CN107908466A - A kind of fast dispatch method of BoT tasks under cloud environment - Google Patents
A kind of fast dispatch method of BoT tasks under cloud environment Download PDFInfo
- Publication number
- CN107908466A CN107908466A CN201711023219.3A CN201711023219A CN107908466A CN 107908466 A CN107908466 A CN 107908466A CN 201711023219 A CN201711023219 A CN 201711023219A CN 107908466 A CN107908466 A CN 107908466A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- msubsup
- time
- beginning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/485—Task life-cycle, e.g. stopping, restarting, resuming execution
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a kind of fast dispatch method of BoT tasks under cloud environment.This method step is as follows：Step 1, information is read：Read all parameters of BoT tasks；Step 2, job scheduling：Parameter based on reading, between determining at the beginning of all operations, dispatching method from the subsequent time at the failure moment of operation at the beginning of definite operation between, it is necessary to consider master budget and meet formula constrain；Step 3, job execution：Each operation is performed according to the result of step 2.The fast dispatch method of BoT tasks under cloud environment of the present invention, it is contemplated that the processing time of each operation in BoT tasks, between can quickly and efficiently determining at the beginning of each operation and minimizes makespan.
Description
Technical field
The present invention relates to a kind of fast dispatch method of BoT tasks under field of cloud computer technology, particularly cloud environment.
Background technology
Cloud computing is intended to available resources (including computing resource, platform resource and application resource) by internet to service
Form be supplied to user.User by " with it is how many, pay it is how many " in a manner of use these resources.Therefore, cloud computing is particularly suitable for
Handling those needs the task of a large amount of computing resources.
Bag-of-Tasks (BoT) task be it is a kind of comprising it is multiple can parallel processing operation task.BoT tasks refer to wrap
Containing it is multiple can parallel processing operation task, be widely present in computer video, picture processing, Monte-Carlo Simulation science and
Engineering field.In fact, the computing device that many users (including enterprise or mechanism) have some limited resources (can be described as
Private clound), but cannot efficiently dispatch BoT tasks and Maximal Makespan (i.e. makespan) is minimized.
The content of the invention
It is an object of the invention to provide a kind of fast dispatch method of BoT tasks under cloud environment, so that fast and effeciently
Between determining at the beginning of each operation and minimize makespan.
The technical solution for realizing the object of the invention is：The fast dispatch method of BoT tasks, step under a kind of cloud environment
It is as follows：
Step 1, information is read：Read all parameters of BoT tasks；
Step 2, job scheduling：Parameter based on reading, between determining at the beginning of all operations；
Step 3, job execution：Each operation is performed according to the result of step 2.
Further, all parameters of the BoT tasks include：The processing time of each operation in all BoT tasks, each
VM types needed for business.
Further, the parameter based on reading, between determining at the beginning of all operations, used job scheduling side
Method is as follows：
An assuming that private clound CP0, private clound provides k VM type, is respectively VM1,VM2,…,VMk；
Each VMqAll described with two configuration parameters：The CPU quantity CPU that the VM can be providedqWith the quantity of memory
Memq, q=1,2 ..., k；
One shares n BoT tasks a1,a2,…,an, each task aiInclude TiA operationWith a ginseng
Number xiqRepresent aiWhether VM is neededq, x if necessaryiq=1；Otherwise, xiq=0；The processing time of each operation is rij, wherein
I=1,2 ..., n, j=1,2 ..., Ti；
Time shaft is expressed as CPU with 1 granularity discretization when small, private clound maximum CPU quantity and amount of memory*With
Mem*, i.e., the resource consumed in any point private clound on a timeline is no more than CPU*And Mem*；
If task aiCompletion date be ci, the maximum time stamp S=max of time shaftI=1,2 ..., nci；If cijFor operation tij
Completion date, therefore have：
If operation tijAt the beginning of between be stij, then cijIt is as follows：
cij=stij+rij (2)
Define decision variable zijs, i=1,2 ..., n, j=1,2 ..., Ti, s=0,1 ..., S；zijs=1 shows operation tij
At the beginning of between be s, otherwise zijs=0, operation tijAt the beginning of between be described as integer programming problem, i.e. Minimize the
makespan(Cmax)：
Cmax=maxI=1,2 ..., nci (3)
s.t.
Above-mentioned job scheduling method from the subsequent time at the failure moment of operation at the beginning of definite operation between, it is necessary to tie
Close private clound maximum CPU quantity CPU*With amount of memory Mem*, and meet the constraint of formula (3)~(5).
Further, job scheduling method is specific as follows：
According to configuration requirement of each operation to VM and processing time, task is rearranged according to the ascending order of processing time
One job execution queue, provides the VM for meeting mission requirements successively, between determining at the beginning of each task, to ensure whole appoint
Business minimizes makespan, and job scheduling comprises the following steps that：
(1) a job execution queue is received；
(2) if queue is sky, go to (7)；Otherwise, go to (3)；
(3) an operation t is taken out from queuei, its desired VM types VMiBy cloud CjThere is provided, processing time timei, turn
To (4)；
(4) remember s=0, go to (5)；
(5) in CjIn, begin look for C from the s momentjIn whether have continuous timeiA period disclosure satisfy that VMiMatch somebody with somebody
Put requirement, if it is possible to find, then it is t to record siEarliest start time, and by the resource of time period according to VMiConfiguration
Distribution, goes to (2)；Otherwise, go to (6)；
(6) remember that last can not meet VMiIt is e at the time of configuration, makes s=e+1, goes to (5)；
(7) terminate.
Compared with prior art, the present invention its remarkable advantage is：The processing time of each operation in BoT tasks is considered,
Between can quickly and efficiently determining at the beginning of each operation and minimize makespan.
Brief description of the drawings
Fig. 1 is the flow chart of the fast dispatch method of BoT tasks under cloud environment of the present invention.
Fig. 2 is that each operation of improved method in the embodiment of the present invention handles schematic diagram.
Embodiment
The fast dispatch method of BoT tasks, problem to be solved can be described below under cloud environment of the present invention：
There is a private clound CP0.Private clound provides k VM type, is respectively VM1,VM2,…,VMk.Each VMq(q=
1,2 ..., k) (CPU all is described with two configuration parametersqAnd Memq), CPU quantity that the VM can be provided and interior is represented respectively
The quantity deposited.One shares n BoT tasks a1,a2,…,an.Each task ai(i=1,2 ..., n) include TiA operationWith a parameter xiqRepresent aiWhether VM is neededq, it is necessary to xiq=1；Otherwise, xiq=0.The place of each operation
The reason time is rij(i=1,2 ..., n, j=1,2 ..., Ti)。
Time shaft is with 1 granularity discretization when small.Private clound maximum CPU quantity and amount of memory are represented by CPU*With
Mem*.That is, the resource consumed in any point private clound on a timeline is no more than CPU*And Mem*。
If task aiCompletion date be ci, the maximum time stamp S=max of time shaftI=1,2 ..., nci.If cijFor operation tij
Completion date, still have
If operation tijAt the beginning of between be stij, then cijIt can be calculated as below
cij=stij+rij (2)
It is defined as follows decision variable zijs(i=1,2 ..., n, j=1,2 ..., Ti, s=0,1 ..., S).zijs=1 shows
Operation tijAt the beginning of between be s, otherwise zijs=0. we this problem can be described as integer programming problem, i.e. Minimize
the makespan(Cmax):
Cmax=maxI=1,2 ..., nci (3)
s.t.
With reference to Fig. 1, the fast dispatch method of BoT tasks, comprises the following steps under cloud environment of the present invention：
Step 1, information is read：Read all parameters of BoT tasks；
Step 2, job scheduling：Parameter based on reading, between determining at the beginning of all operations；
The job scheduling method from the subsequent time at the failure moment of operation at the beginning of definite operation between, it is necessary to tie
Close private clound maximum CPU quantity CPU*With amount of memory Mem*, and meet the constraint of formula (3)~(5).Job scheduling method has
Body is as follows：
According to configuration requirement of each operation to VM and processing time, task is rearranged according to the ascending order of processing time
One job execution queue, provides the VM for meeting mission requirements successively, between determining at the beginning of each task, to ensure whole appoint
Business minimizes makespan, and job scheduling comprises the following steps that：
(1) a job execution queue is received；
(2) if queue is sky, go to (7)；Otherwise, go to (3)；
(3) an operation t is taken out from queuei, its desired VM types VMiBy cloud CjThere is provided, processing time timei, turn
To (4)；
(4) remember s=0, go to (5)；
(5) in CjIn, begin look for C from the s momentjIn whether have continuous timeiA period disclosure satisfy that VMiMatch somebody with somebody
Put requirement, if it is possible to find, then it is t to record siEarliest start time, and by the resource of time period according to VMiConfiguration
Distribution, goes to (2)；Otherwise, go to (6)；
(6) remember that last can not meet VMiIt is e at the time of configuration, makes s=e+1, goes to (5)；
(7) terminate.
Step 3, job execution：Each operation is performed according to the result of step 2.
The present invention is described in further detail with reference to specific embodiment.
Embodiment 1
Step 1, information is read：
Equipped with a BoT task a, task includes 5 operations, i.e. a={ t1,t2,t3,t4,t5}
CP0The resource upper limit it is as shown in table 1, the configuration of various VM is as shown in table 2, in this example the VM of all job requirements
By CP0There is provided.
The resource upper limit that 1 cloud of table provides
| Cloud | CPU quantity | Amount of memory |
| CP0 | 3 | 3 |
The configuration of 2 VM types of table and unit price
| VM types | CPU quantity | Amount of memory |
| VM1 | 1 | 1 |
| VM2 | 2 | 1 |
| VM3 | 3 | 1 |
Each operation is as shown in table 3 to the types entail of VM and processing time
Types entail and processing time of the 3 each operation of table to VM
| Operation | VM types | Processing time |
| t1 | VM1 | 1 |
| t2 | VM1 | 1 |
| t3 | VM2 | 1 |
| t4 | VM3 | 1 |
| t5 | VM1 | 3 |
Step 2, job scheduling：
According to configuration requirement and processing time of each operation of table 2 and table 3 to VM, the liter by task according to processing time
Sequence arrangement (i.e. short job priority) forms a job execution queue.From t1~t5The VM for meeting mission requirements is provided successively, is determined
Between at the beginning of each task, to ensure that whole task minimizes makespan.The specific method of job scheduling is described below：
(1) queue (first in first out structure) of a job execution is received
(2) if queue is sky, go to (7)；Otherwise, go to (3)
(3) an operation t is taken out from queuei, its desired VM types VMiBy cloud CjThere is provided, processing time timei, turn
To (4)
(4) remember s=0, go to (5)
(5) in CjIn, begin look for C from the s momentjIn whether have continuous timeiA period can meet VMiMatch somebody with somebody
Put requirement.If can find, record s is tiEarliest start time, and by the resource of time period according to VMiConfiguration
Distribution, goes to (2)；Otherwise, go to (6)
(6) remember that last can not meet VMiIt is e at the time of configuration, makes s=e+1, goes to (5)
(7) terminate
According to the method described above, at the beginning of each operation between and makespan as shown in Fig. 2, idiographic flow is as follows：
a)t1And t2The configuration for meeting to perform can just be possessed at 0 moment, 0 moment was t1And t2At the beginning of between.
B) there is t at 0 moment1And t2Task will be handled, and remaining CPU only remains one, can not meet t3Configuration requirement.
It can just meet t to 1 moment3Configuration requirement, then 1 moment was t3Earliest processing time.
C) it is same b) equally, 2 moment were t4Earliest processing time
t5Meet that CPU is configured at 0 moment, also meet that CPU is configured at 1 moment, detected at 2 moment, find in 2 to 3 intervals
CPU is occupied, that is, t5It can not meet VM at 2 moment1Configuration requirement, then need again since redefining 1 moment
Time.It is known however that in 2 moment processing failures (moment is known as the failure moment of the operation), thus can directly from 3 when
Carve and find (compared to being begun look for again from 1 moment, the method is more quick) again, then finding can in 3 to 6 time interval
To handle t5, then 3 moment t5It is the time started.
Step 3, job execution：Each operation is performed according to the result of step 2.
To sum up, the present invention considers the processing time of each operation in BoT tasks, can quickly and efficiently determine each work
Between at the beginning of industry and minimize makespan.
Claims (4)
1. a kind of fast dispatch method of BoT tasks under cloud environment, it is characterised in that step is as follows：
Step 1, information is read：Read all parameters of BoT tasks；
Step 2, job scheduling：Parameter based on reading, between determining at the beginning of all operations；
Step 3, job execution：Each operation is performed according to the result of step 2.
2. the fast dispatch method of BoT tasks under cloud environment according to claim 1, it is characterised in that the BoT tasks
All parameters include：The processing time of each operation in all BoT tasks, the VM types of each required by task.
3. the fast dispatch method of BoT tasks under cloud environment according to claim 1, it is characterised in that described based on reading
The parameter taken, between determining at the beginning of all operations, used job scheduling method is as follows：
An assuming that private clound CP0, private clound provides k VM type, is respectively VM1, VM2..., VMk；
Each VMqAll described with two configuration parameters：The CPU quantity CPU that the VM can be providedqWith the quantity Mem of memoryq, q
=1,2 ..., k；
One shares n BoT tasks a1, a2..., an, each task aiInclude TiA operationWith a parameter xiq
Represent aiWhether VM is neededq, x if necessaryiq=1；Otherwise, xiq=0；The processing time of each operation is rij, wherein i=1,
2 ..., n, j=1,2 ..., Ti；
Time shaft is expressed as CPU with 1 granularity discretization when small, private clound maximum CPU quantity and amount of memory*And Mem*,
The resource consumed in any point private clound i.e. on a timeline is no more than CPU*And Mem*；
If task aiCompletion date be ci, the maximum time stamp S=max of time shaftI=1,2 ..., n ci；If ciiFor operation tijIt is complete
Between man-hour, therefore have：
<mrow>
<msub>
<mi>c</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<msub>
<mi>max</mi>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1,2</mn>
<mo>,</mo>
<mo>&CenterDot;</mo>
<mo>&CenterDot;</mo>
<mo>&CenterDot;</mo>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
</mrow>
</msub>
<msub>
<mi>c</mi>
<mi>ij</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
If operation tijAt the beginning of between be stij, then ciiIt is as follows：
cii=stij+rij (2)
Define decision variable zijs, i=1,2 ..., n, j=1,2 ..., Ti, s=0,1 ..., S；zijs=1 shows operation tijOpen
Time beginning is s, otherwise zijs=0, operation tijAt the beginning of between be described as integer programming problem, i.e. Minimize the
makespan(Cmax)：
Cmax=maxi=1,2 ..., n ci (3)
s.t.
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>q</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</msubsup>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
</msubsup>
<msub>
<mi>z</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
<mi>s</mi>
</mrow>
</msub>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>q</mi>
</mrow>
</msub>
<msub>
<mi>CPU</mi>
<mi>q</mi>
</msub>
<mo>&le;</mo>
<msup>
<mi>CPU</mi>
<mo>*</mo>
</msup>
<mo>,</mo>
<mi>s</mi>
<mo>=</mo>
<mn>0</mn>
<mo>,</mo>
<mn>1</mn>
<mo>,</mo>
<mo>...</mo>
<mo>,</mo>
<mi>S</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>q</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</msubsup>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
</msubsup>
<msub>
<mi>z</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
<mi>s</mi>
</mrow>
</msub>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>q</mi>
</mrow>
</msub>
<msub>
<mi>Mem</mi>
<mi>q</mi>
</msub>
<mo>&le;</mo>
<msup>
<mi>Mem</mi>
<mo>*</mo>
</msup>
<mo>,</mo>
<mi>s</mi>
<mo>=</mo>
<mn>0</mn>
<mo>,</mo>
<mn>1</mn>
<mo>,</mo>
<mo>...</mo>
<mo>,</mo>
<mi>S</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
Above-mentioned job scheduling method from the subsequent time at the failure moment of operation at the beginning of definite operation between, it is necessary to combine private
There is cloud maximum CPU quantity CPU*With amount of memory Mem*, and meet the constraint of formula (3)~(5).
4. the fast dispatch method of BoT tasks under cloud environment according to claim 3, it is characterised in that job scheduling side
Method is specific as follows：
According to configuration requirement of each operation to VM and processing time, task is rearranged one according to the ascending order of processing time
Job execution queue, provides the VM for meeting mission requirements, between determining at the beginning of each task, to ensure whole task most successively
Smallization makespan, job scheduling comprise the following steps that：
(1) a job execution queue is received；
(2) if queue is sky, go to (7)；Otherwise, go to (3)；
(3) an operation t is taken out from queuei, its desired VM types VMiBy cloud CjThere is provided, processing time timei, go to
(4)；
(4) remember s=0, go to (5)；
(5) in CjIn, begin look for C from the s momentjIn whether have continuous timeiA period disclosure satisfy that VMiConfiguration will
Ask, if it is possible to find, then it is t to record siEarliest start time, and by the resource of time period according to VMiConfiguration distribution,
Go to (2)；Otherwise, go to (6)；
(6) remember that last can not meet VMiIt is e at the time of configuration, makes s=e+1, goes to (5)；
(7) terminate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711023219.3A CN107908466A (en) | 2017-10-27 | 2017-10-27 | A kind of fast dispatch method of BoT tasks under cloud environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711023219.3A CN107908466A (en) | 2017-10-27 | 2017-10-27 | A kind of fast dispatch method of BoT tasks under cloud environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107908466A true CN107908466A (en) | 2018-04-13 |
Family
ID=61842006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711023219.3A Pending CN107908466A (en) | 2017-10-27 | 2017-10-27 | A kind of fast dispatch method of BoT tasks under cloud environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107908466A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112506634A (en) * | 2020-12-08 | 2021-03-16 | 天津津航计算技术研究所 | Fairness operation scheduling method based on reservation mechanism |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140006534A1 (en) * | 2012-06-27 | 2014-01-02 | Nilesh K. Jain | Method, system, and device for dynamic energy efficient job scheduling in a cloud computing environment |
| CN104657221A (en) * | 2015-03-12 | 2015-05-27 | 广东石油化工学院 | Multi-queue peak-alternation scheduling model and multi-queue peak-alteration scheduling method based on task classification in cloud computing |
| CN105893158A (en) * | 2016-06-08 | 2016-08-24 | 北京工业大学 | Big data hybrid scheduling model on private cloud condition |
| CN106027617A (en) * | 2016-05-11 | 2016-10-12 | 广东浪潮大数据研究有限公司 | Method for implementing dynamic scheduling of tasks and resources in private cloud environment |
| CN107229514A (en) * | 2017-06-01 | 2017-10-03 | 南京邮电大学盐城大数据研究院有限公司 | The dispatching method of BoT tasks with budgetary restraints under a kind of mixing cloud environment |
-
2017
- 2017-10-27 CN CN201711023219.3A patent/CN107908466A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140006534A1 (en) * | 2012-06-27 | 2014-01-02 | Nilesh K. Jain | Method, system, and device for dynamic energy efficient job scheduling in a cloud computing environment |
| CN104657221A (en) * | 2015-03-12 | 2015-05-27 | 广东石油化工学院 | Multi-queue peak-alternation scheduling model and multi-queue peak-alteration scheduling method based on task classification in cloud computing |
| CN106027617A (en) * | 2016-05-11 | 2016-10-12 | 广东浪潮大数据研究有限公司 | Method for implementing dynamic scheduling of tasks and resources in private cloud environment |
| CN105893158A (en) * | 2016-06-08 | 2016-08-24 | 北京工业大学 | Big data hybrid scheduling model on private cloud condition |
| CN107229514A (en) * | 2017-06-01 | 2017-10-03 | 南京邮电大学盐城大数据研究院有限公司 | The dispatching method of BoT tasks with budgetary restraints under a kind of mixing cloud environment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112506634A (en) * | 2020-12-08 | 2021-03-16 | 天津津航计算技术研究所 | Fairness operation scheduling method based on reservation mechanism |
| CN112506634B (en) * | 2020-12-08 | 2022-07-12 | 天津津航计算技术研究所 | Fairness operation scheduling method based on reservation mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8719833B2 (en) | Adaptive demand-driven load balancing | |
| US8595735B2 (en) | Holistic task scheduling for distributed computing | |
| Land | Parameters and sensitivity in workload control | |
| CN103309946B (en) | Multimedia file processing method, Apparatus and system | |
| Land | Cobacabana (control of balance by card-based navigation): A card-based system for job shop control | |
| CN104657220A (en) | Model and method for scheduling for mixed cloud based on deadline and cost constraints | |
| CN107493314A (en) | Service provider system and method based on multi-tenant | |
| Qian et al. | Fast algorithms for scheduling with learning effects and time-dependent processing times on a single machine | |
| Fernandes et al. | Aligning workload control theory and practice: lot splitting and operation overlapping issues | |
| CN103257896B (en) | A kind of Max-D job scheduling method under cloud environment | |
| Thürer et al. | Card-based delivery date promising in high-variety manufacturing with order release control | |
| CN106776025A (en) | A kind of computer cluster job scheduling method and its device | |
| CN109034611A (en) | Back-end software development approach, platform and the system quickly delivered | |
| JP6447611B2 (en) | Schedule management apparatus and method | |
| US8494893B2 (en) | Production plan making program, device, and method | |
| CN107908466A (en) | A kind of fast dispatch method of BoT tasks under cloud environment | |
| CN109214645A (en) | Development approach and system based on module metering | |
| CN104794000B (en) | A kind of job scheduling method and system | |
| US20130235416A1 (en) | Bin-packing approach to production job scheduling on a cloud infrastructure | |
| CN109118059A (en) | Front-end software development approach, platform and the system quickly delivered | |
| US8559040B2 (en) | Method and system for performing statistical throughput control in a print shop | |
| US20210365872A1 (en) | A system and method for management of project portfolios | |
| CN108521524B (en) | Agent collaborative task management method and device, computer equipment and storage medium | |
| US11656611B2 (en) | Production plan creation device and non-transitory computer readable medium | |
| US20140240737A1 (en) | Methods and systems for designing cells in a production environment |
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 |