CN109800071A - A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA - Google Patents
A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA Download PDFInfo
- Publication number
- CN109800071A CN109800071A CN201910004317.5A CN201910004317A CN109800071A CN 109800071 A CN109800071 A CN 109800071A CN 201910004317 A CN201910004317 A CN 201910004317A CN 109800071 A CN109800071 A CN 109800071A
- Authority
- CN
- China
- Prior art keywords
- task
- population
- follows
- fitness
- chromosome
- 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
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 21
- 238000004364 calculation method Methods 0.000 title claims abstract description 15
- 210000000349 chromosome Anatomy 0.000 claims abstract description 59
- 238000000034 method Methods 0.000 claims abstract description 37
- 230000035772 mutation Effects 0.000 claims abstract description 23
- 238000011156 evaluation Methods 0.000 claims abstract description 16
- 238000005457 optimization Methods 0.000 claims abstract description 10
- 108090000623 proteins and genes Proteins 0.000 claims description 17
- 238000005265 energy consumption Methods 0.000 claims description 13
- 230000006978 adaptation Effects 0.000 claims description 7
- 238000003066 decision tree Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000004043 dyeing Methods 0.000 claims 1
- 230000002068 genetic effect Effects 0.000 abstract description 10
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000009399 inbreeding Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 239000000975 dye Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000011664 nicotinic acid Substances 0.000 description 1
- 235000001968 nicotinic acid Nutrition 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA, using the evaluation method of the weighted optimization based on multiple target as fitness function;The mode of real number direct coding is taken to be encoded;Generate using state algorithm and at random the method generation initial population combined;Classification selection strategy is taken to carry out selection operation;Crossover operation is carried out according to crossover probability and chromosome diversity factor;Mutation operation is carried out according to dynamic variation probability;Whether terminate according to double termination condition determining programs;It finds preferred plan and is allocated.Improvement is optimized to multiple steps of traditional genetic algorithm in the present invention, improves user satisfaction and algorithm execution efficiency.
Description
Technical field
The present invention relates to cloud computing task scheduling technique field more particularly to a kind of cloud computings based on improved adaptive GA-IAGA
Method for scheduling task.
Background technique
With the fast development of information age, computer technology and network technology etc. are more and more perfect, have pushed internet
Application is popularized.Cloud computing be grow up on the basis of distributed computing, parallel computation, grid computing one kind it is novel
Calculating mode, cloud computing obtain various resources, including various applications and IT service etc. by network.As a kind of new calculating mould
Formula, the core of cloud computing are also to be connected computing resource, Service Source and storage resource etc. by network, form a resource
Pond, while according to the demand of user, United Dispatching and the relevant resource of management.
In face of increasing service request, realize that the scheduling efficient and rational to task in cloud helps to be promoted user to clothes
The satisfaction of business is one of the emphasis paid close attention in cloud computing application research.The cloud computing technology computing discipline emerging as one,
It is the stage developed using relevant many researchs still in one to it, not perfect.What many task schedules were still continued to use is net
Method in lattice calculating, though it is simple and practical, due to the considerations of lacking to factors such as the environmental characteristics of cloud computing, scheduling effect
Fruit is often unsatisfactory.Therefore reasonable scheduling model how is taken out, and realizes cloud computing task using preferable algorithm
Effective scheduling, meet the needs of users, have great significance to the practical cloud computing service that improves to a greater degree.
Currently, with bionic rapid development, people increasingly pay attention to bionics techniques, and thus produce new calculation
Method, genetic algorithm are one of them.Genetic algorithm is classical intelligent algorithm, it can preferably solve np problem, people
Be gradually concerned about the concurrency and intelligence of genetic algorithm, and be introduced into during the scheduling of resource of cloud computing, effect compares
It is good.But traditional genetic algorithm, when handling cloud computing task schedule, inevitably some defects, limit it
Effect.Therefore how improved adaptive GA-IAGA, the shortcomings that overcoming traditional algorithm, its reliability of lift-off and efficiency, by genetic algorithm
It is dissolved into the task schedule field of cloud computing, is met the needs of users, the effect of genetic algorithm is more effectively played, becomes mesh
Preceding research emphasis.
Summary of the invention
It is an object of the invention to provide a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA, including following step
It is rapid:
S0. it pre-processes: proposing particular problem description or model, determine chromosome coding mode, determine that chromosome is preferably commented
Valence method and fitness function;
S1. it initializes: setting maximum number of iterations or other termination conditions, population scale, crossover probability, mutation probability etc.
Parameter;
S2. it generates initial population: state algorithm and random algorithm is combined, form first generation population, and iteration meter
Number device zero;
S3. it calculates initial population fitness: calculating the fitness value for going out all individuals in contemporary population, and it is optimal to find value
Fitness value, as current contemporary optimal solution and current globally optimal solution;
S4. judge whether to terminate: being judged whether to meet two termination conditions according to the information that previous step obtains, if meeting
Output global optimum's individual is optimal solution, and algorithm terminates;Otherwise enter in next step;
S5. selection operation: taking classification selection strategy, and advantage individual is avoided to be lost during evolving by generation;
S6. the individual of current population crossover operation: is selected according to crossover probability to execute improved crossover operation;
S7. improved mutation operation mutation operation: is executed according to the individual of the current population of dynamic variation probability selection;
S8. it generates next-generation population: passing through step S5, S6, S7, generate the population of a new generation;
S9. it calculates current population's fitness: first calculating all ideal adaptation angle value of contemporary population, and find out wherein most
That big label is optimum individual.It is compared to each other with global optimum individual again, if its value is more excellent, by it
As global optimum's individual;Otherwise it does not change;
S10. algorithm terminates: judge whether two conditions for meeting iteration ends, if it is not, iteration count adds 1 certainly, algorithm
Return back to S5;If so, output is as a result, algorithm terminates;
S11. it exports optimal solution: according to optimal solution, task being assigned to virtual machine, and export optimal solution;
The particular problem of the step S0 describes or model is as follows:
The different cloud task (Tasks) of M length is distributed to N number of performance different virtual machine (VMs) to execute, it is reasonable to find
Allocation plan so that allocation result meet the cloud computing task schedule decision tree based on multiple target, problem include it is following about
Beam condition is using MapReduce model, and overall tasks have been split into several small tasks, and nothing between each subtask
It is directly linked;Two are mapped to virtual machine for all cloud computing resources;Three be the delay and loss for ignoring data transmission;Four be to mention
The number of tasks of friendship is greater than virtual machine number;
The coding mode about chromosome of the step S0 is as follows:
Take the mode of real number direct coding: every chromosome length (i.e. contained gene dosage) is equal to task quantity M, base
The value of cause is the serial number { 0,1,2 ..., N-1 } of virtual machine, therefore each gene position means that task symbol, gene
Value, which means that distribute to, is worth virtual machine marked as this.It clearer can show the mapping relations between Task and VM in this way.
Of the invention is further described, chromosome optimizing evaluation method in the step S0 are as follows: adding based on multiple target
Optimizing evaluation method, the method overall merit task completion time are weighed, the factor of three aspect of amount consumption and load balancing uses
Person can determine corresponding weight according to demand, obtain corresponding fitness function, specific as follows:
(1) it about the fitness function component F 1 of task completion time, is obtained by following procedure:
Execution time TaskTime (i, j) of the note task i on virtual machine j are as follows:
Wherein Task_length (i) indicates the length of task i, and VM_Mips (j) indicates the execution speed of virtual machine j.
After task is distributed, each VM executes task execution time the sum of of the time for all distribution on the VM, it may be assumed that
Wherein Num (j) indicates the task quantity distributed on j-th of VM.
The then task completion time of certain allocation plan are as follows:
Thus, the fitness function component F 1 about task completion time are as follows:
F1=Fitnesstime=Timetotal;
(2) it about the fitness function component F 2 of energy consumption, is obtained by following procedure:
The specific consumption for remembering VM (j) is Power (j), then the energy input of VM (j) are as follows:
The total energy consumption of all virtual machines of certain allocation plan are as follows:
Thus, the fitness function component F 2 about energy consumption are as follows:
F2=Fitnesspower=Powertotal;
(3) it about the fitness function component F 3 of load balancing, is obtained by following procedure:
The execution time average of all VMs are as follows:
Load balancing function are as follows:
Fitness function about target 3 (load balancing) at this time are as follows:
F3=Fitnessload=LoadBalance;
(4) the weighted optimization evaluation method based on multiple target, comprehensive fitness degree function are as follows:
F=α F1+βF2+γF3;
Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
For the step S1 in initialization, setting maximum number of iterations is MaxGen, and other termination conditions are global optimum
Chromosome is in continuous X generation without more excellent solution, population scale Scale, crossover probability Pc, mutation probability Pm。
Of the invention is further described, the step S2 is when generating initialization population, first with a variety of static methods
Choosing generates chromosome dyad, then generates other chromosomes at random according to coding rule, thus forms a population.State algorithm
The characteristics of be execute speed be exceedingly fast, solution can be quickly found, these solution typically more excellent solution.This operates introducing that can be purposive
The more excellent solution in part, increases the probability of advantage chromosome, can provide initial population more better than random selection initial population.Example
Such as: using state algorithm, such as max_min, min_min scheduling algorithm first acquires chromosome dyad, then generates other dyes at random
Thus colour solid forms initial population.
Of the invention is further described, the step S3 can be obtained when calculating fitness according to following formula:
F=α F1+βF2+γF3;(formula 12);Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
Specific weight α, beta, gamma are selected according to actual needs by user.Such as when only considering the deadline, it can use: α
=1, β=0, γ=0;If it is considered that deadline and energy consumption are more important, load balance is more secondary, then can use: α=0.4,
β=0.4, γ=0.2, and so on.
Of the invention is further described, the step S4 is when judging termination condition, using two kinds of termination conditions, respectively
It is as follows:
If global optimum's chromosome regard this solution as optimal solution in the continuous X generation more excellent solution of nothing, terminator;
If two do not occur situation (1), maximum number of iterations MaxGen is gone to always, and export globally optimal solution,
When judging termination condition, (1) is first determined whether, then judge (2).Such termination condition is added, can be reduced not
Necessary the number of iterations improves executing efficiency.
As an improvement of the present invention, the step S5 takes classification back-and-forth method, specifically such as when carrying out selection operation
Under:
Find out the mean value of all chromosome fitness in populationFitness value is less than in the population
(being better than) average value, that is, meetChromosome it is selected, remaining chromosome is according to individual adaptation degree in current population
The details of value operate to execute roulette selection.The select probability of each chromosome when executing roulette selection operation are as follows:Wherein Scale is population scale, and the method effectively prevents excellent individual by generation evolution
It is lost in the process;It can accelerate convergence rate.
Of the invention is further described, when the step S6 carries out crossover operation, takes two neighboring position one in population
It is right, the mode of random single point crossing, with probability PcIntersected.The similarity of each two chromosome of centering is checked before intersecting
(same position gene equal probabilities), do not intersect when similarity is larger (more than 60%), otherwise normal to intersect.It can avoid in this way
" inbreeding " generates more different individuals, it is ensured that the diversification of group.
Of the invention is further described, when the step S7 carries out mutation operation, mutation probability is function Pm=0.15-
0.1e-(i/1000), wherein i is the number of iterations.It can guarantee that mutation probability is not too big in this way, and can be adaptive.
Of the invention is further described, step S11 assigns the task to corresponding void according to the optimal solution for inputting algorithm
Quasi- machine, terminates program.
By adopting the above technical scheme, it has the following beneficial effects:
The present invention has comprehensively considered the plurality of target demand of user, makes user that can adjust objective appraisal according to the demand of oneself
Function (fitness function), improves user satisfaction;And the Optimal improvements by being carried out to the multiple steps of traditional genetic algorithm,
Improve algorithm execution efficiency;Use the weighted optimization evaluation method based on multiple target as the fitness function of genetic algorithm,
It is capable of the effect of the multiple targets of overall merit, and selects optimal scheduling scheme accordingly;To several processes of traditional genetic algorithm: raw
Optimal improvements appropriate have been carried out at initial population, selection, intersection, variation, termination condition etc., have improved efficiency of algorithm.
Detailed description of the invention
Fig. 1 is flowage structure schematic diagram of the invention;
Specific embodiment
The present invention will be further described with reference to the accompanying drawing.
A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA, steps are as follows:
S0. it pre-processes: proposing particular problem description or model, determine chromosome coding mode, determine that chromosome is preferably commented
Valence method and fitness function;
Problem description are as follows: the different cloud task (Tasks) of M length is distributed into N number of performance different virtual machine (VMs) and is held
Row, finds reasonable allocation plan, so that allocation result meets the cloud computing task schedule decision tree based on multiple target.
The chromosome coding mode takes the mode of real number direct coding, and chromosome length is number of tasks M, each position
On genic value indicate the corresponding resource number of the task.
Fitness function is the weighted optimization evaluation function based on multiple target, it may be assumed that
F=α F1+βF2+γF3;Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
Specific weight α, beta, gamma are selected according to actual needs by user, such as herein desirable α=0.4, and β=0.4, γ=
0.2, it indicates in evaluation of result, deadline and energy consumption are more important, and load balancing is more secondary, and so on.
S1. initialize: setting maximum number of iterations is MaxGen, and other termination conditions are global optimum's chromosome continuous
In X generation, is without more excellent solution (such as settable X=MaxGen × 10%), population scale Scale, crossover probability Pc, mutation probability
Etc. parameters be Pm。
S2. generate initial population: first with a variety of static methods select generate chromosome dyad, then according to coding rule with
Machine generates other chromosomes, thus forms a population, and iteration count is zeroed.
The characteristics of state algorithm is to execute speed to be exceedingly fast, and can quickly find solution, these solutions are typically all more excellent solution.This operation
The more excellent solution of introducing portion that can be purposive increases the probability of advantage chromosome, can provide than random selection initial population more
Good initial population.Such as: using state algorithm, such as max_min, min_min scheduling algorithm first acquires chromosome dyad, then
Other chromosomes are generated at random, thus form initial population.
S3. it calculates initial population fitness: calculating the fitness value for going out all individuals in contemporary population, and it is optimal to find value
Fitness value, as current contemporary optimal solution and current globally optimal solution;Fitness function formula is by base in step S0
It is provided in the weighted optimization evaluation function of multiple target.
S4. judge whether to terminate: being judged whether to meet two termination conditions according to the information that previous step obtains, if meeting
Output global optimum's individual is optimal solution, and algorithm terminates;Otherwise enter in next step.
Two kinds of termination conditions are as follows respectively:
(1) if global optimum's chromosome is in the continuous X generation more excellent solution of nothing, terminator, and it regard this solution as optimal solution
(2) if not occurring situation (1), maximum number of iterations MaxGen is gone to always, and export globally optimal solution.
When judging termination condition, (1) is first determined whether, then judge (2).
S5. selection operation: taking classification selection strategy, and advantage individual is avoided to be lost during evolving by generation.
Classification selection method particularly includes: find out the mean value of all chromosome fitness in populationIt should
Fitness value is less than (being better than) average value in population, that is, meetsChromosome it is selected, remaining chromosome is according to working as
The details of ideal adaptation angle value operate in preceding population to execute roulette selection.Each chromosome when executing roulette selection operation
Select probability are as follows:
S6. the individual of current population crossover operation: is selected according to crossover probability to execute improved crossover operation;
Crossover probability and method are as follows: take two neighboring position in population a pair of, the mode of random single point crossing, with probability P c
Intersected.Check the similarity (same position gene equal probabilities) of each two chromosome of centering before intersecting, similarity compared with
Do not intersect when big (more than 60%), it is otherwise normal to intersect.It can avoid " inbreeding " in this way, generate more different individuals,
Ensure the diversification of group.
S7. improved mutation operation mutation operation: is executed according to the individual of the current population of dynamic variation probability selection.Become
Different probability is function Pm=0.15-0.1e-(i/1000), wherein i is the number of iterations.It can guarantee that mutation probability is not too big in this way, and
It can be adaptive.
S8. it generates next-generation population: passing through step S5, S6, S7, generate the population of a new generation.
S9. it calculates current population's fitness: first calculating all ideal adaptation angle value of contemporary population, and find out wherein most
That big label is optimum individual.It is compared to each other with global optimum individual again, if its value is more excellent, by it
As global optimum's individual;Otherwise it does not change;Fitness function formula is by the weighted optimization based on multiple target in step S0
Evaluation function provides.
S10. algorithm terminates: judge whether two conditions for meeting iteration ends, if it is not, iteration count adds 1 certainly, algorithm
Return back to S5;If so, output is as a result, algorithm terminates.Two conditions are provided by step S4.
S11. it exports optimal solution: according to optimal solution, task being assigned to virtual machine, and export optimal solution.
The particular problem of the step S0 describes or model is as follows:
The different cloud task (Tasks) of M length is distributed to N number of performance different virtual machine (VMs) to execute, it is reasonable to find
Allocation plan so that allocation result meet the cloud computing task schedule decision tree based on multiple target, problem include it is following about
Beam condition is using MapReduce model, and overall tasks have been split into several small tasks, and nothing between each subtask
It is directly linked;Two are mapped to virtual machine for all cloud computing resources;Three be the delay and loss for ignoring data transmission;Four be to mention
The number of tasks of friendship is greater than virtual machine number;
The coding mode about chromosome of the step S0 is as follows:
Take the mode of real number direct coding: every chromosome length (i.e. contained gene dosage) is equal to task quantity M, base
The value of cause is the serial number { 0,1,2 ..., N-1 } of virtual machine, therefore each gene position means that task symbol, gene
Value, which means that distribute to, is worth virtual machine marked as this.It clearer can show the mapping relations between Task and VM in this way.
Take the mode of real number direct coding: every chromosome length (i.e. contained gene dosage) is equal to task quantity M, base
The value of cause is the serial number { 0,1,2 ..., N-1 } of virtual machine, therefore each gene position means that task symbol, gene
Value, which means that distribute to, is worth virtual machine marked as this.It clearer can show the mapping relations between Task and VM in this way.
For example: it is assumed that virtual machine quantity is 5, number of tasks is 10, then the gene number of its every chromosome is
10, genic value be serial number { 0,1,2,3,4 } corresponding to 5 virtual machines one of them.If be encoded to (2,1,3,0,4,1,3,
2,4,2) chromosome, then its indicate virtual machine and task corresponding relationship are as follows:
VM0:Task { 3 }
VM1:Task { 1,5 }
VM2:Task { 0,7,9 }
VM3:Task { 2,6 }
VM4:Task { 4,8 }
Chromosome optimizing evaluation method in the step S0 are as follows: the weighted optimization evaluation method based on multiple target, the method
Overall merit task completion time, the factor of three aspect of amount consumption and load balancing, user can determine phase according to demand
The weight answered obtains corresponding fitness function, specific as follows:
(1) it about the fitness function component F 1 of task completion time, is obtained by following procedure:
Execution time TaskTime (i, j) of the note task i on virtual machine j are as follows:
Wherein Task_length (i) indicates the length of task i, and VM_Mips (j) indicates the execution speed of virtual machine j.
After task is distributed, each VM executes task execution time the sum of of the time for all distribution on the VM, it may be assumed that
Wherein Num (j) indicates the task quantity distributed on j-th of VM,
The then task completion time of certain allocation plan are as follows:
Thus, the fitness function component F 1 about task completion time are as follows:
F1=Fitnesstime=Timetotal;
(2) it about the fitness function component F 2 of energy consumption, is obtained by following procedure:
The specific consumption for remembering VM (j) is Power (j), then the energy input of VM (j) are as follows:
The total energy consumption of all virtual machines of certain allocation plan are as follows:
Thus, the fitness function component F 2 about energy consumption are as follows:
F2=Fitnesspower=Powertotal;
(3) it about the fitness function component F 3 of load balancing, is obtained by following procedure:
The execution time average of all VMs are as follows:
Load balancing function are as follows:
Fitness function about target 3 (load balancing) at this time are as follows:
F3=Fitnessload=LoadBalance;
(4) the weighted optimization evaluation method based on multiple target, comprehensive fitness degree function are as follows:
F=α F1+βF2+γF3;
Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
For the step S1 in initialization, setting maximum number of iterations is MaxGen, and other termination conditions are global optimum
Chromosome is in continuous X generation without more excellent solution, population scale Scale, crossover probability Pc, mutation probability Pm。
The step S2 is selected first with a variety of static methods when generating initialization population and is generated chromosome dyad, then
It generates other chromosomes at random according to coding rule, thus forms a population.The characteristics of state algorithm is to execute speed to be exceedingly fast,
Solution can be quickly found, these solutions are typically all more excellent solution.This operates the more excellent solution of introducing portion that can be purposive, increases advantage dye
The probability of colour solid can provide initial population more better than random selection initial population.Such as: using state algorithm, such as
Max_min, min_min scheduling algorithm first acquire chromosome dyad, then generate other chromosomes at random, thus form initial kind
Group.
Of the invention is further described, the step S3 can be obtained when calculating fitness according to following formula:
F=α F1+βF2+γF3;(formula 12);Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
Specific weight α, beta, gamma are selected according to actual needs by user.Such as when only considering the deadline, it can use: α
=1, β=0, γ=0;If it is considered that deadline and energy consumption are more important, load balance is more secondary, then can use: α=0.4,
β=0.4, γ=0.2, and so on.
The step S4 is as follows respectively using two kinds of termination conditions when judging termination condition:
If global optimum's chromosome regard this solution as optimal solution in the continuous X generation more excellent solution of nothing, terminator;
If two do not occur situation (1), maximum number of iterations MaxGen is gone to always, and export globally optimal solution,
When judging termination condition, (1) is first determined whether, then judge (2).Such termination condition is added, can be reduced not
Necessary the number of iterations improves executing efficiency.
The step S5 takes classification back-and-forth method when carrying out selection operation, specific as follows:
Find out the mean value of all chromosome fitness in populationFitness value is less than in the population
(being better than) average value, that is, meetChromosome it is selected, remaining chromosome is according to individual adaptation degree in current population
The details of value operate to execute roulette selection.The select probability of each chromosome when executing roulette selection operation are as follows:Wherein Scale is population scale, and the method effectively prevents excellent individual by generation evolution
It is lost in the process, convergence rate can be accelerated.
When the step S6 carries out crossover operation, take two neighboring position in population a pair of, the mode of random single point crossing,
With probability PcIntersected.The similarity (same position gene equal probabilities) of each two chromosome of centering is checked before intersecting,
Do not intersect when similarity is larger (more than 60%), it is otherwise normal to intersect.It can avoid " inbreeding " in this way, generate more different
Individual, it is ensured that the diversification of group.
When the step S7 carries out mutation operation, mutation probability is function Pm=0.15-0.1e-(i/1000), wherein i is repeatedly
Generation number.It can guarantee that mutation probability is not too big in this way, and can be adaptive.
Step S11 assigns the task to respective virtual machine according to the optimal solution for inputting algorithm, terminates program.
The foregoing describe basic principles and main features of the invention, It should be understood by those skilled in the art that of the invention
It is not restricted to the described embodiments, the above embodiments and description only illustrate the principle of the present invention, is not departing from
Under the premise of spirit and scope of the invention, various changes and improvements may be made to the invention, these changes and improvements both fall within requirement
In the scope of the invention of protection, invents claimed range and be defined by the appending claims and its equivalent thereof.
Claims (9)
1. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA, which comprises the following steps:
S0. it pre-processes: proposing particular problem description or model, determine chromosome coding mode, determine chromosome optimizing evaluation side
Method and fitness function;
S1. initialize: setting maximum number of iterations or other termination conditions, population scale, crossover probability, mutation probability etc. are joined
Number;
S2. it generates initial population: state algorithm and random algorithm is combined, form first generation population, and iteration count
Zero;
S3. it calculates initial population fitness: calculating in contemporary population and go out the fitness value of all individuals, and find value adaptive optimal control
Angle value, as current contemporary optimal solution and current globally optimal solution;
S4. judge whether to terminate: being judged whether to meet two termination conditions according to the information that previous step obtains, be exported if meeting
Global optimum's individual is optimal solution, and algorithm terminates;Otherwise enter in next step;
S5. selection operation: taking classification selection strategy, and advantage individual is avoided to be lost during evolving by generation;
S6. the individual of current population crossover operation: is selected according to crossover probability to execute improved crossover operation;
S7. improved mutation operation mutation operation: is executed according to the individual of the current population of dynamic variation probability selection;
S8. it generates next-generation population: passing through step S5, S6, S7, generate the population of a new generation;
S9. it calculates current population's fitness: first calculating all ideal adaptation angle value of contemporary population, and find out maximum
That label is optimum individual, then it is compared to each other with global optimum individual, if its value is more excellent, as
Global optimum's individual;Otherwise it does not change;
S10. algorithm terminates: judging whether two conditions for meeting iteration ends, if it is not, iteration count is from adding 1, algorithm is turned round
To S5;If so, output is as a result, algorithm terminates;
S11. it exports optimal solution: according to optimal solution, task being assigned to virtual machine, and export optimal solution;
The particular problem of the step S0 describes or model is as follows:
The different cloud task (Tasks) of M length is distributed to N number of performance different virtual machine (VMs) to execute, finds reasonable point
With scheme, so that allocation result meets the cloud computing task schedule decision tree based on multiple target, problem includes following constraint item
Part is using MapReduce model, and overall tasks have been split into several small tasks, and without directly between each subtask
Association;Two are mapped to virtual machine for all cloud computing resources;Three be the delay and loss for ignoring data transmission;Four be to submit
Number of tasks is greater than virtual machine number;
The coding mode about chromosome of the step S0 is as follows:
Take the mode of real number direct coding: every chromosome length (i.e. contained gene dosage) is equal to task quantity M, gene
Value is the serial number { 0,1,2 ..., N-1 } of virtual machine, therefore each gene position means that task symbol, and genic value is just
Expression, which is distributed to, is worth virtual machine marked as this.It clearer can show the mapping relations between Task and VM in this way.
2. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
Chromosome optimizing evaluation method in the step S0 are as follows: the weighted optimization evaluation method based on multiple target, the method overall merit
Task completion time, the factor of three aspect of amount consumption and load balancing, user can determine corresponding weight according to demand,
Corresponding fitness function is obtained, specific as follows:
(1) it about the fitness function component F 1 of task completion time, is obtained by following procedure:
Execution time TaskTime (i, j) of the note task i on virtual machine j are as follows:
Wherein Task_length (i) indicates the length of task i, and VM_Mips (j) indicates the execution speed of virtual machine j.
After task is distributed, each VM executes task execution time the sum of of the time for all distribution on the VM, it may be assumed that
Wherein Num (j) indicates the task quantity distributed on j-th of VM.
The then task completion time of certain allocation plan are as follows:
Thus, the fitness function component F 1 about task completion time are as follows:
F1=Fitnesstime=Timetotal;
(2) it about the fitness function component F 2 of energy consumption, is obtained by following procedure:
The specific consumption for remembering VM (j) is Power (j), then the energy input of VM (j) are as follows:
The total energy consumption of all virtual machines of certain allocation plan are as follows:
Thus, the fitness function component F 2 about energy consumption are as follows:
F2=Fitnesspower=Powertotal;
(3) it about the fitness function component F 3 of load balancing, is obtained by following procedure:
The execution time average of all VMs are as follows:
Load balancing function are as follows:
Fitness function about target 3 (load balancing) at this time are as follows:
F3=Fitnessload=LoadBalance;
(4) the weighted optimization evaluation method based on multiple target, comprehensive fitness degree function are as follows:
F=α F1+βF2+γF3;
Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1]
For the step S1 in initialization, setting maximum number of iterations is MaxGen, and other termination conditions are global optimum's dyeing
Body is in continuous X generation without more excellent solution, population scale Scale, crossover probability Pc, mutation probability Pm。
3. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
The step S2 is selected first with a variety of static methods when generating initialization population and is generated chromosome dyad, then according to coding
Regular random generates other chromosomes, thus forms a population.
4. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
The step S3 can be obtained when calculating fitness according to following formula:
F=α F1+βF2+γF3;(formula 12);Wherein alpha+beta+γ=1, and α, beta, gamma ∈ [0,1];
Specific weight α, beta, gamma are selected according to actual needs by user.
α=1, β=0, γ=0;If it is considered that deadline and energy consumption are more important, load balance is more secondary, then can use: α
=0.4, β=0.4, γ=0.2, and so on.
5. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
The step S4 is as follows respectively using two kinds of termination conditions when judging termination condition:
If global optimum's chromosome regard this solution as optimal solution in the continuous X generation more excellent solution of nothing, terminator;
If two do not occur situation (1), maximum number of iterations MaxGen is gone to always, and export globally optimal solution;
When judging termination condition, (1) is first determined whether, then judge (2).Such termination condition is added, it is unnecessary to reduce
The number of iterations, improve executing efficiency.
6. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
The step S5 takes classification back-and-forth method when carrying out selection operation, specific as follows:
Find out the mean value of all chromosome fitness in populationFitness value is less than (excellent in the population
In) average value, that is, meetChromosome it is selected, remaining chromosome is according to ideal adaptation angle value in current population
Details come execute roulette selection operation.The select probability of each chromosome when executing roulette selection operation are as follows:Wherein Scale is population scale.
7. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
When the step S6 carries out crossover operation, take two neighboring position in population a pair of, the mode of random single point crossing, with probability Pc
Intersected;Check the similarity (same position gene equal probabilities) of each two chromosome of centering before intersecting, similarity compared with
Do not intersect when big (more than 60%), it is otherwise normal to intersect.
8. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
When the step S7 carries out mutation operation, mutation probability is function Pm=0.15-0.1e-(i/1000), wherein i is the number of iterations, can
Guarantee that mutation probability is not too big, and can be adaptive.
9. a kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA according to claim 1, which is characterized in that
Step S11 assigns the task to respective virtual machine according to the optimal solution for inputting algorithm, terminates program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910004317.5A CN109800071A (en) | 2019-01-03 | 2019-01-03 | A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910004317.5A CN109800071A (en) | 2019-01-03 | 2019-01-03 | A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109800071A true CN109800071A (en) | 2019-05-24 |
Family
ID=66558451
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910004317.5A Pending CN109800071A (en) | 2019-01-03 | 2019-01-03 | A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109800071A (en) |
Cited By (47)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110389819A (en) * | 2019-06-24 | 2019-10-29 | 华中科技大学 | A kind of dispatching method and system of computation-intensive batch processing task |
CN110458291A (en) * | 2019-08-09 | 2019-11-15 | 成都理工大学 | A kind of best common recognition node selecting method based on genetic algorithm |
CN110569117A (en) * | 2019-08-26 | 2019-12-13 | 南瑞集团有限公司 | Power supply station intelligent cloud platform task scheduling algorithm and system based on optimized particle swarm |
CN110609739A (en) * | 2019-09-19 | 2019-12-24 | 湘潭大学 | Train period information scheduling method and system based on multi-objective evolutionary algorithm |
CN110751292A (en) * | 2019-10-12 | 2020-02-04 | 浙江财经大学 | Cloud manufacturing multi-view collaborative scheduling optimization method based on genetic algorithm |
CN110837413A (en) * | 2019-11-20 | 2020-02-25 | 福州大学 | Computing migration scheduling method for deep neural network application in edge environment |
CN110851277A (en) * | 2019-11-08 | 2020-02-28 | 中国石油大学(华东) | Task scheduling strategy based on edge cloud cooperation in augmented reality scene |
CN110851272A (en) * | 2019-10-30 | 2020-02-28 | 内蒙古农业大学 | Cloud task scheduling method based on phagocytic particle swarm genetic hybrid algorithm |
CN110851257A (en) * | 2019-11-07 | 2020-02-28 | 中国石油大学(华东) | Genetic and differential hybrid evolution cloud computing task scheduling algorithm based on early-stage catastrophe strategy |
CN110889552A (en) * | 2019-11-26 | 2020-03-17 | 中国计量大学 | Automatic apple boxing path optimization method based on optimal parameter genetic algorithm |
CN110928669A (en) * | 2019-12-10 | 2020-03-27 | 浙江工商大学 | Energy consumption perception cloud workflow scheduling optimization method based on multi-population genetic algorithm |
CN111079229A (en) * | 2019-12-20 | 2020-04-28 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment unbalance dual-target optimization method based on cloud self-adaptive genetic algorithm |
CN111125904A (en) * | 2019-12-20 | 2020-05-08 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment blade sorting method based on multi-target regulation |
CN111274030A (en) * | 2020-01-16 | 2020-06-12 | 中国人民解放军国防科技大学 | Efficient multiprocessor system-on-chip design space mining method oriented to application features |
CN111290853A (en) * | 2020-01-14 | 2020-06-16 | 西安建筑科技大学 | Cloud data center scheduling method based on self-adaptive improved genetic algorithm |
CN111338749A (en) * | 2020-02-10 | 2020-06-26 | 长沙学院 | Task scheduling method, system and storage medium |
CN111427688A (en) * | 2020-03-23 | 2020-07-17 | 武汉轻工大学 | Cloud task multi-target scheduling method and device, electronic equipment and storage medium |
CN111641891A (en) * | 2020-04-16 | 2020-09-08 | 北京邮电大学 | Task peer-to-peer unloading method and device in multi-access edge computing system |
CN111651434A (en) * | 2019-10-22 | 2020-09-11 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN111985841A (en) * | 2020-08-31 | 2020-11-24 | 华中科技大学 | Injection molding workshop scheduling method and system based on improved genetic algorithm |
CN112035224A (en) * | 2020-07-17 | 2020-12-04 | 中国科学院上海微系统与信息技术研究所 | Fog calculation scheduling method suitable for intelligent factory |
CN112069081A (en) * | 2020-09-24 | 2020-12-11 | 中国人民解放军国防科技大学 | Genetic algorithm-based directory controller test excitation generation method |
CN112083967A (en) * | 2020-08-18 | 2020-12-15 | 深圳供电局有限公司 | Unloading method of cloud edge computing task, computer equipment and storage medium |
CN112181598A (en) * | 2020-10-12 | 2021-01-05 | 内蒙古农业大学 | Cloud computing task scheduling method based on improved genetic algorithm |
CN112364528A (en) * | 2020-12-09 | 2021-02-12 | 南京长峰航天电子科技有限公司 | Microwave darkroom multipath effect evaluation algorithm based on genetic algorithm |
CN112379996A (en) * | 2020-10-14 | 2021-02-19 | 华南理工大学 | Cloud computing resource scheduling optimization method based on dynamic target strategy and application |
CN112598333A (en) * | 2021-01-07 | 2021-04-02 | 北京城市大数据研究院有限公司 | Community empty nest old people assisting and dispatching method based on genetic algorithm |
CN112631612A (en) * | 2020-12-03 | 2021-04-09 | 杭州谐云科技有限公司 | Optimization method for kubernetes cloud platform configuration based on genetic algorithm |
CN112819242A (en) * | 2021-02-22 | 2021-05-18 | 西北工业大学 | Civil transportation type airplane flight test task allocation optimization method |
CN113127167A (en) * | 2021-03-18 | 2021-07-16 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
CN113132445A (en) * | 2020-01-10 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
CN113268376A (en) * | 2021-04-16 | 2021-08-17 | 济南轨道交通集团有限公司 | Data center object storage method and system based on genetic algorithm |
CN113326917A (en) * | 2021-04-29 | 2021-08-31 | 开放智能机器(上海)有限公司 | Method and system for automatically optimizing operator based on genetic algorithm |
CN113411369A (en) * | 2020-03-26 | 2021-09-17 | 山东管理学院 | Cloud service resource collaborative optimization scheduling method, system, medium and equipment |
CN113656073A (en) * | 2021-08-18 | 2021-11-16 | 哈尔滨工业大学 | Parallel scheduling mode recombination method based on large mode set, computer and storage medium |
CN113848725A (en) * | 2021-10-13 | 2021-12-28 | 长春工业大学 | Mechanical arm time optimal trajectory planning and optimizing method based on adaptive genetic algorithm |
CN113986518A (en) * | 2021-12-28 | 2022-01-28 | 华南理工大学 | Real-time cloud task multi-objective optimization scheduling method based on ant colony algorithm |
CN114581058A (en) * | 2022-03-10 | 2022-06-03 | 杭州电子科技大学 | Personnel organization structure optimization method based on business process |
CN114943489A (en) * | 2022-07-22 | 2022-08-26 | 浙江中控技术股份有限公司 | Production scheduling method and device based on improved genetic algorithm |
CN115185655A (en) * | 2022-06-23 | 2022-10-14 | 郑州轻工业大学 | Genetic task scheduling method based on gene frequency improvement |
CN115293449A (en) * | 2022-08-23 | 2022-11-04 | 中国空间技术研究院 | Distribution optimization method for tasks with time windows and coupling constraints |
CN116401037A (en) * | 2023-06-05 | 2023-07-07 | 杭州谐云科技有限公司 | Genetic algorithm-based multi-task scheduling method and system |
CN116562582A (en) * | 2023-05-23 | 2023-08-08 | 广东建设职业技术学院 | Intelligent distribution method for building engineering detection tasks |
CN117057485A (en) * | 2023-10-11 | 2023-11-14 | 山东天鼎舟工业科技有限公司 | Scheduling method and system for casting machining production |
CN117130622A (en) * | 2023-10-26 | 2023-11-28 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Distributed online code compiling and running method and system |
CN117272838A (en) * | 2023-11-17 | 2023-12-22 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
CN117290077A (en) * | 2023-11-24 | 2023-12-26 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130308570A1 (en) * | 2012-05-17 | 2013-11-21 | Beijing University Of Posts And Telecommunications | Method for joint optimization of schedule and resource allocation based on the genetic algorithm |
CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
-
2019
- 2019-01-03 CN CN201910004317.5A patent/CN109800071A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130308570A1 (en) * | 2012-05-17 | 2013-11-21 | Beijing University Of Posts And Telecommunications | Method for joint optimization of schedule and resource allocation based on the genetic algorithm |
CN103902375A (en) * | 2014-04-11 | 2014-07-02 | 北京工业大学 | Cloud task scheduling method based on improved genetic algorithm |
Non-Patent Citations (4)
Title |
---|
任金霞等: "基于遗传算法的云任务调度改进算法", 《江西理工大学学报》 * |
姚虎等: "Cloud Task Scheduling Algorithm based on Improved Genetic Algorithm", 《INTERNATIONAL JOURNAL OF PERFORMABILITY ENGINEERING》 * |
李超等: "云计算环境下基于改进遗传算法的多维约束任务调度研究", 《小型微型计算机系统》 * |
胡艳华等: "基于改进遗传算法的云计算任务调度算法", 《计算机技术与发展》 * |
Cited By (76)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110389819A (en) * | 2019-06-24 | 2019-10-29 | 华中科技大学 | A kind of dispatching method and system of computation-intensive batch processing task |
CN110458291A (en) * | 2019-08-09 | 2019-11-15 | 成都理工大学 | A kind of best common recognition node selecting method based on genetic algorithm |
CN110569117A (en) * | 2019-08-26 | 2019-12-13 | 南瑞集团有限公司 | Power supply station intelligent cloud platform task scheduling algorithm and system based on optimized particle swarm |
CN110609739A (en) * | 2019-09-19 | 2019-12-24 | 湘潭大学 | Train period information scheduling method and system based on multi-objective evolutionary algorithm |
CN110609739B (en) * | 2019-09-19 | 2022-04-08 | 湘潭大学 | Train period information scheduling method and system based on multi-objective evolutionary algorithm |
CN110751292A (en) * | 2019-10-12 | 2020-02-04 | 浙江财经大学 | Cloud manufacturing multi-view collaborative scheduling optimization method based on genetic algorithm |
CN110751292B (en) * | 2019-10-12 | 2022-05-03 | 浙江财经大学 | Cloud manufacturing multi-view collaborative scheduling optimization method based on genetic algorithm |
CN111651434A (en) * | 2019-10-22 | 2020-09-11 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN111651434B (en) * | 2019-10-22 | 2023-10-24 | 摩登汽车有限公司 | Millimeter wave radar data processing method, device and equipment based on genetic algorithm |
CN110851272B (en) * | 2019-10-30 | 2022-02-11 | 内蒙古农业大学 | Cloud task scheduling method based on phagocytic particle swarm genetic hybrid algorithm |
CN110851272A (en) * | 2019-10-30 | 2020-02-28 | 内蒙古农业大学 | Cloud task scheduling method based on phagocytic particle swarm genetic hybrid algorithm |
CN110851257A (en) * | 2019-11-07 | 2020-02-28 | 中国石油大学(华东) | Genetic and differential hybrid evolution cloud computing task scheduling algorithm based on early-stage catastrophe strategy |
CN110851277A (en) * | 2019-11-08 | 2020-02-28 | 中国石油大学(华东) | Task scheduling strategy based on edge cloud cooperation in augmented reality scene |
CN110837413B (en) * | 2019-11-20 | 2023-03-24 | 福州大学 | Computing migration scheduling method for deep neural network application in edge environment |
CN110837413A (en) * | 2019-11-20 | 2020-02-25 | 福州大学 | Computing migration scheduling method for deep neural network application in edge environment |
CN110889552B (en) * | 2019-11-26 | 2023-05-23 | 中国计量大学 | Apple automatic boxing path optimization method based on optimal parameter genetic algorithm |
CN110889552A (en) * | 2019-11-26 | 2020-03-17 | 中国计量大学 | Automatic apple boxing path optimization method based on optimal parameter genetic algorithm |
CN110928669A (en) * | 2019-12-10 | 2020-03-27 | 浙江工商大学 | Energy consumption perception cloud workflow scheduling optimization method based on multi-population genetic algorithm |
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 |
CN111125904A (en) * | 2019-12-20 | 2020-05-08 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment blade sorting method based on multi-target regulation |
CN111125904B (en) * | 2019-12-20 | 2024-01-16 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment blade sequencing method based on multi-target regulation and control |
CN111079229A (en) * | 2019-12-20 | 2020-04-28 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment unbalance dual-target optimization method based on cloud self-adaptive genetic algorithm |
CN113132445A (en) * | 2020-01-10 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
CN111290853A (en) * | 2020-01-14 | 2020-06-16 | 西安建筑科技大学 | Cloud data center scheduling method based on self-adaptive improved genetic algorithm |
CN111290853B (en) * | 2020-01-14 | 2023-06-23 | 西安建筑科技大学 | Cloud data center scheduling method based on self-adaptive improved genetic algorithm |
CN111274030B (en) * | 2020-01-16 | 2022-09-23 | 中国人民解放军国防科技大学 | Efficient multiprocessor system-on-chip design space mining method oriented to application features |
CN111274030A (en) * | 2020-01-16 | 2020-06-12 | 中国人民解放军国防科技大学 | Efficient multiprocessor system-on-chip design space mining method oriented to application features |
CN111338749A (en) * | 2020-02-10 | 2020-06-26 | 长沙学院 | Task scheduling method, system and storage medium |
CN111427688B (en) * | 2020-03-23 | 2023-08-11 | 武汉轻工大学 | Cloud task multi-target scheduling method and device, electronic equipment and storage medium |
CN111427688A (en) * | 2020-03-23 | 2020-07-17 | 武汉轻工大学 | Cloud task multi-target scheduling method and device, electronic equipment and storage medium |
CN113411369A (en) * | 2020-03-26 | 2021-09-17 | 山东管理学院 | Cloud service resource collaborative optimization scheduling method, system, medium and equipment |
CN111641891A (en) * | 2020-04-16 | 2020-09-08 | 北京邮电大学 | Task peer-to-peer unloading method and device in multi-access edge computing system |
CN111641891B (en) * | 2020-04-16 | 2021-11-23 | 北京邮电大学 | Task peer-to-peer unloading method and device in multi-access edge computing system |
CN112035224A (en) * | 2020-07-17 | 2020-12-04 | 中国科学院上海微系统与信息技术研究所 | Fog calculation scheduling method suitable for intelligent factory |
CN112035224B (en) * | 2020-07-17 | 2024-03-12 | 中国科学院上海微系统与信息技术研究所 | Fog calculation scheduling method suitable for intelligent factory |
CN112083967A (en) * | 2020-08-18 | 2020-12-15 | 深圳供电局有限公司 | Unloading method of cloud edge computing task, computer equipment and storage medium |
CN112083967B (en) * | 2020-08-18 | 2023-10-20 | 深圳供电局有限公司 | Cloud edge computing task unloading method, computer equipment and storage medium |
CN111985841B (en) * | 2020-08-31 | 2023-10-24 | 华中科技大学 | Injection workshop scheduling method and system based on improved genetic algorithm |
CN111985841A (en) * | 2020-08-31 | 2020-11-24 | 华中科技大学 | Injection molding workshop scheduling method and system based on improved genetic algorithm |
CN112069081A (en) * | 2020-09-24 | 2020-12-11 | 中国人民解放军国防科技大学 | Genetic algorithm-based directory controller test excitation generation method |
CN112069081B (en) * | 2020-09-24 | 2022-06-03 | 中国人民解放军国防科技大学 | Genetic algorithm-based directory controller test excitation generation method |
CN112181598A (en) * | 2020-10-12 | 2021-01-05 | 内蒙古农业大学 | Cloud computing task scheduling method based on improved genetic algorithm |
CN112181598B (en) * | 2020-10-12 | 2022-12-09 | 内蒙古农业大学 | Cloud computing task scheduling method based on improved genetic algorithm |
CN112379996A (en) * | 2020-10-14 | 2021-02-19 | 华南理工大学 | Cloud computing resource scheduling optimization method based on dynamic target strategy and application |
CN112631612A (en) * | 2020-12-03 | 2021-04-09 | 杭州谐云科技有限公司 | Optimization method for kubernetes cloud platform configuration based on genetic algorithm |
CN112364528B (en) * | 2020-12-09 | 2024-02-13 | 南京长峰航天电子科技有限公司 | Genetic algorithm-based microwave darkroom multipath effect influence assessment method |
CN112364528A (en) * | 2020-12-09 | 2021-02-12 | 南京长峰航天电子科技有限公司 | Microwave darkroom multipath effect evaluation algorithm based on genetic algorithm |
CN112598333A (en) * | 2021-01-07 | 2021-04-02 | 北京城市大数据研究院有限公司 | Community empty nest old people assisting and dispatching method based on genetic algorithm |
CN112819242A (en) * | 2021-02-22 | 2021-05-18 | 西北工业大学 | Civil transportation type airplane flight test task allocation optimization method |
CN112819242B (en) * | 2021-02-22 | 2023-06-09 | 西北工业大学 | Civil transportation airplane flight test task allocation optimization method |
CN113127167B (en) * | 2021-03-18 | 2023-11-03 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
CN113127167A (en) * | 2021-03-18 | 2021-07-16 | 国家卫星气象中心(国家空间天气监测预警中心) | Heterogeneous resource intelligent parallel scheduling method based on improved genetic algorithm |
CN113268376A (en) * | 2021-04-16 | 2021-08-17 | 济南轨道交通集团有限公司 | Data center object storage method and system based on genetic algorithm |
CN113268376B (en) * | 2021-04-16 | 2023-01-31 | 济南轨道交通集团有限公司 | Data center object storage method and system based on genetic algorithm |
CN113326917A (en) * | 2021-04-29 | 2021-08-31 | 开放智能机器(上海)有限公司 | Method and system for automatically optimizing operator based on genetic algorithm |
CN113656073A (en) * | 2021-08-18 | 2021-11-16 | 哈尔滨工业大学 | Parallel scheduling mode recombination method based on large mode set, computer and storage medium |
CN113848725A (en) * | 2021-10-13 | 2021-12-28 | 长春工业大学 | Mechanical arm time optimal trajectory planning and optimizing method based on adaptive genetic algorithm |
CN113986518A (en) * | 2021-12-28 | 2022-01-28 | 华南理工大学 | Real-time cloud task multi-objective optimization scheduling method based on ant colony algorithm |
CN114581058B (en) * | 2022-03-10 | 2023-08-18 | 杭州电子科技大学 | Personnel organization structure optimization method based on business process |
CN114581058A (en) * | 2022-03-10 | 2022-06-03 | 杭州电子科技大学 | Personnel organization structure optimization method based on business process |
CN115185655A (en) * | 2022-06-23 | 2022-10-14 | 郑州轻工业大学 | Genetic task scheduling method based on gene frequency improvement |
CN114943489A (en) * | 2022-07-22 | 2022-08-26 | 浙江中控技术股份有限公司 | Production scheduling method and device based on improved genetic algorithm |
CN114943489B (en) * | 2022-07-22 | 2022-10-21 | 浙江中控技术股份有限公司 | Production scheduling method and device based on improved genetic algorithm |
CN115293449B (en) * | 2022-08-23 | 2023-05-30 | 中国空间技术研究院 | Distribution optimization method for tasks with time window and coupling constraint |
CN115293449A (en) * | 2022-08-23 | 2022-11-04 | 中国空间技术研究院 | Distribution optimization method for tasks with time windows and coupling constraints |
CN116562582A (en) * | 2023-05-23 | 2023-08-08 | 广东建设职业技术学院 | Intelligent distribution method for building engineering detection tasks |
CN116401037B (en) * | 2023-06-05 | 2023-08-22 | 杭州谐云科技有限公司 | Genetic algorithm-based multi-task scheduling method and system |
CN116401037A (en) * | 2023-06-05 | 2023-07-07 | 杭州谐云科技有限公司 | Genetic algorithm-based multi-task scheduling method and system |
CN117057485B (en) * | 2023-10-11 | 2024-01-09 | 山东天鼎舟工业科技有限公司 | Scheduling method and system for casting machining production |
CN117057485A (en) * | 2023-10-11 | 2023-11-14 | 山东天鼎舟工业科技有限公司 | Scheduling method and system for casting machining production |
CN117130622A (en) * | 2023-10-26 | 2023-11-28 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Distributed online code compiling and running method and system |
CN117130622B (en) * | 2023-10-26 | 2024-01-12 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Distributed online code compiling and running method and system |
CN117272838A (en) * | 2023-11-17 | 2023-12-22 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
CN117272838B (en) * | 2023-11-17 | 2024-02-02 | 恒海云技术集团有限公司 | Government affair big data platform data acquisition optimization method |
CN117290077A (en) * | 2023-11-24 | 2023-12-26 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
CN117290077B (en) * | 2023-11-24 | 2024-02-23 | 深圳市飞梵实业有限公司 | Medical platform cloud computing task scheduling method and system based on genetic algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109800071A (en) | A kind of cloud computing method for scheduling task based on improved adaptive GA-IAGA | |
CN107659433B (en) | Cloud resource scheduling method and equipment | |
Liu et al. | Job scheduling model for cloud computing based on multi-objective genetic algorithm | |
CN108182115A (en) | A kind of virtual machine load-balancing method under cloud environment | |
CN104065745A (en) | Cloud computing dynamic resource scheduling system and method | |
CN103631657A (en) | Task scheduling algorithm based on MapReduce | |
Dinesh et al. | Efficient resources allocation for different jobs in cloud | |
CN107203412A (en) | A kind of cloud resource method for optimizing scheduling that particle cluster algorithm is improved based on membranous system | |
CN101582043A (en) | Dynamic task allocation method of heterogeneous computing system | |
Shishira et al. | Survey on meta heuristic optimization techniques in cloud computing | |
CN109213585A (en) | A kind of cloud platform energy optimization management method based on blending inheritance algorithm and ant group algorithm | |
Subramoney et al. | Multi-swarm PSO algorithm for static workflow scheduling in cloud-fog environments | |
CN111813500B (en) | Multi-target cloud workflow scheduling method and device | |
CN109815009A (en) | Scheduling of resource and optimization method under a kind of CSP | |
CN109710372A (en) | A kind of computation-intensive cloud workflow schedule method based on cat owl searching algorithm | |
Samriya et al. | Spider monkey optimization based energy-efficient resource allocation in cloud environment | |
Hosseini et al. | Resource allocation optimization in cloud computing using the whale optimization algorithm | |
Ye et al. | A hybrid instance-intensive workflow scheduling method in private cloud environment | |
CN106802822A (en) | A kind of cloud data center cognitive resources dispatching method based on moth algorithm | |
Zhang et al. | Task scheduling of cloud computing based on Improved CHC algorithm | |
Xu et al. | Hybrid scheduling deadline-constrained multi-DAGs based on reverse HEFT | |
Yu | Research on optimization strategy of task scheduling software based on genetic algorithm in cloud computing environment | |
CN115599522A (en) | Task scheduling method, device and equipment for cloud computing platform | |
Xu et al. | Actor-critic with transformer for cloud computing resource three stage job scheduling | |
CN115185651A (en) | Workflow optimization scheduling algorithm based on cloud computing |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190524 |