CN109240813A - Task schedule and task immigration method in a kind of mobile cloud computing - Google Patents
Task schedule and task immigration method in a kind of mobile cloud computing Download PDFInfo
- Publication number
- CN109240813A CN109240813A CN201810955573.8A CN201810955573A CN109240813A CN 109240813 A CN109240813 A CN 109240813A CN 201810955573 A CN201810955573 A CN 201810955573A CN 109240813 A CN109240813 A CN 109240813A
- Authority
- CN
- China
- Prior art keywords
- task
- state
- child
- node
- num
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
-
- 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
- G06F9/4856—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses the task schedules and task immigration method in a kind of mobile cloud computing, its step includes: to execute all task nodes in mobile terminal, record the energy consumption on current mobile terminal, calculate the energy consumption that mobile terminal is saved after each transportable task immigration to cloud and by the total deadline finish_time of maximum priority scheduling algorithm calculating task, select saving energy consumption at most and meeting time-constrain for task, by the task immigration to cloud, if without qualified task node, then terminating method process, it repeats the above steps until all tasks are fully completed.
Description
Technical field
The present invention relates to mobile field of cloud calculation, more particularly, to the task schedule in a kind of mobile cloud computing and appoint
Business moving method.
Background technique
With the development of internet and mechanics of communication, mobile applications provide the calculating, storage and electricity of mobile device
The demand in source is increasing.However, the resource of mobile device tends not to the high need for meeting mobile application due to the limitation of volume
It asks.However, the development of cloud computing, so that the partial task in mobile device can migrate to cloud execution.Although in mobile device
Task can migrate to cloud, but also bring communication energy consumption and communication delay.Existing solution is largely directed to
A kind of situation in task immigration or task schedule.It is worth aiming at the problem that joint migration in mobile cloud computing is with scheduling
The project further studied.
Summary of the invention
For the combined dispatching and migration problem in mobile cloud computing, existing solution is transported by optimizer
It calculates, although precision has greatly improved, the Resources Consumption of the time and operation consumption that run consumption are too big.The present invention proposes
Task schedule and task immigration method in a kind of mobile cloud computing, the method are executed as at the beginning of using whole tasks in mobile terminal
Begin solution, the energy consumption saving that transportable task is run beyond the clouds is gradually calculated, successively by the maximum task immigration of saving to cloud
End, according to the call duration time between task, updates the energy consumption saving of each task.
The purpose of the present invention is refer under following constraint condition: in the time range that regulation completes general assignment, moving to
Task on Cloud Server can execute parallel with the task on mobile terminal, can also be parallel with the task of other on Cloud Server
It executes.The multiple tasks of mobile terminal can only be executed serially.Just exist when two task nodes are respectively at different ends, between them logical
Believe time and communication energy consumption, otherwise call duration time and communication energy consumption are 0.Solve the problems, such as be: under above-mentioned constraint condition, look for
To a migration strategy, after completion task, so that the energy consumption in mobile terminal is minimum.
Task schedule and task immigration method in a kind of mobile cloud computing, step include:
S1: all task nodes are executed in mobile terminal;
S2: the energy consumption on current mobile terminal is recorded;
S3: the energy consumption that mobile terminal is saved after each transportable task immigration to cloud is calculated and by maximum priority
The total deadline finish_time of dispatching algorithm calculating task;
S4: selecting saving energy consumption at most and meeting time-constrain for task, by the task immigration to cloud, if without item is met
The task node of part, then terminating method process;
S5: repeat the above steps S1, S2, S3, S4.
Further, a kind of maximum priority scheduling algorithm, step described in step S3 include:
S3.1: defining the priority of task node first, marks by the sequence of breadth first traversal to each task node
Serial number;
S3.2: defining task node, there are five types of states, indicate the state that task node i is presently in State (i);
The total deadline finish_time of S3.3 initialization task;
S3.4 is executed and is gathered around top-priority ready task in all ready tasks and mobile terminal on cloud;
S3.5 updates the state of each task, total deadline of more new task when there is a task node to complete;
S3.6 calculates and returns total deadline finish_time of task when all tasks are completed.
Further, the priority that task node is defined described in step S3.1, by the sequence of breadth first traversal to every
A task node marking serial numbers, specifically includes the following steps:
If the serial number of i before the serial number of j, is denoted as i < j;Child's mesh of task node is denoted as Child_num (i), with
And child's number of task node j is denoted as Child_num (j);The priority of task node i is denoted as prior (i), task node j
Priority be denoted as prior (j);The priority of task node defines;
(1) as Child_num (i) > Child_num (j), then prior (i) > prior (j);
(2) as Child_num (i)=Child_num (j) and i>j, then prior (i)<prior (j);
(3) if Child_num (i)=Child_num (j) and i<j, prior (i)>prior (j).
Further, task node is defined described in step S3.2, and there are five types of states, are defined as follows: non-active
Indicate that unactivated state, active indicate that state of activation, ready expression are ready, runnable expression can be run, finished table
Show and is completed;
After if all father nodes of task node i had executed, State (i)=active;
After if all father nodes of task i had executed on cloud, State (i)=runnable;
If after all father nodes of task i have executed in ready set, and possessing highest priority, then State (i)=
runnable;
If priority is not highest after all father nodes of task i have executed in ready set, then State (i)=
ready;
If all father nodes of task i are without all having executed, State (i)=non-active;
If task node i has had been carried out, State (i)=finished.
Compared with prior art, the beneficial effect of technical solution of the present invention is:
The runing time of inventive algorithm is much smaller than the runing time of traditional optimizer, and consumption resource substantially reduces, simultaneously
Keep preferable accuracy.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Fig. 2 is a task image with 15 subtasks in mobile device.
Specific embodiment
The following further describes the technical solution of the present invention with reference to the accompanying drawings and examples.
Embodiment
Task schedule and task immigration method in mobile cloud computing, the thought of dispatching algorithm are as follows: migrated in Given task
In the case of, can be executed parallel with other tasks on cloud in the task of ready state, on mobile terminal in ready state and
The possessing highest priority of the task first carries out.When one task of every completion, the completion status of more new task is found out and is currently in just
The task of not-ready status.It constantly repeats the above process, finally finds out the task scheduling approach under time restriction, and calculating task
Total deadline.The heuritic approach thought of transition process: a task is migrated to cloud from cloud every time, will be ensured
After migration, the total energy consumption of mobile terminal is smaller than before, and total deadline after migration will meet time-constrain.By time
All nodes on task image are gone through, are repeated the above process, obtain final migration, scheduling scheme.
The algorithm is divided into two parts, and first part is mainly dispatching algorithm, for providing one after the completion of calculating general assignment
Kind approximation method, second part moving method, wherein algorithm needs to call the dispatching algorithm of front.
Specific steps include: as shown in Fig. 2, moving a task image with 15 subtasks in mobile device.It is solid
Node indicates that the task can only execute on mobile terminal.In addition to solid node, cloud execution can migrate to.Hollow node indicates
The task is completed beyond the clouds;Shaded nodes indicate that the task is completed on mobile terminal.For there is the two of communication tasks, if with
Mobile terminal is realized together beyond the clouds, then the call duration time and communication energy consumption between the two tasks are ignored;If there is communication
Two tasks realize at different ends, then the call duration time and communication energy consumption between calculating task.
Referring to Fig. 1, task schedule and task immigration method in a kind of mobile cloud computing, step includes:
S1: all task nodes are executed in mobile terminal;
S2: the energy consumption on current mobile terminal is recorded;
S3: the energy consumption that mobile terminal is saved after each transportable task immigration to cloud is calculated and by maximum priority
The total deadline finish_time of dispatching algorithm calculating task.
Wherein, the total deadline step of maximum priority scheduling algorithm calculating task includes:
1. defining the priority of task node first, sequence is put on to each task node by the sequence of breadth first traversal
Number.If the serial number of i is denoted as i < j before the serial number of j.Secondly, child's number of task node i is denoted as Child_num (i),
And child's number of task node j is denoted as Child_num (j).Then, the priority of task node i is denoted as prior (i), together
Reason, the priority of task node j are denoted as prior (j).Task node priority size is defined as follows:
A. if Child_num (i) > Child_num (j), prior (i) > prior (j);
B. if Child_num (i)=Child_num (j) and i>j, prior (i)<prior (j);
C. if Child_num (i)=Child_num (j) and i<j, prior (i)>prior (j).
2. to several states locating for node of going out on missions: inactive (non-active), activating (active), is ready
(ready), (runnable) can be run, (finished) is completed.We indicate that task node i is presently in State (i)
State:
A. if after all father nodes of task node i have executed, State (i)=active;
B. if after all father nodes of task i have executed on cloud, State (i)=runnable;
C. if after all father nodes of task i have executed in ready set, and possess highest priority, then State (i)
=runnable;
D. if priority is not highest after all father nodes of task i have executed in ready set, then State (i)=
read;
E. if all father nodes of task i are without all having executed, State (i)=non-acti;
F. if task node i has had been carried out, State (i)=finished.
3. the total deadline finish_time of initialization task.
4. executing and gathering around top-priority ready task in all ready tasks and mobile terminal on cloud.
5. updating the state of each task, total deadline of more new task when there is a task node to complete.
6. calculating and returning total deadline finish_time of task when all tasks are completed.
S4: selecting saving energy consumption at most and meeting time-constrain for task, by the task immigration to cloud, if without item is met
The task node of part, then terminating method process.
S5: repeat the above steps S1, S2, S3, S4, until all tasks are fully completed.
Claims (4)
1. a kind of task schedule in mobile cloud computing and task immigration method, which is characterized in that its step includes:
S1: all task nodes are executed in mobile terminal;
S2: the energy consumption on current mobile terminal is recorded;
S3: the energy consumption that mobile terminal is saved after each transportable task immigration to cloud is calculated and by maximum priority scheduling
The total deadline finish_time of algorithm calculating task;
S4: saving energy consumption at most and meeting time-constrain for task is selected, by the task immigration to cloud, if without qualified
Task node, then terminating method process;
S5: repeat the above steps S1, S2, S3, S4.
2. task schedule and task immigration method in a kind of mobile cloud computing according to claim 1, which is characterized in that
Maximum priority scheduling algorithm, step described in step S3 include:
S3.1: defining the priority of task node first, gives each task node marking serial numbers by the sequence of breadth first traversal;
S3.2: defining task node, there are five types of states, indicate the state that task node i is presently in State (i);
The total deadline finish_time of S3.3 initialization task;
S3.4 is executed and is gathered around top-priority ready task in all ready tasks and mobile terminal on cloud;
S3.5 updates the state of each task, total deadline of more new task when there is a task node to complete;
S3.6 calculates and returns total deadline finish_time of task when all tasks are completed.
3. a kind of maximum priority scheduling algorithm according to claim 2, which is characterized in that defined described in step S3.1
The priority of task node gives each task node marking serial numbers by the sequence of breadth first traversal, specifically includes the following steps:
If the serial number of i before the serial number of j, is denoted as i < j;Child's mesh of task node is denoted as Child_num (i), Yi Jiren
Child's number of business node j is denoted as Child_num (j);The priority of task node i is denoted as prior (i), and task node j's is excellent
First power is denoted as prior (j);The priority of task node defines;
(1) as Child_num (i) > Child_num (j), then prior (i) > prior (j);
(2) as Child_num (i)=Child_num (j) and i>j, then prior (i)<prior (j);
(3) if Child_num (i)=Child_num (j) and i<j, prior (i)>prior (j).
4. a kind of maximum priority scheduling algorithm according to claim 2, which is characterized in that defined described in step S3.2
Task node is defined as follows there are five types of state: non-active indicate unactivated state, active indicate state of activation,
Ready indicates that ready, runnable expression can be run, finished expression is completed;
After if all father nodes of task node i had executed, State (i)=active;
After if all father nodes of task i had executed on cloud, State (i)=runnable;
If after all father nodes of task i have executed in ready set, and possessing highest priority, then State (i)=
runnable;
If priority is not highest after all father nodes of task i have executed in ready set, then State (i)=ready;
If all father nodes of task i are without all having executed, State (i)=non-active;
If task node i has had been carried out, State (i)=finished.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810955573.8A CN109240813B (en) | 2018-08-21 | 2018-08-21 | Task scheduling and task migration method in mobile cloud computing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810955573.8A CN109240813B (en) | 2018-08-21 | 2018-08-21 | Task scheduling and task migration method in mobile cloud computing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109240813A true CN109240813A (en) | 2019-01-18 |
CN109240813B CN109240813B (en) | 2021-12-24 |
Family
ID=65070123
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810955573.8A Active CN109240813B (en) | 2018-08-21 | 2018-08-21 | Task scheduling and task migration method in mobile cloud computing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109240813B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111402353A (en) * | 2020-03-11 | 2020-07-10 | 浙江大学 | Cloud-end drawing computing framework based on self-adaptive virtual drawing assembly line |
CN114281426A (en) * | 2021-12-21 | 2022-04-05 | 中国联合网络通信集团有限公司 | Task unloading method and device, electronic equipment and readable storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102467532A (en) * | 2010-11-12 | 2012-05-23 | 中国移动通信集团山东有限公司 | Task processing method and task processing device |
CN103293967A (en) * | 2012-02-29 | 2013-09-11 | 陕西省地方电力(集团)有限公司 | Multi-task control method for intelligent meter reading terminal |
WO2016045515A1 (en) * | 2014-09-23 | 2016-03-31 | 同济大学 | Cloud task scheduling algorithm based on user satisfaction |
CN107360235A (en) * | 2017-07-17 | 2017-11-17 | 广东工业大学 | A kind of task immigration method based on reliability classification |
CN107436811A (en) * | 2017-07-07 | 2017-12-05 | 广东工业大学 | It is related to the task immigration method of task scheduling in mobile cloud problem |
CN107562527A (en) * | 2017-08-28 | 2018-01-09 | 北京翼辉信息技术有限公司 | A kind of Real-Time Task Schedule Algorithm of SMP on RTOS |
-
2018
- 2018-08-21 CN CN201810955573.8A patent/CN109240813B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102467532A (en) * | 2010-11-12 | 2012-05-23 | 中国移动通信集团山东有限公司 | Task processing method and task processing device |
CN103293967A (en) * | 2012-02-29 | 2013-09-11 | 陕西省地方电力(集团)有限公司 | Multi-task control method for intelligent meter reading terminal |
WO2016045515A1 (en) * | 2014-09-23 | 2016-03-31 | 同济大学 | Cloud task scheduling algorithm based on user satisfaction |
CN107436811A (en) * | 2017-07-07 | 2017-12-05 | 广东工业大学 | It is related to the task immigration method of task scheduling in mobile cloud problem |
CN107360235A (en) * | 2017-07-17 | 2017-11-17 | 广东工业大学 | A kind of task immigration method based on reliability classification |
CN107562527A (en) * | 2017-08-28 | 2018-01-09 | 北京翼辉信息技术有限公司 | A kind of Real-Time Task Schedule Algorithm of SMP on RTOS |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111402353A (en) * | 2020-03-11 | 2020-07-10 | 浙江大学 | Cloud-end drawing computing framework based on self-adaptive virtual drawing assembly line |
CN111402353B (en) * | 2020-03-11 | 2020-12-25 | 浙江大学 | Cloud-end drawing calculation method based on self-adaptive virtualization drawing production line |
WO2021179780A1 (en) * | 2020-03-11 | 2021-09-16 | 浙江大学 | Cloud-terminal rendering calculation method based on self-adaptive virtualization rendering pipeline |
CN114281426A (en) * | 2021-12-21 | 2022-04-05 | 中国联合网络通信集团有限公司 | Task unloading method and device, electronic equipment and readable storage medium |
CN114281426B (en) * | 2021-12-21 | 2023-05-16 | 中国联合网络通信集团有限公司 | Task unloading method and device, electronic equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109240813B (en) | 2021-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10031775B2 (en) | Backfill scheduling for embarrassingly parallel jobs | |
CN105956021A (en) | Automated task parallel method suitable for distributed machine learning and system thereof | |
CN104536811B (en) | Method for scheduling task based on HIVE tasks and device | |
CN104506620A (en) | Extensible automatic computing service platform and construction method for same | |
CN105719126A (en) | System and method for internet big data task scheduling based on life cycle model | |
CN107220780B (en) | Heterogeneous task execution sequence optimization method in crowdsourcing system | |
CN105205105A (en) | Data ETL (Extract Transform Load) system based on storm and treatment method based on storm | |
CN103729748B (en) | Progress computational methods and its algorithm based on comprehensive network plans are realized | |
CN102364447A (en) | Operation scheduling method for optimizing communication energy consumption among multiple tasks | |
CN103577474A (en) | Method and system for updating database | |
CN109240813A (en) | Task schedule and task immigration method in a kind of mobile cloud computing | |
CN102945516A (en) | Multistage network planned schedule analysis method | |
CN111062467A (en) | Automatic neural network subgraph segmentation method applied to AI heterogeneous compiler | |
CN103914556A (en) | Large-scale graph data processing method | |
CN106095552A (en) | A kind of Multi-Task Graph processing method based on I/O duplicate removal and system | |
CN114356578A (en) | Parallel computing method, device, equipment and medium for natural language processing model | |
CN104360906B (en) | A kind of High Level Synthesis dispatching method based on difference constrained system Yu iteration mould | |
Liu et al. | An independent task scheduling algorithm in heterogeneous multi-core processor environment | |
Bansal et al. | Application of artificial bee colony algorithm using hadoop | |
CN103049310A (en) | Multi-core simulation parallel accelerating method based on sampling | |
CN102253837A (en) | Object tree-based software framework designing technology | |
CN110502337B (en) | Optimization system for shuffling stage in Hadoop MapReduce | |
Yuan et al. | Dynamic parallel machine scheduling using the learning agent | |
CN109902403A (en) | A kind of integrated dispatch method based on Petri network and heuristic value | |
CN112149826B (en) | Profile graph-based optimization method in deep neural network inference calculation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |