CN108416465B - Workflow optimization method in mobile cloud environment - Google Patents
Workflow optimization method in mobile cloud environment Download PDFInfo
- Publication number
- CN108416465B CN108416465B CN201810095728.5A CN201810095728A CN108416465B CN 108416465 B CN108416465 B CN 108416465B CN 201810095728 A CN201810095728 A CN 201810095728A CN 108416465 B CN108416465 B CN 108416465B
- Authority
- CN
- China
- Prior art keywords
- workflow
- mobile cloud
- tasks
- cloud environment
- voltage frequency
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000005457 optimization Methods 0.000 title claims abstract description 26
- 238000005265 energy consumption Methods 0.000 claims abstract description 37
- 230000002068 genetic effect Effects 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 230000003044 adaptive effect Effects 0.000 claims description 20
- 230000005540 biological transmission Effects 0.000 claims description 18
- 230000035772 mutation Effects 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000006978 adaptation Effects 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 210000000349 chromosome Anatomy 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0633—Workflow analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Water Supply & Treatment (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Primary Health Care (AREA)
- Evolutionary Biology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Power Sources (AREA)
Abstract
The invention discloses a workflow optimization method under a mobile cloud environment, which comprises the following steps: step S1: preprocessing a workflow submitted by a user; step S2: constructing a workflow model under a mobile cloud environment; step S3: generating an optimal workflow scheduling scheme based on an improved genetic algorithm; step S4: and distributing the tasks to the equipment in the mobile cloud according to the scheduling result. Compared with the prior art, the method and the device have the advantages that the optimal voltage frequency of the device is configured by adjusting the voltage frequency, the obtained voltage frequency is used as a code to carry out optimization search through a genetic algorithm, optimal task scheduling of the device in the mobile cloud environment with respect to reliability and energy consumption is sought, the purpose of multi-objective optimization can be achieved by fast search, and the method and the device are particularly suitable for terminal device users to process a large amount of data and calculate mass data.
Description
Technical Field
The invention relates to the field of mobile cloud computing environments, in particular to a workflow optimization method in a mobile cloud environment.
Background
The mobile cloud computing is a product of the popularization of mobile equipment, the improvement of the performance of the mobile equipment, the improvement of the stability of a mobile network and the like and the development and fusion of a commercial mode anytime and anywhere. The cloud platform is used for reasonably and restrictively distributing tasks in a workflow to terminal users, so that energy consumption and calculation burden of the user terminal are reduced, and the user can conveniently manage and process a large amount of timely information at any time and any place by using mobile equipment.
Compared with the traditional server and desktop equipment, the mobile equipment has limitations on power consumption, storage space and computing capacity, and how to integrate information resources, process mass data and reduce equipment energy consumption is a problem that cannot be avoided in the prior workflow scheduling technology under the mobile cloud environment. In the existing solution to the problem, namely, a mode of providing platform services is adopted, a large amount of information is processed by using the characteristics of resource sharing and high computing capacity of a cloud server, and unlike a workflow processing and scheduling mode of a cloud environment, the problems of task processing time, equipment reliability and energy consumption in a workflow are key points and difficulties in processing workflow scheduling of the environment. A great deal of research work has been done at home and abroad on the workflow scheduling problem in the mobile cloud environment, but most of the workflow scheduling algorithms in the current mobile cloud computing environment only consider unilateral factors, and do not comprehensively consider that the energy consumption of equipment is minimized and the reliability is maximized under the condition that the preset workflow deadline constraint is met in the mobile cloud environment.
Therefore, it is necessary to provide a technical solution to solve the technical problems of the prior art.
Disclosure of Invention
In view of this, it is necessary to provide a workflow optimization method in a mobile cloud environment, where transmission and computational energy consumption are reduced by preprocessing nodes of a workflow in the mobile cloud environment, and an improved genetic algorithm is used to seek optimal task scheduling of energy consumption and reliability, so that the method can combine the advantages of the genetic algorithm and a voltage frequency adjustment technology, and organically integrate the two technologies, so that the method can maximally meet the requirements of users on the reliability and energy consumption of the workflow within constraint completion time.
In order to overcome the defects of the prior art, the technical scheme of the invention is as follows:
a workflow optimization method in a mobile cloud environment comprises the following steps:
step S1: preprocessing a workflow submitted by a user;
step S2: constructing a workflow model under a mobile cloud environment;
step S3: generating an optimal workflow scheduling scheme based on an improved genetic algorithm;
step S4: distributing tasks to the equipment in the mobile cloud according to the scheduling result;
wherein the step S1 further comprises the steps of:
step S11: merging the nodes with overlarge workflow transmission cost according to the workflow transmission cost weight;
step S12: the mobile cloud environment equipment calls a voltage adjustment technology to adjust the most appropriate voltage frequency of the equipment;
the step S11 further includes the steps of:
(1) calculating a weight tau according to the Trans _ P matrix;
(2) according to the sequence of the tasks and the transmission capability of the tasks is higher than the weight tau, the tasks are distributed to the same processor;
the calculation formula of the node merging reference weight in the workflow is as follows:
in the above formula, m represents the number of mobile devices, Trans _ P [ i [ ]][j]Representing a processor piTo pjThe data transmission capability of (2);
in step S12, the voltage frequency is calculated as follows:
wherein f represents the voltage frequency of the mobile cloud device, VddIndicating the voltage, V, supported by the systemtsRepresenting the system voltage threshold, LdRepresents the minimum time for workflow execution, and Z and β represent a constant;
the step 2 further comprises the following steps:
task set of workflow in mobile cloud environment is passed through oneThe directed acyclic graph of sideband weights represents G ═ { T, E, P }, where T ═ T1，t2...，tnIs a set of N tasks, E { (i, j) | ti<tjIs the precedence constraint between tasks, i.e. tjCan not be at tiStarting before completion, P ═ P1,p2,...,pmcRepresents equipment deployed at a mobile cloud; the optimization conditions comprise { Tim, Ene and R }, wherein Tim, Ene and R respectively represent user demand values of a user on time, energy consumption and reliability;
mc and n in the model represent the number of processors used in the cluster and the number of tasks in the task set respectively, wherein each device supports a dynamic voltage frequency adjustment technology, i and j represent task numbers, so that i is greater than or equal to 1 and j is greater than or equal to n;
the step S3 further includes the steps of:
step S31: after encoding the workflow, the adaptive value of each individual of the new population needs to be calculated, and the calculation formula of the adaptive value is as follows:
where f (n, mr) denotes an adaptation value in the scheduling scheme, Rn,mrRepresenting the reliability of the task n on the device mr, RhighRepresents the maximum reliability, R, in the scheduling schemelowIndicating minimum reliability in the scheduling scheme, En,mrRepresenting the energy consumption of the task n on the device mr, EnormalIndicating the energy consumption of the device, psi, produced without the use of dynamic voltage frequency adjustment techniques1And psi2Representing two constants, and1+ψ2＝1；
step S32: dividing the population according to the following steps:
and calculating the adaptive value of each individual, wherein the larger the adaptive value is, the better the individual is, so that the individual with higher adaptability is selected to be divided into elite group members, and the rest are classified into common group members.
Step S33: performing crossing and mutation operations on the population, wherein the specific operations are as follows:
in the cross operation, single-node exchange is carried out on every two individual nodes in the population in sequence to obtain new individuals, and the individuals with higher adaptive values are selected as new solutions according to the adaptive values;
in the mutation operation, firstly, a set needing mutation is separated from the population according to the probability, single nodes on individuals in the set are subjected to mutation operation in sequence to obtain new individuals, and the individuals with higher adaptive values are selected as new solutions according to the adaptive values.
As a preferable technical solution, the step S4 further includes:
and in a preset iteration frequency range, after continuous iteration, the reliability, completion time and energy consumption in the generated new population are in accordance with the expectation of a user, and the tasks and the equipment are mapped and the proper voltage frequency of the mobile cloud equipment is distributed according to the obtained scheduling scheme.
As a preferred technical solution, in the step S1, the local device voltage frequency is determined by the device voltage and the device manufacturing process, and the device energy consumption is reduced by reasonably allocating the voltage frequency to the mobile cloud device.
Compared with the prior art, the invention has the following beneficial effects:
(1) the method has the advantages that the optimal voltage frequency of the equipment is corrected by adjusting the voltage frequency, the obtained voltage frequency is used as a code to carry out optimization search through a genetic algorithm, the energy consumption is reduced by 40% compared with the method without adopting the DVFS technology, and the energy consumption is reduced by 30% compared with the method only adopting the DVFS technology.
(2) The method has high reliability, seeks optimal task scheduling of the equipment in the mobile cloud environment with respect to reliability and energy consumption by utilizing a genetic algorithm, can search at a high speed to achieve the aim of multi-objective optimization, also ensures the estimation of the execution time of the task on the terminal, can improve the equipment reliability by nearly 10 percent, and is particularly suitable for terminal equipment users to process a large amount of data and calculate mass data.
Drawings
Fig. 1 is a schematic flow chart of a workflow optimization method in a mobile cloud environment according to the present invention;
FIG. 2 is a schematic flow chart of the genetic algorithm after improvement of the present invention;
FIG. 3 is a diagram of population coding modes in a genetic algorithm under a mobile cloud environment;
FIG. 4 is a graph of energy consumption versus the algorithm of the present invention and two other algorithms;
FIG. 5 is a run-time comparison graph of the algorithm of the present invention and two other algorithms;
fig. 6 is a graph of the reliability of the algorithm of the present invention and two other algorithms.
The following specific embodiments will further illustrate the invention in conjunction with the above-described figures.
Detailed Description
The technical solution provided by the present invention will be further explained with reference to the accompanying drawings.
The workflow scheduling in the mobile cloud environment is implemented in two stages, in the first stage, tasks with high transmission cost are combined according to the transmission cost among the tasks, and meanwhile, each device is allocated with proper voltage frequency, so that an initial scheduling scheme of the workflow is obtained. And the second stage is a dispatching stage, the initial dispatching scheme is adjusted by taking the requirement of meeting the deadline of the preset workflow as a constraint condition and taking the requirement of minimum energy consumption of equipment and the reliability of a user as a target, and the final workflow dispatching scheme is obtained. The method specifically comprises the following steps:
step S1: preprocessing a workflow submitted by a user;
step S2: constructing a workflow model under a mobile cloud environment;
step S3: generating an optimal workflow scheduling scheme based on an improved genetic algorithm;
step S4: distributing tasks to the equipment in the mobile cloud according to the scheduling result;
wherein the step S1 further comprises the steps of:
step S11: merging the nodes with overlarge workflow transmission cost according to the workflow transmission cost weight;
step S12: the mobile cloud environment equipment calls a voltage adjustment technology to adjust the most appropriate voltage frequency of the equipment;
in step S11, referring to fig. 1, a flowchart of a workflow optimization method in a mobile cloud environment provided by the present invention is shown, where the workflow optimization method merges data-intensive workflow similar nodes according to a workflow transmission cost threshold, and a formula of the calculation is:
in the Trans _ P [ i ] [ j ] matrix, if two task nodes do not have transmission, the two task nodes are set to be 0; otherwise, defining the weight value according to the communication cost between the nodes. Tasks with overlarge transmission cost are distributed to the same processor, so that not only can time overhead generated during transmission among the tasks be reduced, but also energy consumption generated during transmission can be reduced, the DAG graph can be converted into a new DAG graph, and the next step is to perform optimization on the basis of the new DAG graph.
In step S12, the most suitable voltage and frequency of the device is adjusted according to the performance of the mobile cloud environment device, and the energy consumption of the device can be effectively reduced by reasonably allocating the voltage and frequency to the mobile cloud device, so that the suitable voltage and frequency is crucial to the mobile cloud device, and the calculation formula of the voltage and frequency is as follows:
wherein Z is 51 × 10 when defined in the present invention under a 0.18 μm process-12；β＝1.5。
In step S2, the method further includes the steps of:
step s21, representing the task set of the workflow in the mobile cloud environment by a sideband weighted directed acyclic graph G ═ T, E, P, where T ═ T1，t2...，tnIs a set of N tasks, E { (i, j) | ti<tjIs between tasksFirst order constraints, i.e. tjCan not be at tiStarting before completion, P ═ P1,p2,...,pmcDenotes a device deployed at the mobile cloud. The optimization conditions comprise { Tim, Ene, R }, wherein Tim, Ene, R respectively represent QoS request values of time, energy consumption and reliability of the user, namely the total time T must be reached after the workflow is executedtotalLess than or equal to Tim and total cost EtotalEne and reliability RtotalNot less than R, total time T for executing initialization tasktotalTotal cost C of 0total0 and total reliability Rtotal＝0；
Step S22, it can be known from step S21 that the workflow model in the mobile cloud environment needs to define an energy consumption model, and therefore, the deadline time of the task needs to be estimated, and the following formula is calculated in combination with the failure rate of the device and the voltage frequency of the task on the device calculated from step S21:
where Ene represents the energy consumption of the device by voltage frequency regulation, PstaticRepresenting the quiescent voltage, P, of the devicedynamicWhich represents the dynamic voltage of the device,
representing the time it takes task n to run at fk on an mr device.
Step S23, it can be known from step S21 that the workflow model in the mobile cloud environment needs to define a reliability model, that is, the success rate of running the task under the device, the reliability of the device mr is affected by the instantaneous failure rate lambda and the calculation cost c of the device at the frequency fiiInfluence. The calculation formula for the reliability of the device is thus:
wherein R ismr(fi) Indicating that the device mr is on frequencyRate fiReliability of time, λ represents instantaneous failure rate of the device, d represents a constant and d is 0.1, ciIndicating the device at frequency fiThe computational cost of time.
In step S3, the method further includes the steps of:
s31, generating an optimal workflow scheduling scheme by using an improved genetic algorithm, and firstly encoding a workflow scheduling task in a mobile cloud environment;
referring to fig. 2, a flow chart of the improved genetic algorithm for solving the optimization of energy consumption and reliability in a workflow provided by the present invention is shown, which is specifically as follows:
after the energy consumption model, the reliability model and the input workflow are established, two components, namely mapping and scheduling character strings, are set for each chromosome in order to better perform algorithm optimization on a scheduling scheme of the workflow. Scheduling represents DAG topological ordering, ensuring preferential constraints. The length of the chromosome is equal to the task number set T ═ T { (T) }1，t2..，tnMapping the representative node to distribute tasks to the processor and the voltage frequency required to be adjusted; the coding pattern of individuals in a particular population is shown in FIG. 3.
After encoding the individuals, the adaptive value of each individual in the population needs to be calculated, and the formula for calculating the adaptive value is as follows:
and (4) calculating the adaptive value of each individual, wherein the larger the adaptive value is, the more excellent the individual is, so that the population is divided into an elite group and a common group in a descending order according to the fitness.
Step S32 dividing the population into elite groups and common groups;
and dividing an elite group and a common group in the population, and sorting each group of individuals in a descending order according to the fitness. New individuals are continuously generated in the process of iteration on the population, so that the situation that the member group is full of members occurs in both the normal group and the elite group, and the members of the elite group and the normal group need to be processed.
If new better individuals are generated, the new better individuals are added into the elite group, if the elite group is full, the fitness of the new better individuals is compared with the fitness of members of the elite group, inferior individuals are removed from the elite group, and if the common group is full, the worst individuals are removed from the common group.
Step S33, performing cross operation on individuals in the elite group and the common group according to the cross probability, and performing mutation operation according to the mutation probability to generate a new group;
according to the cross variation rule of the genetic algorithm, the selection reflects the approach to the optimal solution, the cross reflects the generation of the optimal solution, and the variation reflects the coverage of the global optimal solution. Therefore, the crossover operation is performed according to the probability of 2% and the mutation operation is performed according to the probability of 5% for the elite group, whereas the crossover operation is performed according to the probability of 5% and the mutation operation is performed according to the probability of 2% for the common group.
Step S34, judging whether the genetic finishing condition is satisfied, if yes, obtaining an optimal solution set, and if not, returning to execute step S33;
the genetic end condition is that after N iterations are set, the optimal solution of the individuals of the elite group can keep lower energy consumption and higher reliability under the condition of meeting time constraint.
In step S4, tasks are allocated to the devices in the mobile cloud according to the scheduling result, an optimal scheduling scheme is selected and sent to the mobile terminal, the tasks and the devices are mapped according to the scheduling scheme, and the devices perform voltage frequency adjustment according to the plan, so that the whole workflow scheduling can be completed.
Through the process, the optimization of the workflow under the cloud mobile environment is realized, the reliability of the equipment is ensured and the energy consumption of the equipment is reduced on the premise of meeting the requirement of time constraint.
In order to verify the technical effect of the technical scheme of the invention, a mobile cloud environment is simulated in a laboratory, and the workflow optimization effect of the invention is explained in detail as follows:
we set 3 servers with 3 cores in a mobile cloud environment, where the voltages of three devices at the maximum voltage frequency are respectively set to P1-1J, P2-3J, P3-5J; and randomly generating a DAG graph with 50-150 task numbers, taking the 50 task numbers as a first group of scheduling task targets, sequentially increasing by taking 10 as a unit, and totally considering 10 groups of data. Respectively taking energy consumption, time and reliability as observation targets, and sequentially comparing and explaining the common genetic algorithm optimization (MCC-GA) operated by a DAG (directed-edge graph), the genetic algorithm optimization (MCC-DVGA) combined with energy consumption and the algorithm (MCC-ODVGA) proposed in the invention, but the invention is not limited by the description.
Figures 4 to 6 show a comparison of three algorithms running on three mobile processors.
Fig. 4 represents the energy consumption of each algorithm during execution of the workflow. Of these three algorithms, the MCC-ODVGA has better performance than the other algorithms. With the increase of the number of tasks, the energy boosting effect of the MCC-ODVGA tends to be more and more obviously better than that of other algorithms.
Fig. 5 compares the completion times of the three algorithms at different numbers of tasks. Compared with MCC-GA and MCC-DVGA, MCC-ODVGA has better performance than other algorithms under the limitation of expiration date. Therefore, this method is novel, and it is effective not only in adjusting the time cost but also in improving the performance of energy consumption.
Fig. 6 shows the performance of the three algorithms on reliability. To determine the relationship between these algorithms, a third experimental search for sample reliability was conducted. Experiments show that as the complexity of tasks increases, the probability of errors also increases, the difficulty of task scheduling is also influenced by the increase of the number of tasks and the complexity of task dependence, and the MCC-ODVGA can work better when a user needs higher reliability and lower energy consumption under the deadline constraint.
The above description of the embodiments is only intended to facilitate the understanding of the method of the invention and its core idea. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (3)
1. A workflow optimization method in a mobile cloud environment is characterized by comprising the following steps:
step S1: preprocessing a workflow submitted by a user;
step S2: constructing a workflow model under a mobile cloud environment;
step S3: generating an optimal workflow scheduling scheme based on an improved genetic algorithm;
step S4: distributing tasks to the equipment in the mobile cloud according to the scheduling result;
wherein the step S1 further comprises the steps of:
step S11: merging the nodes with overlarge workflow transmission cost according to the workflow transmission cost weight;
step S12: the mobile cloud environment equipment calls a voltage adjustment technology to adjust the most appropriate voltage frequency of the equipment;
the step S11 further includes the steps of:
(1) calculating a weight tau according to the Trans _ P matrix;
(2) according to the sequence of the tasks and the transmission capability of the tasks is higher than the weight tau, the tasks are distributed to the same processor;
the calculation formula of the node merging reference weight in the workflow is as follows:
in the above formula, m represents the number of mobile devices, Trans _ P [ i [ ]][j]Representing a processor piTo pjThe data transmission capability of (2);
in step S12, the voltage frequency is calculated as follows:
wherein f represents the voltage frequency of the mobile cloud device, VddIndicating the voltage, V, supported by the systemtsRepresenting the system voltage threshold, LdRepresents the minimum time for workflow execution, and Z and β represent a constant;
the step S2 further includes the steps of:
a task set of a workflow in a mobile cloud environment is represented by a directed acyclic graph of sideband weights, wherein G is { T, E, P }, and T is { T }1，t2...，tnIs a set of N tasks, E { (i, j) | ti<tjIs the precedence constraint between tasks, i.e. tjCan not be at tiStarting before completion, P ═ P1,p2,...,pmcRepresents equipment deployed at a mobile cloud; the optimization conditions comprise { Tim, Ene and R }, wherein Tim, Ene and R respectively represent user demand values of a user on time, energy consumption and reliability;
mc and n in the model represent the number of processors used in the cluster and the number of tasks in the task set respectively, wherein each device supports a dynamic voltage frequency adjustment technology, i and j represent task numbers, so that i is greater than or equal to 1 and j is greater than or equal to n;
the step S3 further includes the steps of:
step S31: after encoding the workflow, the adaptive value of each individual of the new population needs to be calculated, and the calculation formula of the adaptive value is as follows:
where f (n, mr) denotes an adaptation value in the scheduling scheme, Rn,mrRepresenting the reliability of the task n on the device mr, RhighRepresents the maximum reliability, R, in the scheduling schemelowIndicating minimum reliability in the scheduling scheme, En,mrRepresenting the energy consumption of the task n on the device mr, EnormalIndicating the energy consumption of the device, psi, produced without the use of dynamic voltage frequency adjustment techniques1And psi2Representing two constants, and1+ψ2＝1；
step S32: dividing the population according to the following steps:
calculating the adaptive value of each individual, wherein the larger the adaptive value is, the more excellent the individual is, so that the scheduling scheme is classified into an elite group and a common group according to the descending division of the fitness;
step S33: performing crossing and mutation operations on the population, wherein the specific operations are as follows:
in the cross operation, single-node exchange is carried out on every two individual nodes in the population in sequence to obtain new individuals, and the individuals with higher adaptive values are selected as new solutions according to the adaptive values;
in the mutation operation, firstly, a set needing mutation is separated from the population according to the probability, single nodes on individuals in the set are subjected to mutation operation in sequence to obtain new individuals, and the individuals with higher adaptive values are selected as new solutions according to the adaptive values.
2. The workflow optimization method under a mobile cloud environment according to claim 1,
the step S4 further includes:
and in a preset iteration frequency range, after continuous iteration, the reliability, completion time and energy consumption in the generated new population are in accordance with the expectation of a user, and the tasks and the equipment are mapped and the proper voltage frequency of the mobile cloud equipment is distributed according to the obtained scheduling scheme.
3. The workflow optimization method in a mobile cloud environment according to claim 1 or 2,
in step S1, the local device voltage frequency is determined by the device voltage and the device manufacturing process, and the device energy consumption is reduced by reasonably allocating the local device voltage frequency to the mobile device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810095728.5A CN108416465B (en) | 2018-01-31 | 2018-01-31 | Workflow optimization method in mobile cloud environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810095728.5A CN108416465B (en) | 2018-01-31 | 2018-01-31 | Workflow optimization method in mobile cloud environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108416465A CN108416465A (en) | 2018-08-17 |
| CN108416465B true CN108416465B (en) | 2021-08-31 |
Family
ID=63127295
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810095728.5A Active CN108416465B (en) | 2018-01-31 | 2018-01-31 | Workflow optimization method in mobile cloud environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108416465B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783213B (en) * | 2018-12-28 | 2021-02-19 | 杭州电子科技大学 | Workflow fault tolerance scheduling method for reliability in edge computing environment |
| CN109992355B (en) * | 2019-01-30 | 2021-04-13 | 北京理工大学 | Multi-target cloud workflow scheduling method based on improved non-dominated genetic algorithm |
| CN110851247B (en) * | 2019-10-12 | 2023-07-25 | 华东师范大学 | Cost optimization scheduling method for cloud workflow with constraint |
| CN110928669B (en) * | 2019-12-10 | 2022-05-20 | 浙江工商大学 | Energy consumption perception cloud workflow scheduling optimization method based on multi-population genetic algorithm |
| CN113361833B (en) | 2020-03-02 | 2022-05-24 | 联芯集成电路制造(厦门)有限公司 | Chemical mechanical polishing system and related dispatching management method |
| CN112328380B (en) * | 2020-11-10 | 2024-06-25 | 武汉理工大学 | Task scheduling method and device based on heterogeneous computation |
| CN112600872B (en) * | 2020-11-20 | 2023-03-31 | 南京理工大学 | Efficient scheduling method for crowdsourcing tasks in hybrid cloud environment |
| CN114881320A (en) * | 2022-04-29 | 2022-08-09 | 哈尔滨理工大学 | Multi-objective optimization scheduling method based on virtual linear production process |
| CN116308220B (en) * | 2023-05-25 | 2023-08-15 | 北京联讯星烨科技有限公司 | Online debugging optimization method and system for workflow data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
| CN104102532A (en) * | 2013-04-15 | 2014-10-15 | 同济大学 | Low-energy-consumption-based scientific workflow scheduling method in heterogeneous cluster |
| CN105260005A (en) * | 2015-09-22 | 2016-01-20 | 浙江工商大学 | Energy consumption-oriented cloud workflow scheduling optimization method |
| CN105427063A (en) * | 2016-01-04 | 2016-03-23 | 厦门大学 | Micro-grid scheduling decision method and micro-grid scheduling decision system |
| CN107168770A (en) * | 2017-04-14 | 2017-09-15 | 中国人民解放军国防科学技术大学 | A kind of cloud data center workflow schedule of low energy consumption and resource provision method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8682686B2 (en) * | 2008-01-11 | 2014-03-25 | General Electric Company | System and method to manage a workflow in delivering healthcare |
| US9020829B2 (en) * | 2008-05-07 | 2015-04-28 | International Business Machines Corporation | Quality of service aware scheduling for composite web service workflows |
-
2018
- 2018-01-31 CN CN201810095728.5A patent/CN108416465B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102532A (en) * | 2013-04-15 | 2014-10-15 | 同济大学 | Low-energy-consumption-based scientific workflow scheduling method in heterogeneous cluster |
| CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
| CN105260005A (en) * | 2015-09-22 | 2016-01-20 | 浙江工商大学 | Energy consumption-oriented cloud workflow scheduling optimization method |
| CN105427063A (en) * | 2016-01-04 | 2016-03-23 | 厦门大学 | Micro-grid scheduling decision method and micro-grid scheduling decision system |
| CN107168770A (en) * | 2017-04-14 | 2017-09-15 | 中国人民解放军国防科学技术大学 | A kind of cloud data center workflow schedule of low energy consumption and resource provision method |
Non-Patent Citations (3)
| Title |
|---|
| A Workflow Scheduling Algorithm for Optimizing Energy-Efficient Grid Resources Usage;Fabio Coutinho等;《IEEE》;20111231;第642-649页 * |
| 云环境下基于多目标的多科学工作流调度算法;袁友伟等;《软件学报》;20171206;第3326-3339页 * |
| 移动云计算环境下任务调度的多目标优化方法;胡海洋等;《计算机研究与发展》;20170930;第54卷(第9期);第1909-1919页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108416465A (en) | 2018-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416465B (en) | Workflow optimization method in mobile cloud environment | |
| CN109818786B (en) | Method for optimally selecting distributed multi-resource combined path capable of sensing application of cloud data center | |
| CN112286677A (en) | Resource-constrained edge cloud-oriented Internet of things application optimization deployment method | |
| CN113778683B (en) | Handle identification system analysis load balancing method based on neural network | |
| CN110830562A (en) | Limited load consistency Hash load balancing strategy based on virtual nodes | |
| CN104375897A (en) | Cloud computing resource scheduling method based on minimum relative load imbalance degree | |
| CN111813506A (en) | Resource sensing calculation migration method, device and medium based on particle swarm algorithm | |
| CN110162390B (en) | Task allocation method and system for fog computing system | |
| CN113194489A (en) | Minimum-maximum cost optimization method for effective federal learning in wireless edge network | |
| CN112187535B (en) | Server deployment method and device in fog computing environment | |
| CN113794748A (en) | Performance-aware service function chain intelligent deployment method and device | |
| Gu et al. | A multi-objective fog computing task scheduling strategy based on ant colony algorithm | |
| CN112231085A (en) | Mobile terminal task migration method based on time perception in collaborative environment | |
| CN113672372B (en) | Multi-edge collaborative load balancing task scheduling method based on reinforcement learning | |
| CN111506431A (en) | Method for optimizing perception load performance of cloud server under energy consumption constraint | |
| CN114785692A (en) | Virtual power plant aggregation regulation and control communication network flow balancing method and device | |
| CN114090239A (en) | Model-based reinforcement learning edge resource scheduling method and device | |
| CN106610866A (en) | Service value constrained task scheduling algorithm in cloud storage environment | |
| CN113159539A (en) | Joint green energy scheduling and dynamic task allocation method in multilayer edge computing system | |
| CN107948330A (en) | Load balancing based on dynamic priority under a kind of cloud environment | |
| Ren et al. | Smig-rl: An evolutionary migration framework for cloud services based on deep reinforcement learning | |
| CN116431281A (en) | Virtual machine migration method based on whale optimization algorithm | |
| Guo et al. | PARA: Performability‐aware resource allocation on the edges for cloud‐native services | |
| CN113296893B (en) | Cloud platform low-resource-loss virtual machine placement method based on hybrid sine and cosine particle swarm optimization algorithm | |
| CN114980216A (en) | Dependent task unloading system and method based on mobile edge 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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240220 Address after: 230000 floor 1, building 2, phase I, e-commerce Park, Jinggang Road, Shushan Economic Development Zone, Hefei City, Anhui Province Patentee after: Dragon totem Technology (Hefei) Co.,Ltd. Country or region after: China Address before: 310018 Xiasha Higher Education Zone, Hangzhou, Zhejiang Patentee before: HANGZHOU DIANZI University Country or region before: China |