computer multi-process equity dispatching, comprising the following steps:
Step 1: process composition scheduling collection is taken out from process sets, by scheduled process does not form standby in process sets
Scheduling queue；Step 2: judging whether process scheduling request terminates, and is such as not finished and performs the next step suddenly, otherwise finishing scheduling；Step
Rapid three: scheduling collection is scheduled；Step 4: the process after dispatching concentration is dispatched is transferred to the tail of the queue of standby scheduling queue；
Step 5: scheduling collection is emptied；Step 6: process constructive scheduling collection again is chosen from standby scheduling queue, and is gone to step
Two.
Wherein, scheduling collection is made of the process that can obtain resource in process sets simultaneously in step 1.
Wherein, further, the composition of collection is dispatched in step 1 the following steps are included:
Step A: process replication in process sets to interim process is concentrated, and is initialized scheduling and integrated as empty set；Step B: with
Machine concentrates one process of any taking-up from interim process；Step C: process is put into dispatching concentration；Step D: interim process is deleted
The impact factor of process concentrate；Step E: interim process integrates not as empty set, then jumps to step B or interim process integrates as empty set,
Then terminate the composition of scheduling collection.
Wherein, the process of dispatching concentration is chosen from the head of the queue of standby scheduling queue.
Wherein, further, dispatch the composition of collection the following steps are included:
Step F: process is chosen from standby scheduling queue and is put into dispatching concentration；Step G: by standby scheduling queue not
The process being selected is arranged by former sequence.
Wherein, further, the process that collection is dispatched in step 6 is successively chosen from the head of the queue of standby scheduling queue to tail of the queue.
Wherein, further, the process chosen in step 6 is not the impact factor of any process of dispatching concentration.
Wherein, scheduling collection operates generation by set operation with standby scheduling queue.
Wherein, scheduling collection and standby scheduling queue can be converted mutually.
The present invention is to realize that computer processor to the concurrency of multiprocess scheduling, Nondeadlock and fairness, proposes
A kind of method of computer multi-process equity dispatching dispatches team using the scheduling collection and standby of set operation operation acquisition process
Column, then schedulable Process Calls Sequence is generated, realize the fairness dispatch that computer processor operates multi-course concurrency.
Detailed description of the invention
Fig. 1 is the process status transition diagram of the method process scheduling process of computer multi-process equity dispatching of the present invention.
Fig. 2 is the flow chart of scheduling collection D construction algorithm in the method for computer multi-process equity dispatching of the present invention.
Fig. 3 is the flow chart of the method for computer multi-process equity dispatching of the present invention.
Fig. 4 is the method using computer multi-process equity dispatching of the present invention to 5 times of 5 processes and 2 shared resources
The simulation drawing of concurrent scheduling.
Fig. 5 is the method using computer multi-process equity dispatching of the present invention to 7 times of 6 processes and 2 shared resources
The simulation drawing of concurrent scheduling.
Specific embodiment
In conjunction with following specific embodiments and attached drawing, the present invention is described in further detail.Implement process of the invention,
Condition, experimental method etc. are among the general principles and common general knowledge in the art, this hair in addition to what is specifically mentioned below
It is bright that there are no special restrictions to content.
Fig. 1 is the process status transition diagram of process scheduling process in the present invention.In computer multi-process justice tune of the present invention
In the method for degree, all processes are divided into three kinds of states:
1) process sets state when initial；
2) the scheduling collection state of process；
3) standby scheduling queue state when process is not scheduled.
The method of computer multi-process equity dispatching according to the present invention, process sets state are converted to scheduling collection state and standby
Scheduling queue state.It is mutually converted between scheduling collection state and standby scheduling queue state, until process scheduling request terminates.
Fig. 2 is the flow chart of scheduling collection D construction algorithm in the method for computer multi-process equity dispatching of the present invention.It is defeated first
Enter process sets A, then by all process replications in process sets A to interim process collection C, initialization scheduling integrates D as empty set.When interim
When process sets C is not empty, following operation is repeated:
1) a process a in interim process collection C is obtained at random；
2) set operation operation C=C-f (a)-{ a } is executed；
3)
When interim process integrates C as empty set, output scheduling collection D.Due to being the process a taken in interim process collection C at random, because
The scheduling collection D that this is finally formed is not unique.
Fig. 3 is the flow chart of the method for computer multi-process equity dispatching of the present invention.Computer multi-process equity dispatching side
The step of method, is as follows:
Step 1: can carry out the process of concurrent scheduling by selecting in process sets A, and composition scheduling collection D will be remained in process sets A
Yu Wei scheduled process presses former sequential arrangement and forms standby scheduling queue L；
Step 2: judging whether process scheduling request terminates, and if being not finished, performs the next step suddenly, otherwise finishing scheduling；
Step 3: concurrent scheduling is carried out to all processes in scheduling collection D；
Step 4: the process being scheduled is inserted into the tail of the queue of standby scheduling queue L；
Step 5: scheduling collection D is emptied；
Step 6: from the head of the queue of standby scheduling queue L successively choose backward it is all can concurrent scheduling process, construct it is next
Secondary scheduling collection D, and go to step two.
The operation of set operation is utilized in the method for computer multi-process equity dispatching of the present invention, so that computer processor
Concurrent scheduling can be carried out to all processes in scheduling collection D, standby scheduling cannot temporarily will be put by the process of concurrent scheduling
Then process after scheduling is equally put into standby scheduling queue L, then according to queue first in first out feature, takes out standby by queue L
All process constructive scheduling collection D that can carry out concurrent scheduling in scheduling queue L, therefore all processes all can be by fair tune
Degree.
In the step of computer multi-process equity dispatching method one, the composition of scheduling collection D is the following steps are included: step A: will
Process replication in process sets A is concentrated to interim process；Step B: it is concentrated from interim process take out a process at random；Step C:
Process is put into scheduling collection D；Step D: the impact factor of interim process process concentrate is deleted；Step E: interim process collection is not
Empty set, then jump to step B or interim process integrates the composition for then terminating scheduling collection as empty set.
In the step of computer multi-process equity dispatching method six, the composition of scheduling collection D is the following steps are included: step F: from
Process is chosen in standby scheduling queue to be put into the scheduling collection D；Step G: will be unselected in the standby scheduling queue
Process arranged by former sequence.
Embodiment 1:
Fig. 4 is the method using computer multi-process equity dispatching of the present invention to 5 times of 5 processes and 2 shared resources
Concurrent scheduling simulation drawing, wherein process is 1,2,3,4 and 5, and there are impact factor function f (1)={ 5,2 }, f (2)={ 1,3 }, f
(3)={ 2,4 }, f (4)={ 3,5 } and f (5)={ Isosorbide-5-Nitrae }.
According to the generating algorithm of scheduling collection D, process 1 is randomly selected, scheduling collection D={ 1,3 } that can be obtained is unscheduled
Process 2,4 and 5 is put into standby scheduling queue L (queue is referred to as in Fig. 4), and 5 scheduling are as follows:
1st scheduling: the process 1 and 3 in scheduling collection D={ 1,3 } is scheduled, then again by standby scheduling queue L
Scheduling collection D={ 2,4 }, and the deletion process 2 and 4 from standby scheduling queue L are obtained, the process 1 and 3 after scheduling is inserted into standby
The tail of the queue of scheduling queue L, then the process in standby scheduling queue L is adjusted to 5,1 and 3.
2nd scheduling: the process 2 and 4 in scheduling collection D={ 2,4 } is scheduled, then again by standby scheduling queue L
Scheduling collection D={ 5,3 }, and the deletion process 5 and 3 from standby scheduling queue L are obtained, the process 2 and 4 after scheduling is inserted into standby
The tail of the queue of scheduling queue L, then the process in standby scheduling queue L is adjusted to 1,2 and 4.
3rd scheduling: the process 5 and 3 in scheduling collection D={ 5,3 } is scheduled, then again by standby scheduling queue L
Scheduling collection D={ Isosorbide-5-Nitrae }, and the deletion process 1 and 4 from standby scheduling queue L are obtained, the process 5 and 3 after scheduling is inserted into standby
The tail of the queue of scheduling queue L, then the process in standby scheduling queue L is adjusted to 2,5 and 3.
The 4th scheduling: the process 1 and 4 in scheduling collection D={ Isosorbide-5-Nitrae } is scheduled, then again by standby scheduling queue L
Scheduling collection D={ 2,5 }, and the deletion process 2 and 5 from standby scheduling queue L are obtained, the process 1 and 4 after scheduling is inserted into standby
The tail of the queue of scheduling queue L, then the process in standby scheduling queue L is adjusted to 3,1 and 4.
The 5th scheduling: the process 2 and 5 in scheduling collection D={ 2,5 } is scheduled.
During five scheduling, process 1,2,3,4 and 5 has been scheduled 2 times.Scheduling collection D and standby scheduling queue L
It is successively interactive appearance, and since not scheduled process is first put into standby scheduling queue L, after being inserted into after the process being scheduled
Standby scheduling queue L, while being to choose process from standby scheduling queue L in the way of first in first out again.
Therefore, the method for computer multi-process equity dispatching of the present invention can be realized the fairness dispatch of multi-process.
Embodiment 2:
Fig. 5 is that the present invention utilizes the method for computer multi-process equity dispatching of the present invention to 6 processes and 2 shared resources
7 process scheduling simulation drawings, wherein process is 1,2,3,4,5 and 6, and there are impact factor function f (1)={ 2,3,4 }, f
(2)={ 3,5 }, f (3)={ 1,2,4 }, f (4)={ 2,3,6 }, f (5)={ 4,6 } and f (6)={ 1,5 }.
According to the generating algorithm of scheduling collection D, if randomly selecting process 1, scheduling collection D={ 1,5 } that can be obtained is unscheduled
Process 2,3,4 and 6 be put into standby scheduling queue L (in Fig. 5 be referred to as queue), 7 scheduling are as follows:
1st scheduling: the process 1 and 5 in scheduling collection D={ 1,5 } is scheduled, is then obtained by standby scheduling queue L
Process 1 and 5 after scheduling is inserted into standby scheduling by scheduling collection D={ 2,6 }, and the deletion process 2 and 6 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 3,4,1 and 5.
2nd scheduling: the process 2 and 6 in scheduling collection D={ 2,6 } is scheduled, is then obtained by standby scheduling queue L
Process 2 and 6 after scheduling is inserted into standby scheduling by scheduling collection D={ 3,5 }, and the deletion process 3 and 5 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 4,1,2 and 6.
3rd scheduling: the process 3 and 5 in scheduling collection D={ 3,5 } is scheduled, is then obtained by standby scheduling queue L
Process 3 and 5 after scheduling is inserted into standby scheduling by scheduling collection D={ 4,2 }, and the deletion process 4 and 2 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 1,6,3 and 5.
The 4th scheduling: the process 4 and 2 in scheduling collection D={ 4,2 } is scheduled, is then obtained by standby scheduling queue L
Process 4 and 2 after scheduling is inserted into standby scheduling by scheduling collection D={ 1,5 }, and the deletion process 1 and 5 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 6,3,4 and 2.
The 5th scheduling: the process 1 and 5 in scheduling collection D={ 1,5 } is scheduled, is then obtained by standby scheduling queue L
Process 1 and 5 after scheduling is inserted into standby scheduling by scheduling collection D={ 6,3 }, and the deletion process 6 and 3 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 4,2,1 and 5.
6th scheduling: the process 6 and 3 in scheduling collection D={ 6,3 } is scheduled, is then obtained by standby scheduling queue L
Process 6 and 3 after scheduling is inserted into standby scheduling by scheduling collection D={ 4,2 }, and the deletion process 4 and 2 from standby scheduling queue L
The tail of the queue of queue L, then the process in standby scheduling queue L is adjusted to 1,5,6 and 3.
7th scheduling: the process 4 and 2 in scheduling collection D={ 4,2 } is scheduled.
7 process schedulings in Fig. 5 are counted, statistical result is as shown in table 1.
As shown in Table 1, in addition to process 2 is invoked 3 times, other processes are all invoked 2 times.
Therefore, the method for computer multi-process equity dispatching of the present invention realizes the fairness to multiprocess scheduling.
The scheduled number of each process during 1:7 process concurrent scheduling of table
| Process | 1 | 2 | 3 | 4 | 5 | 6 |
| Scheduled number | 2 | 3 | 2 | 2 | 2 | 2 |
Protection content of the invention is not limited to above embodiments.Under the spirit and scope without departing substantially from invention, this field
Technical staff it is conceivable that variation and advantage be all included in the present invention, and with appended claims be protection model
It encloses.
Claims (6)
1. a kind of method of computer multi-process equity dispatching, which comprises the following steps:
Step 1: the process composition scheduling collection that can carry out concurrent scheduling is taken out from process sets, will not be adjusted in the process sets
The process of degree forms standby scheduling queue；
Step 2: judging whether process scheduling request terminates, and is such as not finished and performs the next step suddenly, otherwise finishing scheduling；
Step 3: the process of the dispatching concentration is scheduled；
Step 4: the process that scheduling is completed enters the tail of the queue of the standby scheduling queue；
Step 5: scheduling collection is emptied；
Step 6: the process that selection can carry out concurrent scheduling from the standby scheduling queue constructs new scheduling collection again, and
Go to step two；
Scheduling collection described in the step 1 is made of the process that can obtain resource in the process sets simultaneously；Further, institute
State the composition that collection is dispatched in step 1 the following steps are included:
Step A: the process replication in the process sets to interim process is concentrated；
Step B: it is concentrated from the interim process take out a process at random；
Step C: the process is put into the dispatching concentration；
Step D: the impact factor of process described in the interim process collection is deleted；The impact factor is when a certain process a quilt
After calling, k resource in resource set B can be occupied, and then the money collected in A in other j processes acquisition resource set B that influences the course
Source then claims this j process ai+1..., ai+jFor impact factor of the process a in resource set B；
Step E: the interim process integrates not as empty set, then jumps to step B or the interim process integrates as empty set, then terminate
The composition of the scheduling collection.
2. the method for computer multi-process equity dispatching as described in claim 1, which is characterized in that dispatched described in step 6
The composition of collection the following steps are included:
Step F: process is chosen from the standby scheduling queue and is put into the dispatching concentration；
Step G: process unselected in the standby scheduling queue is arranged by former sequence.
3. the method for computer multi-process equity dispatching as claimed in claim 2, which is characterized in that further, the step
In rapid six, the process of the scheduling collection is successively chosen from the head of the queue of the standby scheduling queue to tail of the queue.
4. the method for computer multi-process equity dispatching as claimed in claim 3, which is characterized in that further, the step
In rapid six, the process of selection is not the impact factor of any process of the dispatching concentration.
5. the method for computer multi-process equity dispatching as described in claim 1, which is characterized in that scheduling collection and described
Standby scheduling queue operates generation by set operation.
6. the method for computer multi-process equity dispatching as described in claim 1, which is characterized in that scheduling collection and described
Standby scheduling queue can be converted mutually.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310567625.1A CN104636200B (en) | 2013-11-14 | 2013-11-14 | A kind of method of computer multi-process equity dispatching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310567625.1A CN104636200B (en) | 2013-11-14 | 2013-11-14 | A kind of method of computer multi-process equity dispatching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636200A CN104636200A (en) | 2015-05-20 |
| CN104636200B true CN104636200B (en) | 2019-03-29 |
Family
ID=53214991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310567625.1A Active CN104636200B (en) | 2013-11-14 | 2013-11-14 | A kind of method of computer multi-process equity dispatching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636200B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1790270A (en) * | 2005-12-14 | 2006-06-21 | 浙江大学 | Java virtual machine implementation method supporting multi-process |
| CN101145112A (en) * | 2007-11-01 | 2008-03-19 | 北京工业大学 | Real-time system task scheduling method |
| CN105204938A (en) * | 2015-11-02 | 2015-12-30 | 重庆大学 | Data-intensive process scheduling method of memory access |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102591712B (en) * | 2011-12-30 | 2013-11-20 | 大连理工大学 | Decoupling parallel scheduling method for rely tasks in cloud computing |
-
2013
- 2013-11-14 CN CN201310567625.1A patent/CN104636200B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1790270A (en) * | 2005-12-14 | 2006-06-21 | 浙江大学 | Java virtual machine implementation method supporting multi-process |
| CN101145112A (en) * | 2007-11-01 | 2008-03-19 | 北京工业大学 | Real-time system task scheduling method |
| CN105204938A (en) * | 2015-11-02 | 2015-12-30 | 重庆大学 | Data-intensive process scheduling method of memory access |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636200A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106358003B (en) | A kind of video analysis accelerated method based on thread level production line | |
| US9798830B2 (en) | Stream data multiprocessing method | |
| CN102473118B (en) | Information processing system | |
| CN103324525B (en) | Method for scheduling task under a kind of cloud computing environment | |
| CN100576177C (en) | Bidirectional grade gridding resource scheduling method based on the QoS constraint | |
| CN102364447B (en) | Operation scheduling method for optimizing communication energy consumption among multiple tasks | |
| CN102053870A (en) | Systems and methods for affinity driven distributed scheduling of parallel computations | |
| CN103823706B (en) | A kind of plant model analog simulation real-time scheduling method based on RTLinux | |
| CN105677461B (en) | Mixing key task dispatching method based on criticality | |
| CN102597950A (en) | Hardware-based scheduling of GPU work | |
| CN102193779A (en) | MPSoC (multi-processor system-on-chip)-oriented multithread scheduling method | |
| CN103078941A (en) | Task scheduling method and system for distributed computing system | |
| CN104239147B (en) | One kind processing deadlock round-robin method and system | |
| CN102495722A (en) | XML (extensible markup language) parallel parsing method for multi-core fragmentation | |
| CN109840149A (en) | Method for scheduling task, device, equipment and storage medium | |
| CN106201676A (en) | Task allocation method and device | |
| CN103279622A (en) | Method for dispatching distributed simulation models in aircraft simulation system | |
| CN101833439B (en) | Parallel computing hardware structure based on separation and combination thought | |
| CN104090826B (en) | Task optimization deployment method based on correlation | |
| CN105786610B (en) | The method that computation-intensive task is unloaded into Cloud Server | |
| CN104156505B (en) | A kind of Hadoop cluster job scheduling method and devices based on user behavior analysis | |
| CN104636200B (en) | A kind of method of computer multi-process equity dispatching | |
| CN103810041A (en) | Parallel computing method capable of supporting dynamic compand | |
| CN107220114A (en) | Distributed resource scheduling method based on resource United Dispatching | |
| CN104077182B (en) | Strategy for scheduling tasks of same priority |
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 |