SlideShare una empresa de Scribd logo
1 de 4
Independent Tasks Scheduling Based on Genetic
                Algorithm in Cloud Computing
                    Chenhong Zhao1                             Shanshan Zhang1                                     Qingfeng Liu1
               hongchenzhao@163.com                       lvzongping2006@163.com                             sithlqf@gmail.com
                                               Jian Xie1                                  Jicheng Hu2
                                     xiejian1990@gmail.com                             jicheng@whu.edu.cn
                                      1
                                        International School of Software, Wuhan University
                                                          Wuhan 430079, China
                                      2
                                        State Key Laboratory of Software Engineering, Wuhan University
                                                          Wuhan 430072, China

 Abstract — Task scheduling algorithm, which is an NP-complete-                          Genetic algorithm, based on natural selection and
 ness problem, plays a key role in cloud computing systems. In                       inheritance theory, has been widely and successfully applied in
 this paper, we propose an optimized algorithm based on genetic                      scheduling problems.
 algorithm to schedule independent and divisible tasks adapting to                       In this paper, we introduce an application of GA in task
 different computation and memory requirements. We prompt the                        scheduling in order to adapt to the memory constraints and
 algorithm in heterogeneous systems, where resources (including                      high request of performance in cloud computing. The paper is
 CPUs) are of computational and communication heterogeneity.                         organized as follows: the second part will be depicting
 Dynamic scheduling is also in consideration. Though GA is                           dynamic scheduling and some relative researches in this field;
 designed to solve combinatorial optimization problem, it’s
                                                                                     the third part gives a brief representation of cloud computing;
 inefficient for global optimization. So we conclude with further
                                                                                     the forth part prompts our algorithm; the last part lays out our
 researches in optimized genetic algorithm.
                                                                                     experiment results and concludes with a possible optimized
 Keywords - Cloud computing; Dynamic scheduling; Genetic                             algorithm.
 Algorithm; Heterogeneous System
                                                                                                           II.     DYNAMIC SCHEDULING

                            I.     INTRODUCTION                                      A. Introduction
                                                                                         Scheduling problem is aimed at resources distribution and
     Task scheduling, one of the most famous combinatorial
                                                                                     utilization in a limited environment. It can be defined into two
 optimization problems, plays a key role to improve flexible and
                                                                                     kinds: static scheduling and dynamic scheduling. Static
 reliable systems. The main purpose is to schedule tasks to the                      scheduling uses so-called prescheduling technology to schedule
 adaptable resources in accordance with adaptable time, which                        known tasks in foregone environment, while dynamic
 involves finding out a proper sequence in which tasks can be                        scheduling must depend on not only the foregone tasks and
 executed under transaction logic constraints.                                       environment but also the current system state to make
     There have been a lot of algorithms put forward for task                        scheduling plan[5~7].
 scheduling in many different research fields, whereas few
 covers the arisen field -- cloud computing.                                         B. Relative Rearches
     Cloud computing, which means assigning computation on a                             Dynamic scheduling problem is an NP-complete problem,
 dynamic resource pool composed of massive computers, makes                          whose cost of precise computation will increase exponentially
 users gain computation capability, memory space and software                        with the increase of the problem scale. In fact, we can just find
 services online according to different requirements. In cloud                       out an approximate solution to NP problem [10].
 computing, resources (including CPUs) are of computational                              At present, three general algorithms become popular to
 and communication heterogeneity and are always dynamically                          deal with such problems: Simulated Annealing; Tabu Search;
 collocated. This leads task scheduling in cloud computing to be                     Genetic Algorithm. Among them, GA gains wide attention
 a dynamic scheduling problem. There are mainly two factors of                       because of its simpleness and parallel search.
 uncertainties [1]:
     • Task flow is uncertain. Instances of task flow are                                                   III.    CLOUD COMPUTING
          uncertain; the execution path and times of the flow are                        In this paper computing resources in a cloud computing
          also uncertain.                                                            system is modeled to have a topological space which each node
                                                                                     (computing resource) is assigned as a local resource for some
     •     Resources are uncertain. During the possible long time                    other nodes by connecting weights. Mapping and scheduling of
           of execution, available resources with their quantity                     macro data-flow graphs to a series of suitable resources from
           and form are changing all the way; resources'                             the clouds is a fundamental and challenging problem. We will
           capability, current load, interests and tasks' requests,                  propose an architecture models to minimize the application
           which can effect the scheduling a lot, are dynamic too.                   execution time.


     Sponsored by the National Innovative Project




                                                         978-1-4244-3693-4/09/$25.00 ©2009 IEEE
Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
Looking through the lens of an architect, architecting                                              IV.     SOLUTION BASED ON GA
 software as an application in cloud computing system is an area
 sorely in need of better modeling. Cloud computing system                           A. Fundamental Concepts
 delivers computation as a service. In the mean time SaaS                              1) Task Model
 delivery mechanism will become a main software business                                Suppose that a cloud computing system consist of hetero-
 model and will change the economics of the software industry                        geneous process unit possess of m(m>1) units. The tasks in this
 in a way that allows online information and computing to be                         system features are as follows:
 accessible to many more people in emerging cooperation works
 such as India, China and US, by the technology of cloud                                  a) Tasks are aperiodic; i.e., task arrival times are not
 computing.                                                                          known a priori. Every task has the attributes arrival time (ai ),
     Cloud computing system scales applications by maximizing                        worst case computation time (ci), and deadline (di). The ready
 concurrency and using computing resources more efficiently.                         time of a task is equal to its arrival time.
 One must optimize locking duration, statelessness, sharing                               b) Tasks are nonpreemptive; each of them is
 pooled resources such as task threads and network connections                       independent.
 bus, cache reference data and partition large databases for                              c) Each task has two ways of access to a process unit:(1)
 scaling services to a large number of users.                                        exclusive access, in which case no other task can use the
     Our system model is based on SOA (Service Oriented                              resource with it, or (2) shared access, in which case it can share
 Architecture). Each basic element of the system is considered                       the resource with another task (the other task also should be
 to be a service node. Every service has a hierarchical
                                                                                     willing to share the resource) [2].
 hypergraph structure. Under this model we have two kinds of
 basic procedures: EP (Enwrapping Procedure) and SP                                    2) Scheduler Model
 (Solution Procedure). The EP phase is to enwrap computing                               As with [3,4] in references, this paper also assume
 resource into cloud computing system in which a computing                           centralized scheduling scheme; i.e., a master processor unit in
 resource is semantically parsed and enwrapped as a node                             cloud, collecting all tasks, will take charge of dispatching them
 service. The SP phase is to discretize the input user’s                             to other process units. Each process unit has its own dispatch
 computational problems to be a series of services through a                         queue (DQ). The master unit communicates with other process
 semantic translator. This series is sent to a solver to be solved                   units through these dispatch queues. This organization ensures
 and the result is sent back to semantic translator and the                          that processor units always find some tasks in the dispatch
 composed to be a solution to the input user problem.                                queue when the finish the execution of their current task. The
                                                                                     master unit works in parallel with other units, scheduling the
                                                                                     newly arrived tasks, and periodically updating the dispatch
                            Computing resource                                       queues. Tasks are sorted ascending by the value of deadlines.
                                                                                     B. Genetic Algorithm Scheduling
                                                                                       1) Primary Match
                               Semantic enwrap
                                                                                         Definition 1: CRV (Capability Requirement Vector) is a m-
                                    Wrapper                                          component vector for every task. CRVj(Ti) is the jth value of
                                                                                     the capability task i requires. For example, CRV1can be the
                                                                                     arrival time of task 1
                                 Node service                                             Definition 2: CSV (Capability Supply Vector) is a m-
                                                                                     component vector for every process unit. CSV(Pj) is the jth
                       Figure 1. Enwrapping Procedure                                value of the capability process unit i supplies. For example,
                                                                                     CSV1can be the available time of process unit 1.
                                User problem                                            We now use the two vectors to get possible solutions.
                                                                                        In this phase, we suppose any process unit is available to
                                                                                     each task. Then the primary match algorithm will be:
                          discretizing / composition                                     For task i
                                                                                          For processor unit j
                                                                                           If CRVj(Ti) ⊂ CSV(Pj)
                                                                                                  Put Pj into Candidate Queue of Ti
                            Semantic Translator
                                                                                           Else
                                                                                                 Combine with next CSV
                                                                                                 If CRVj(Ti) ⊂ CombinedCSV1
                          Management of solver                                                   Put {Pj,Pj+1} into Candidate Queue of Ti
                                                                                               Else
                         Figure 2. Solution Procedure                                               Combine with next CSV
                                                                                                   If CRVj(Ti) ⊂ CombinedCSV2
                                                                                                   Put { Pj,Pj+1, Pj+2} into CQ of Ti
                                                                                                      End If




Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
End IF                                                             POP[ChromNum], and initialize the gene number of the
              End IF                                                                 chromosome ,geneNum = 0.
           End For
                                                                                          5) IF geneNum = n, turn to step 3), otherwise the
        End For
                                                                                     implementation of the following steps;
    Comments:
       a) Candidate Queue (i)is for task i from which task i can                           6)set the number of gene of the chromosomes for
 choose a process unit to be executed.                                               geneNum = geneNum +1, to obtain TAgeneNum available process
       b) The combination between CSVs is simply a                                   units number ngeneNum, and then randomly generate a integer
 mathematic addition of the vectors multiplying a coefficient                        between 1 to the integer ngeneNum as the location of the gene;
 matrix.                                                                             duplicate 5).
       c) We consider at most three combinations for one task
 taking the cost of cooperation and communication with more                               c)  fitness function
 combination into account.                                                               Dynamic JSS problem must meet the deadline, so this
    2) Schedule Algorithm                                                            paper assumes if a task fails to meet the deadline, it will get a
    Genetic algorithm is a new technology in the artificial                          punishment for its delay.
 intelligence, recently developed to the very popular inter-                             Objective function:
 discipline .Genetic algorithms and genetic rules are in essence                                                                             n
 a search optimization technology, which starts from one initial                                F = min{ max {ck } + ∑ f1 ( d i − Ti )}
 population, in compliance with the "survival of the fittest                                                  k∈(1, 2 , 3,..., m )
                                                                                                                                            i =1
 survival "principle, through operators such as reproduction,                              Ck is the finish time of the kth process unit. Punish
 hybridization, transformation, and forced to run continuously                                                −W L X , X < 0
 and iterate to approximate to the optimal solution[8]. Solution
 to the problem of the concrete steps are as follows:
                                                                                     function f1 ( x ) =    { WeX , X ≥ 0
                                                                                                                                 ,     di is the deadline of task i

      a) encoding:                                                                   while Ti denotes the actual finish time of task i. WL regulates
                                                                                     the cost of delay and We regulates the cost of store for
       Coding program is to perform optimizing the selection of
                                                                                     advanced tasks.
 the solution of a problem by a way of code, so that it
 corresponds the problem solution space and genetic algorithm                               From above , F is an overall function for all tasks
 code space.                                                                              d) selection operator
       Suppose tasks including (TA1, TA2, ∧, TAn), for ∀TAi ,                              Here we choose a competitive mechanism based on local
                                                                                     selection of the method .
 exists a total of N process unit or combination of process units
 to meet the conditions of its execution. Encoding with digital                            Stage1 Randomly select from among a certain number of
 string, as shown in Figure 3.                                                       individuals (known as the marriage of scale),
                                                                                           Stage2 Compare the relative fitness value as selection
                                                                                     criteria, and choose the high fitness ones into the next
      TA1       TA2        ···             TAi        ···            TAn             generation. General scale selection for marriage 2, and the
                                                                                     numerical value of the fitness value is not directly proportional
       P1        P2        ···              Pi        ···             Xn             so this avoids the effects of super-individual, to a certain extent,
              Figure 3.   Encoding with digital string                               avoids premature convergence and stagnation.
    For TAi, there is a selection of a total of N process units                           e) crossover operator
 available. Each task can only select one process unit, so Pi                              Suppose the crossover rate for bad individual is C1 , and
 represent that task TAi will be executed by the ith of provided                     for good one is C2 . The formula to compute C is:
 N process units up for selection ,and 1 ≤ i ≤N. At the same
                                                                                                             f *C1 , f ≤ − f
 time, introducing the conflict measurement and counteract
 results to the code to ensure that each chromosome in the                                     C ={                    ( C 1 − C 2 )*( f − − f )               (1)
 feasible solution space.                                                                                    C1 −                                   , f > −f
                                                                                                                                     f max − − f
      b) the initial population
                                                                                           The procedure of crossover is:
       In this paper, the algorithm generates initial population as                       1) Get two chromosomes POP[i], POP[j] from previous
 follows:                                                                            generation and work out CFitness[i], CFitness[j].
       1) initialization: set the size of population POP as                               2) Get the crossover ratio by (1) for the two
 popsize.                                                                              chromosomes.
                                                                                          3) Generate a random digit c=Random(0,1);
       2) begin execution: the initial number of chromosome in
                                                                                           If C<c, generate an assembly as D composed of gene
 the population is ChromNum = 0.
                                                                                     points with difference, set D={d1, d2, ∧ , dm }.
         3) IF ChromNum= popsize, end algorithm, output the                               If D is empty, leave the chromosomes to the next
 initial population, otherwise execute following steps.                              generation directly.
                                                                                          Else, randomly crossover two gene points with difference,
     4) Chromosome number for ChromNum = ChromNum                                    and hold the chromosome with higher fitness value in the next
 +1, the ChromNumth chromosome in the population is                                  generation in order to reduce degradation.




Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
f) mutation algorithm
       Suppose a formula for self-adaptive mutation rate M:                                                          REFERENCES
                                        −                                            [1] XIAO Zhi-Jiao, CHANG Hui-You,YI Yang “An Optimization
                          f *M 1 , f ≤ f
           M ={                                         −
                                                                   −   (2)
                                                                                          M ethod of W orkflow Dynamic Scheduling Based on Heuristic
                                                                                          GA ”, Computer Science, Vo1.34 No.2 2007.
                                 (M1−M      2   )*( f − f )                          [2] Indranil Gupta , G. Manimaran , and C. Siva Ram Murthy "A
                          M1−                     −
                                                              ,f > f
                                                                                          New Strategy for Improving the Effectiveness of Resource
                                        f max − f                                         Reclaiming Algorithms in Multiprocessor Real-Time Systems",
       The procedure of mutation is:                                                      Journal of Parallel and Distributed Computing 60, 113133
         1) Work out the mutation ratio for every chromosome                              (2000).
 according to formula (2).                                                           [3] Ramamritham K.,Stankovic A.J.."Efficient scheduling
                                                                                          algorithms for real time multiprocessor systems . IEEE
         2) Select a chromosome randomly using the selection                              Transactions on Parallel and Distributed Systems" 1990 ,
 algorithm mentioned above, set it as POP[i];                                             1(2):184~194
         3) At the mutation position i, generate a new random                        [4] Qiao Ying.Wang Hong An, Dai Guo Zhong. "Developing a
 digit for it and save as POP[new];                                                       new dynamic scheduling algorithm for real-time multiprocessor
                                                                                          systems" . Journal of Software , 2002 , 3(1):5l ~ 58(in
         4) Compute CFitness[new];                                                        Chinese).
         5) IF CFitness[new] ≥ CFitness[i], POP[new] is                              [5] LeeCY,Piramuthu S.Tsai YK."Job shop scheduling with a
 added into the population, else, save the previous chromosome                            genetic algorithm and machine 1earning "           Inr J. Pred
      g) termination algorithm                                                            Res.1997.35(4):1171--1191.
                                                                                     [6] Wang L C, Chen H M, Liu C M."Intelligent scheduling of
     In this paper, the total selection process will be able to copy                      FMS with inductive learning capability using neural networks"[J]
 the chromosome that has the highest fit in the population .The                           lnt.J FMS 1995.7:147 一 l75.
 termination of genetic algorithm is to reserve the chromosome                       [7] Nakasuka S ,Yoshida T "Dynamic scheduling system utilizing
 that 20 generation consecutively keeps the highest fit value,                            machine learning as a knowledge acquisition tool " lnt. J Prod
 when the conditions meet the termination will be based on                                Res. 1992,30 (2): 411—431
 fitness value of the size of sorted chromosomes, output the                         [8] L Ozdamar . A genetic algorithm approach to a general
 results to the decision makers to make decisions.                                        category    project scheduling problem[J].IEEE Trana Syst
                                                                                          Man Cyber,1999,
                                                                                          29(1):44—59.
                V.      EXPERIMENT AND FUTURE WORK                                   [9] Saswati Sarkar “ Optimum Scheduling and Memory
     Our experiment is designed to across different platforms                             Management inInput Queued Switches with Finite Buffer
 where two tasks with distinct CRVS to be dispatched on two                               Space”, 2003 IEEE 1373
 process units, one of which is 32-bit while the other is 64-bit.                    [10] Radoslaw Szymanek and Krzysztof Kuchcinski, “ Task
                                                                                          Assignment and Scheduling under Memory Constraints ” ,
 The two process units can be presumed as mini cloud                                      1089-6503/00$ 10.00 0 2000 IEEE
 computing system.
     The primary matches were first collocated, we continued
 our experiment by setting population size per generation: 20,
 crossover rate for bad individual: 0.9, crossover rate for better
 ones: 0.4, the largest crossover rate: 0.2, the least crossover rate:
 0.1, WL: 1.0, We: 1.0. The algorithm ended at the 85th
 generation. Experiment sums up as follows:
    1) The optimal fitness value at the 100th generation
         reached 2.9.
    2) The utilization of resources is highly developed.
    3) The ration between expected execution time and react
         execution time is 15/8.
     The algorithm in this paper synthetically takes both time
 utilization and resource utilization into consideration, so the
 result is of high signification.
      Researches of task scheduling in cloud computing tends to
 be more and more popular in the future, because the more
 efficient algorithm is used, the more benefits we can get.
 Traditional ways of Job-shop scheduling often leads to over-
 costed problem in cloud computing, while genetic algorithms
 can gain overall scheduling optimization.
      In the future, more attention will be paid on reducing the
 solution space in GA, which is of especial importance in cloud
 computing for its integration of innumerable resources and
 tasks.




Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.

Más contenido relacionado

La actualidad más candente

Optimized Assignment of Independent Task for Improving Resources Performance ...
Optimized Assignment of Independent Task for Improving Resources Performance ...Optimized Assignment of Independent Task for Improving Resources Performance ...
Optimized Assignment of Independent Task for Improving Resources Performance ...ijgca
 
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...ijcsit
 
Scheduling in cloud computing
Scheduling in cloud computingScheduling in cloud computing
Scheduling in cloud computingijccsa
 
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...IDES Editor
 
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing IJORCS
 
RSDC (Reliable Scheduling Distributed in Cloud Computing)
RSDC (Reliable Scheduling Distributed in Cloud Computing)RSDC (Reliable Scheduling Distributed in Cloud Computing)
RSDC (Reliable Scheduling Distributed in Cloud Computing)IJCSEA Journal
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmenteSAT Publishing House
 
Energy-aware VM Allocation on An Opportunistic Cloud Infrastructure
Energy-aware VM Allocation on An Opportunistic Cloud InfrastructureEnergy-aware VM Allocation on An Opportunistic Cloud Infrastructure
Energy-aware VM Allocation on An Opportunistic Cloud InfrastructureMario Jose Villamizar Cano
 
Job Resource Ratio Based Priority Driven Scheduling in Cloud Computing
Job Resource Ratio Based Priority Driven Scheduling in Cloud ComputingJob Resource Ratio Based Priority Driven Scheduling in Cloud Computing
Job Resource Ratio Based Priority Driven Scheduling in Cloud Computingijsrd.com
 
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...csandit
 
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...ijgca
 
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...rahulmonikasharma
 
Task Scheduling methodology in cloud computing
Task Scheduling methodology in cloud computing Task Scheduling methodology in cloud computing
Task Scheduling methodology in cloud computing Qutub-ud- Din
 
Sharing of cluster resources among multiple Workflow Applications
Sharing of cluster resources among multiple Workflow ApplicationsSharing of cluster resources among multiple Workflow Applications
Sharing of cluster resources among multiple Workflow Applicationsijcsit
 

La actualidad más candente (17)

Optimized Assignment of Independent Task for Improving Resources Performance ...
Optimized Assignment of Independent Task for Improving Resources Performance ...Optimized Assignment of Independent Task for Improving Resources Performance ...
Optimized Assignment of Independent Task for Improving Resources Performance ...
 
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...
Estimation of Optimized Energy and Latency Constraint for Task Allocation in ...
 
Scheduling in cloud computing
Scheduling in cloud computingScheduling in cloud computing
Scheduling in cloud computing
 
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
Multilevel Hybrid Cognitive Load Balancing Algorithm for Private/Public Cloud...
 
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing
Max Min Fair Scheduling Algorithm using In Grid Scheduling with Load Balancing
 
RSDC (Reliable Scheduling Distributed in Cloud Computing)
RSDC (Reliable Scheduling Distributed in Cloud Computing)RSDC (Reliable Scheduling Distributed in Cloud Computing)
RSDC (Reliable Scheduling Distributed in Cloud Computing)
 
genetic paper
genetic papergenetic paper
genetic paper
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environment
 
Energy-aware VM Allocation on An Opportunistic Cloud Infrastructure
Energy-aware VM Allocation on An Opportunistic Cloud InfrastructureEnergy-aware VM Allocation on An Opportunistic Cloud Infrastructure
Energy-aware VM Allocation on An Opportunistic Cloud Infrastructure
 
Job Resource Ratio Based Priority Driven Scheduling in Cloud Computing
Job Resource Ratio Based Priority Driven Scheduling in Cloud ComputingJob Resource Ratio Based Priority Driven Scheduling in Cloud Computing
Job Resource Ratio Based Priority Driven Scheduling in Cloud Computing
 
Dn32717720
Dn32717720Dn32717720
Dn32717720
 
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
Fault-Tolerance Aware Multi Objective Scheduling Algorithm for Task Schedulin...
 
C1803052327
C1803052327C1803052327
C1803052327
 
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...
PERFORMANCE FACTORS OF CLOUD COMPUTING DATA CENTERS USING [(M/G/1) : (∞/GDM O...
 
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...
Challenges in Dynamic Resource Allocation and Task Scheduling in Heterogeneou...
 
Task Scheduling methodology in cloud computing
Task Scheduling methodology in cloud computing Task Scheduling methodology in cloud computing
Task Scheduling methodology in cloud computing
 
Sharing of cluster resources among multiple Workflow Applications
Sharing of cluster resources among multiple Workflow ApplicationsSharing of cluster resources among multiple Workflow Applications
Sharing of cluster resources among multiple Workflow Applications
 

Destacado

Dynamic framed slotted aloha algorithms using fast tag estimation
Dynamic framed slotted aloha algorithms using fast tag estimationDynamic framed slotted aloha algorithms using fast tag estimation
Dynamic framed slotted aloha algorithms using fast tag estimationambitlick
 
A novel estimation based backoff algorithm in the ieee
A novel estimation based backoff algorithm in the ieeeA novel estimation based backoff algorithm in the ieee
A novel estimation based backoff algorithm in the ieeeambitlick
 
Label based dv-hop localization against wormhole attacks in wireless sensor n...
Label based dv-hop localization against wormhole attacks in wireless sensor n...Label based dv-hop localization against wormhole attacks in wireless sensor n...
Label based dv-hop localization against wormhole attacks in wireless sensor n...ambitlick
 
An adaptive power controlled mac protocol for
An adaptive power controlled mac protocol forAn adaptive power controlled mac protocol for
An adaptive power controlled mac protocol forambitlick
 
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...ambitlick
 
Backbone nodes based stable routing for mobile ad hoc networks
Backbone nodes based stable routing for mobile ad hoc networksBackbone nodes based stable routing for mobile ad hoc networks
Backbone nodes based stable routing for mobile ad hoc networksambitlick
 
Effective and efficient shape based pattern
Effective and efficient shape based patternEffective and efficient shape based pattern
Effective and efficient shape based patternambitlick
 
Multi hop wireless-networks
Multi hop wireless-networksMulti hop wireless-networks
Multi hop wireless-networksambitlick
 

Destacado (8)

Dynamic framed slotted aloha algorithms using fast tag estimation
Dynamic framed slotted aloha algorithms using fast tag estimationDynamic framed slotted aloha algorithms using fast tag estimation
Dynamic framed slotted aloha algorithms using fast tag estimation
 
A novel estimation based backoff algorithm in the ieee
A novel estimation based backoff algorithm in the ieeeA novel estimation based backoff algorithm in the ieee
A novel estimation based backoff algorithm in the ieee
 
Label based dv-hop localization against wormhole attacks in wireless sensor n...
Label based dv-hop localization against wormhole attacks in wireless sensor n...Label based dv-hop localization against wormhole attacks in wireless sensor n...
Label based dv-hop localization against wormhole attacks in wireless sensor n...
 
An adaptive power controlled mac protocol for
An adaptive power controlled mac protocol forAn adaptive power controlled mac protocol for
An adaptive power controlled mac protocol for
 
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...
Estimating Parameters of Multiple Heterogeneous Target Objects Using Composit...
 
Backbone nodes based stable routing for mobile ad hoc networks
Backbone nodes based stable routing for mobile ad hoc networksBackbone nodes based stable routing for mobile ad hoc networks
Backbone nodes based stable routing for mobile ad hoc networks
 
Effective and efficient shape based pattern
Effective and efficient shape based patternEffective and efficient shape based pattern
Effective and efficient shape based pattern
 
Multi hop wireless-networks
Multi hop wireless-networksMulti hop wireless-networks
Multi hop wireless-networks
 

Similar a Independent tasks scheduling based on genetic

A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENT
A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENTA HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENT
A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENTieijjournal1
 
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENT
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENTA STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENT
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENTpharmaindexing
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmenteSAT Journals
 
Task Scheduling using Hybrid Algorithm in Cloud Computing Environments
Task Scheduling using Hybrid Algorithm in Cloud Computing EnvironmentsTask Scheduling using Hybrid Algorithm in Cloud Computing Environments
Task Scheduling using Hybrid Algorithm in Cloud Computing Environmentsiosrjce
 
Ieeepro techno solutions 2014 ieee dotnet project - a hyper-heuristic sched...
Ieeepro techno solutions   2014 ieee dotnet project - a hyper-heuristic sched...Ieeepro techno solutions   2014 ieee dotnet project - a hyper-heuristic sched...
Ieeepro techno solutions 2014 ieee dotnet project - a hyper-heuristic sched...ASAITHAMBIRAJAA
 
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions   2014 ieee java project - a hyper-heuristic schedulingIeeepro techno solutions   2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic schedulinghemanthbbc
 
A Review: Metaheuristic Technique in Cloud Computing
A Review: Metaheuristic Technique in Cloud ComputingA Review: Metaheuristic Technique in Cloud Computing
A Review: Metaheuristic Technique in Cloud ComputingIRJET Journal
 
International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)IJERD Editor
 
Simulation Based Workflow Scheduling for Scientific Application
Simulation Based Workflow Scheduling for Scientific ApplicationSimulation Based Workflow Scheduling for Scientific Application
Simulation Based Workflow Scheduling for Scientific ApplicationIJCSIS Research Publications
 
Exploiting dynamic resource allocation for
Exploiting dynamic resource allocation forExploiting dynamic resource allocation for
Exploiting dynamic resource allocation foringenioustech
 
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...Editor IJCATR
 
An efficient scheduling policy for load balancing model for computational gri...
An efficient scheduling policy for load balancing model for computational gri...An efficient scheduling policy for load balancing model for computational gri...
An efficient scheduling policy for load balancing model for computational gri...Alexander Decker
 
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...IRJET Journal
 

Similar a Independent tasks scheduling based on genetic (20)

D04573033
D04573033D04573033
D04573033
 
A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENT
A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENTA HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENT
A HYPER-HEURISTIC METHOD FOR SCHEDULING THEJOBS IN CLOUD ENVIRONMENT
 
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENT
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENTA STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENT
A STUDY ON JOB SCHEDULING IN CLOUD ENVIRONMENT
 
A survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environmentA survey of various scheduling algorithm in cloud computing environment
A survey of various scheduling algorithm in cloud computing environment
 
Task Scheduling using Hybrid Algorithm in Cloud Computing Environments
Task Scheduling using Hybrid Algorithm in Cloud Computing EnvironmentsTask Scheduling using Hybrid Algorithm in Cloud Computing Environments
Task Scheduling using Hybrid Algorithm in Cloud Computing Environments
 
N0173696106
N0173696106N0173696106
N0173696106
 
Ieeepro techno solutions 2014 ieee dotnet project - a hyper-heuristic sched...
Ieeepro techno solutions   2014 ieee dotnet project - a hyper-heuristic sched...Ieeepro techno solutions   2014 ieee dotnet project - a hyper-heuristic sched...
Ieeepro techno solutions 2014 ieee dotnet project - a hyper-heuristic sched...
 
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions   2014 ieee java project - a hyper-heuristic schedulingIeeepro techno solutions   2014 ieee java project - a hyper-heuristic scheduling
Ieeepro techno solutions 2014 ieee java project - a hyper-heuristic scheduling
 
A Review: Metaheuristic Technique in Cloud Computing
A Review: Metaheuristic Technique in Cloud ComputingA Review: Metaheuristic Technique in Cloud Computing
A Review: Metaheuristic Technique in Cloud Computing
 
20 26
20 26 20 26
20 26
 
International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)
 
Simulation Based Workflow Scheduling for Scientific Application
Simulation Based Workflow Scheduling for Scientific ApplicationSimulation Based Workflow Scheduling for Scientific Application
Simulation Based Workflow Scheduling for Scientific Application
 
10.1.1.85.7969
10.1.1.85.796910.1.1.85.7969
10.1.1.85.7969
 
Grid computing for load balancing strategies
Grid computing for load balancing strategiesGrid computing for load balancing strategies
Grid computing for load balancing strategies
 
Exploiting dynamic resource allocation for
Exploiting dynamic resource allocation forExploiting dynamic resource allocation for
Exploiting dynamic resource allocation for
 
International Journal of Engineering Inventions (IJEI)
International Journal of Engineering Inventions (IJEI)International Journal of Engineering Inventions (IJEI)
International Journal of Engineering Inventions (IJEI)
 
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...
Cost-Efficient Task Scheduling with Ant Colony Algorithm for Executing Large ...
 
An efficient scheduling policy for load balancing model for computational gri...
An efficient scheduling policy for load balancing model for computational gri...An efficient scheduling policy for load balancing model for computational gri...
An efficient scheduling policy for load balancing model for computational gri...
 
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...
IRJET-Framework for Dynamic Resource Allocation and Efficient Scheduling Stra...
 
335 340
335 340335 340
335 340
 

Más de ambitlick

DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...
DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...
DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...ambitlick
 
Low cost Java 2013 IEEE projects
Low cost Java 2013 IEEE projectsLow cost Java 2013 IEEE projects
Low cost Java 2013 IEEE projectsambitlick
 
Ambitlick ns2 2013
Ambitlick ns2 2013Ambitlick ns2 2013
Ambitlick ns2 2013ambitlick
 
Low cost Java IEEE Projects 2013
Low cost Java IEEE Projects 2013Low cost Java IEEE Projects 2013
Low cost Java IEEE Projects 2013ambitlick
 
Handling selfishness in replica allocation
Handling selfishness in replica allocationHandling selfishness in replica allocation
Handling selfishness in replica allocationambitlick
 
Mutual distance bounding protocols
Mutual distance bounding protocolsMutual distance bounding protocols
Mutual distance bounding protocolsambitlick
 
Moderated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsModerated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsambitlick
 
Efficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsEfficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsambitlick
 
IEEE -2012-13 Projects IN NS2
IEEE -2012-13 Projects IN NS2  IEEE -2012-13 Projects IN NS2
IEEE -2012-13 Projects IN NS2 ambitlick
 
Adaptive weight factor estimation from user review 1
Adaptive weight factor estimation from user   review 1Adaptive weight factor estimation from user   review 1
Adaptive weight factor estimation from user review 1ambitlick
 
Integrated institutional portal
Integrated institutional portalIntegrated institutional portal
Integrated institutional portalambitlick
 
Mutual distance bounding protocols
Mutual distance bounding protocolsMutual distance bounding protocols
Mutual distance bounding protocolsambitlick
 
Moderated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsModerated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsambitlick
 
Efficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsEfficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsambitlick
 
Comments on “mabs multicast authentication based on batch signature”
Comments on “mabs multicast authentication based on batch signature”Comments on “mabs multicast authentication based on batch signature”
Comments on “mabs multicast authentication based on batch signature”ambitlick
 
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...ambitlick
 
Energy efficient protocol for deterministic
Energy efficient protocol for deterministicEnergy efficient protocol for deterministic
Energy efficient protocol for deterministicambitlick
 
A Privacy-Preserving Location Monitoring System for Wireless Sensor Networks
A Privacy-Preserving Location Monitoring System for Wireless Sensor NetworksA Privacy-Preserving Location Monitoring System for Wireless Sensor Networks
A Privacy-Preserving Location Monitoring System for Wireless Sensor Networksambitlick
 

Más de ambitlick (20)

DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...
DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...
DCIM: Distributed Cache Invalidation Method for Maintaining Cache Consistency...
 
Low cost Java 2013 IEEE projects
Low cost Java 2013 IEEE projectsLow cost Java 2013 IEEE projects
Low cost Java 2013 IEEE projects
 
Ambitlick ns2 2013
Ambitlick ns2 2013Ambitlick ns2 2013
Ambitlick ns2 2013
 
Low cost Java IEEE Projects 2013
Low cost Java IEEE Projects 2013Low cost Java IEEE Projects 2013
Low cost Java IEEE Projects 2013
 
Handling selfishness in replica allocation
Handling selfishness in replica allocationHandling selfishness in replica allocation
Handling selfishness in replica allocation
 
Mutual distance bounding protocols
Mutual distance bounding protocolsMutual distance bounding protocols
Mutual distance bounding protocols
 
Moderated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsModerated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroups
 
Efficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsEfficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secrets
 
IEEE -2012-13 Projects IN NS2
IEEE -2012-13 Projects IN NS2  IEEE -2012-13 Projects IN NS2
IEEE -2012-13 Projects IN NS2
 
Adaptive weight factor estimation from user review 1
Adaptive weight factor estimation from user   review 1Adaptive weight factor estimation from user   review 1
Adaptive weight factor estimation from user review 1
 
Integrated institutional portal
Integrated institutional portalIntegrated institutional portal
Integrated institutional portal
 
Embassy
EmbassyEmbassy
Embassy
 
Crm
Crm Crm
Crm
 
Mutual distance bounding protocols
Mutual distance bounding protocolsMutual distance bounding protocols
Mutual distance bounding protocols
 
Moderated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroupsModerated group authoring system for campus wide workgroups
Moderated group authoring system for campus wide workgroups
 
Efficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secretsEfficient spread spectrum communication without pre shared secrets
Efficient spread spectrum communication without pre shared secrets
 
Comments on “mabs multicast authentication based on batch signature”
Comments on “mabs multicast authentication based on batch signature”Comments on “mabs multicast authentication based on batch signature”
Comments on “mabs multicast authentication based on batch signature”
 
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...
Energy-Efficient Protocol for Deterministic and Probabilistic Coverage In Sen...
 
Energy efficient protocol for deterministic
Energy efficient protocol for deterministicEnergy efficient protocol for deterministic
Energy efficient protocol for deterministic
 
A Privacy-Preserving Location Monitoring System for Wireless Sensor Networks
A Privacy-Preserving Location Monitoring System for Wireless Sensor NetworksA Privacy-Preserving Location Monitoring System for Wireless Sensor Networks
A Privacy-Preserving Location Monitoring System for Wireless Sensor Networks
 

Último

Clinical Pharmacy Introduction to Clinical Pharmacy, Concept of clinical pptx
Clinical Pharmacy  Introduction to Clinical Pharmacy, Concept of clinical pptxClinical Pharmacy  Introduction to Clinical Pharmacy, Concept of clinical pptx
Clinical Pharmacy Introduction to Clinical Pharmacy, Concept of clinical pptxraviapr7
 
M-2- General Reactions of amino acids.pptx
M-2- General Reactions of amino acids.pptxM-2- General Reactions of amino acids.pptx
M-2- General Reactions of amino acids.pptxDr. Santhosh Kumar. N
 
Easter in the USA presentation by Chloe.
Easter in the USA presentation by Chloe.Easter in the USA presentation by Chloe.
Easter in the USA presentation by Chloe.EnglishCEIPdeSigeiro
 
How to Show Error_Warning Messages in Odoo 17
How to Show Error_Warning Messages in Odoo 17How to Show Error_Warning Messages in Odoo 17
How to Show Error_Warning Messages in Odoo 17Celine George
 
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptx
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptxPractical Research 1: Lesson 8 Writing the Thesis Statement.pptx
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptxKatherine Villaluna
 
CAULIFLOWER BREEDING 1 Parmar pptx
CAULIFLOWER BREEDING 1 Parmar pptxCAULIFLOWER BREEDING 1 Parmar pptx
CAULIFLOWER BREEDING 1 Parmar pptxSaurabhParmar42
 
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptxmary850239
 
CapTechU Doctoral Presentation -March 2024 slides.pptx
CapTechU Doctoral Presentation -March 2024 slides.pptxCapTechU Doctoral Presentation -March 2024 slides.pptx
CapTechU Doctoral Presentation -March 2024 slides.pptxCapitolTechU
 
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptxSandy Millin
 
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdf
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdfNOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdf
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdfSumit Tiwari
 
Ultra structure and life cycle of Plasmodium.pptx
Ultra structure and life cycle of Plasmodium.pptxUltra structure and life cycle of Plasmodium.pptx
Ultra structure and life cycle of Plasmodium.pptxDr. Asif Anas
 
Presentation on the Basics of Writing. Writing a Paragraph
Presentation on the Basics of Writing. Writing a ParagraphPresentation on the Basics of Writing. Writing a Paragraph
Presentation on the Basics of Writing. Writing a ParagraphNetziValdelomar1
 
3.21.24 The Origins of Black Power.pptx
3.21.24  The Origins of Black Power.pptx3.21.24  The Origins of Black Power.pptx
3.21.24 The Origins of Black Power.pptxmary850239
 
General views of Histopathology and step
General views of Histopathology and stepGeneral views of Histopathology and step
General views of Histopathology and stepobaje godwin sunday
 
How to Filter Blank Lines in Odoo 17 Accounting
How to Filter Blank Lines in Odoo 17 AccountingHow to Filter Blank Lines in Odoo 17 Accounting
How to Filter Blank Lines in Odoo 17 AccountingCeline George
 
How to Solve Singleton Error in the Odoo 17
How to Solve Singleton Error in the  Odoo 17How to Solve Singleton Error in the  Odoo 17
How to Solve Singleton Error in the Odoo 17Celine George
 
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptx
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptxPISA-VET launch_El Iza Mohamedou_19 March 2024.pptx
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptxEduSkills OECD
 
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdfMaximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdfTechSoup
 

Último (20)

Clinical Pharmacy Introduction to Clinical Pharmacy, Concept of clinical pptx
Clinical Pharmacy  Introduction to Clinical Pharmacy, Concept of clinical pptxClinical Pharmacy  Introduction to Clinical Pharmacy, Concept of clinical pptx
Clinical Pharmacy Introduction to Clinical Pharmacy, Concept of clinical pptx
 
Personal Resilience in Project Management 2 - TV Edit 1a.pdf
Personal Resilience in Project Management 2 - TV Edit 1a.pdfPersonal Resilience in Project Management 2 - TV Edit 1a.pdf
Personal Resilience in Project Management 2 - TV Edit 1a.pdf
 
M-2- General Reactions of amino acids.pptx
M-2- General Reactions of amino acids.pptxM-2- General Reactions of amino acids.pptx
M-2- General Reactions of amino acids.pptx
 
Easter in the USA presentation by Chloe.
Easter in the USA presentation by Chloe.Easter in the USA presentation by Chloe.
Easter in the USA presentation by Chloe.
 
Finals of Kant get Marx 2.0 : a general politics quiz
Finals of Kant get Marx 2.0 : a general politics quizFinals of Kant get Marx 2.0 : a general politics quiz
Finals of Kant get Marx 2.0 : a general politics quiz
 
How to Show Error_Warning Messages in Odoo 17
How to Show Error_Warning Messages in Odoo 17How to Show Error_Warning Messages in Odoo 17
How to Show Error_Warning Messages in Odoo 17
 
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptx
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptxPractical Research 1: Lesson 8 Writing the Thesis Statement.pptx
Practical Research 1: Lesson 8 Writing the Thesis Statement.pptx
 
CAULIFLOWER BREEDING 1 Parmar pptx
CAULIFLOWER BREEDING 1 Parmar pptxCAULIFLOWER BREEDING 1 Parmar pptx
CAULIFLOWER BREEDING 1 Parmar pptx
 
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx
3.19.24 Urban Uprisings and the Chicago Freedom Movement.pptx
 
CapTechU Doctoral Presentation -March 2024 slides.pptx
CapTechU Doctoral Presentation -March 2024 slides.pptxCapTechU Doctoral Presentation -March 2024 slides.pptx
CapTechU Doctoral Presentation -March 2024 slides.pptx
 
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx
2024.03.23 What do successful readers do - Sandy Millin for PARK.pptx
 
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdf
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdfNOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdf
NOTES OF DRUGS ACTING ON NERVOUS SYSTEM .pdf
 
Ultra structure and life cycle of Plasmodium.pptx
Ultra structure and life cycle of Plasmodium.pptxUltra structure and life cycle of Plasmodium.pptx
Ultra structure and life cycle of Plasmodium.pptx
 
Presentation on the Basics of Writing. Writing a Paragraph
Presentation on the Basics of Writing. Writing a ParagraphPresentation on the Basics of Writing. Writing a Paragraph
Presentation on the Basics of Writing. Writing a Paragraph
 
3.21.24 The Origins of Black Power.pptx
3.21.24  The Origins of Black Power.pptx3.21.24  The Origins of Black Power.pptx
3.21.24 The Origins of Black Power.pptx
 
General views of Histopathology and step
General views of Histopathology and stepGeneral views of Histopathology and step
General views of Histopathology and step
 
How to Filter Blank Lines in Odoo 17 Accounting
How to Filter Blank Lines in Odoo 17 AccountingHow to Filter Blank Lines in Odoo 17 Accounting
How to Filter Blank Lines in Odoo 17 Accounting
 
How to Solve Singleton Error in the Odoo 17
How to Solve Singleton Error in the  Odoo 17How to Solve Singleton Error in the  Odoo 17
How to Solve Singleton Error in the Odoo 17
 
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptx
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptxPISA-VET launch_El Iza Mohamedou_19 March 2024.pptx
PISA-VET launch_El Iza Mohamedou_19 March 2024.pptx
 
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdfMaximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
Maximizing Impact_ Nonprofit Website Planning, Budgeting, and Design.pdf
 

Independent tasks scheduling based on genetic

  • 1. Independent Tasks Scheduling Based on Genetic Algorithm in Cloud Computing Chenhong Zhao1 Shanshan Zhang1 Qingfeng Liu1 hongchenzhao@163.com lvzongping2006@163.com sithlqf@gmail.com Jian Xie1 Jicheng Hu2 xiejian1990@gmail.com jicheng@whu.edu.cn 1 International School of Software, Wuhan University Wuhan 430079, China 2 State Key Laboratory of Software Engineering, Wuhan University Wuhan 430072, China Abstract — Task scheduling algorithm, which is an NP-complete- Genetic algorithm, based on natural selection and ness problem, plays a key role in cloud computing systems. In inheritance theory, has been widely and successfully applied in this paper, we propose an optimized algorithm based on genetic scheduling problems. algorithm to schedule independent and divisible tasks adapting to In this paper, we introduce an application of GA in task different computation and memory requirements. We prompt the scheduling in order to adapt to the memory constraints and algorithm in heterogeneous systems, where resources (including high request of performance in cloud computing. The paper is CPUs) are of computational and communication heterogeneity. organized as follows: the second part will be depicting Dynamic scheduling is also in consideration. Though GA is dynamic scheduling and some relative researches in this field; designed to solve combinatorial optimization problem, it’s the third part gives a brief representation of cloud computing; inefficient for global optimization. So we conclude with further the forth part prompts our algorithm; the last part lays out our researches in optimized genetic algorithm. experiment results and concludes with a possible optimized Keywords - Cloud computing; Dynamic scheduling; Genetic algorithm. Algorithm; Heterogeneous System II. DYNAMIC SCHEDULING I. INTRODUCTION A. Introduction Scheduling problem is aimed at resources distribution and Task scheduling, one of the most famous combinatorial utilization in a limited environment. It can be defined into two optimization problems, plays a key role to improve flexible and kinds: static scheduling and dynamic scheduling. Static reliable systems. The main purpose is to schedule tasks to the scheduling uses so-called prescheduling technology to schedule adaptable resources in accordance with adaptable time, which known tasks in foregone environment, while dynamic involves finding out a proper sequence in which tasks can be scheduling must depend on not only the foregone tasks and executed under transaction logic constraints. environment but also the current system state to make There have been a lot of algorithms put forward for task scheduling plan[5~7]. scheduling in many different research fields, whereas few covers the arisen field -- cloud computing. B. Relative Rearches Cloud computing, which means assigning computation on a Dynamic scheduling problem is an NP-complete problem, dynamic resource pool composed of massive computers, makes whose cost of precise computation will increase exponentially users gain computation capability, memory space and software with the increase of the problem scale. In fact, we can just find services online according to different requirements. In cloud out an approximate solution to NP problem [10]. computing, resources (including CPUs) are of computational At present, three general algorithms become popular to and communication heterogeneity and are always dynamically deal with such problems: Simulated Annealing; Tabu Search; collocated. This leads task scheduling in cloud computing to be Genetic Algorithm. Among them, GA gains wide attention a dynamic scheduling problem. There are mainly two factors of because of its simpleness and parallel search. uncertainties [1]: • Task flow is uncertain. Instances of task flow are III. CLOUD COMPUTING uncertain; the execution path and times of the flow are In this paper computing resources in a cloud computing also uncertain. system is modeled to have a topological space which each node (computing resource) is assigned as a local resource for some • Resources are uncertain. During the possible long time other nodes by connecting weights. Mapping and scheduling of of execution, available resources with their quantity macro data-flow graphs to a series of suitable resources from and form are changing all the way; resources' the clouds is a fundamental and challenging problem. We will capability, current load, interests and tasks' requests, propose an architecture models to minimize the application which can effect the scheduling a lot, are dynamic too. execution time. Sponsored by the National Innovative Project 978-1-4244-3693-4/09/$25.00 ©2009 IEEE Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
  • 2. Looking through the lens of an architect, architecting IV. SOLUTION BASED ON GA software as an application in cloud computing system is an area sorely in need of better modeling. Cloud computing system A. Fundamental Concepts delivers computation as a service. In the mean time SaaS 1) Task Model delivery mechanism will become a main software business Suppose that a cloud computing system consist of hetero- model and will change the economics of the software industry geneous process unit possess of m(m>1) units. The tasks in this in a way that allows online information and computing to be system features are as follows: accessible to many more people in emerging cooperation works such as India, China and US, by the technology of cloud a) Tasks are aperiodic; i.e., task arrival times are not computing. known a priori. Every task has the attributes arrival time (ai ), Cloud computing system scales applications by maximizing worst case computation time (ci), and deadline (di). The ready concurrency and using computing resources more efficiently. time of a task is equal to its arrival time. One must optimize locking duration, statelessness, sharing b) Tasks are nonpreemptive; each of them is pooled resources such as task threads and network connections independent. bus, cache reference data and partition large databases for c) Each task has two ways of access to a process unit:(1) scaling services to a large number of users. exclusive access, in which case no other task can use the Our system model is based on SOA (Service Oriented resource with it, or (2) shared access, in which case it can share Architecture). Each basic element of the system is considered the resource with another task (the other task also should be to be a service node. Every service has a hierarchical willing to share the resource) [2]. hypergraph structure. Under this model we have two kinds of basic procedures: EP (Enwrapping Procedure) and SP 2) Scheduler Model (Solution Procedure). The EP phase is to enwrap computing As with [3,4] in references, this paper also assume resource into cloud computing system in which a computing centralized scheduling scheme; i.e., a master processor unit in resource is semantically parsed and enwrapped as a node cloud, collecting all tasks, will take charge of dispatching them service. The SP phase is to discretize the input user’s to other process units. Each process unit has its own dispatch computational problems to be a series of services through a queue (DQ). The master unit communicates with other process semantic translator. This series is sent to a solver to be solved units through these dispatch queues. This organization ensures and the result is sent back to semantic translator and the that processor units always find some tasks in the dispatch composed to be a solution to the input user problem. queue when the finish the execution of their current task. The master unit works in parallel with other units, scheduling the newly arrived tasks, and periodically updating the dispatch Computing resource queues. Tasks are sorted ascending by the value of deadlines. B. Genetic Algorithm Scheduling 1) Primary Match Semantic enwrap Definition 1: CRV (Capability Requirement Vector) is a m- Wrapper component vector for every task. CRVj(Ti) is the jth value of the capability task i requires. For example, CRV1can be the arrival time of task 1 Node service Definition 2: CSV (Capability Supply Vector) is a m- component vector for every process unit. CSV(Pj) is the jth Figure 1. Enwrapping Procedure value of the capability process unit i supplies. For example, CSV1can be the available time of process unit 1. User problem We now use the two vectors to get possible solutions. In this phase, we suppose any process unit is available to each task. Then the primary match algorithm will be: discretizing / composition For task i For processor unit j If CRVj(Ti) ⊂ CSV(Pj) Put Pj into Candidate Queue of Ti Semantic Translator Else Combine with next CSV If CRVj(Ti) ⊂ CombinedCSV1 Management of solver Put {Pj,Pj+1} into Candidate Queue of Ti Else Figure 2. Solution Procedure Combine with next CSV If CRVj(Ti) ⊂ CombinedCSV2 Put { Pj,Pj+1, Pj+2} into CQ of Ti End If Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
  • 3. End IF POP[ChromNum], and initialize the gene number of the End IF chromosome ,geneNum = 0. End For 5) IF geneNum = n, turn to step 3), otherwise the End For implementation of the following steps; Comments: a) Candidate Queue (i)is for task i from which task i can 6)set the number of gene of the chromosomes for choose a process unit to be executed. geneNum = geneNum +1, to obtain TAgeneNum available process b) The combination between CSVs is simply a units number ngeneNum, and then randomly generate a integer mathematic addition of the vectors multiplying a coefficient between 1 to the integer ngeneNum as the location of the gene; matrix. duplicate 5). c) We consider at most three combinations for one task taking the cost of cooperation and communication with more c) fitness function combination into account. Dynamic JSS problem must meet the deadline, so this 2) Schedule Algorithm paper assumes if a task fails to meet the deadline, it will get a Genetic algorithm is a new technology in the artificial punishment for its delay. intelligence, recently developed to the very popular inter- Objective function: discipline .Genetic algorithms and genetic rules are in essence n a search optimization technology, which starts from one initial F = min{ max {ck } + ∑ f1 ( d i − Ti )} population, in compliance with the "survival of the fittest k∈(1, 2 , 3,..., m ) i =1 survival "principle, through operators such as reproduction, Ck is the finish time of the kth process unit. Punish hybridization, transformation, and forced to run continuously −W L X , X < 0 and iterate to approximate to the optimal solution[8]. Solution to the problem of the concrete steps are as follows: function f1 ( x ) = { WeX , X ≥ 0 , di is the deadline of task i a) encoding: while Ti denotes the actual finish time of task i. WL regulates the cost of delay and We regulates the cost of store for Coding program is to perform optimizing the selection of advanced tasks. the solution of a problem by a way of code, so that it corresponds the problem solution space and genetic algorithm From above , F is an overall function for all tasks code space. d) selection operator Suppose tasks including (TA1, TA2, ∧, TAn), for ∀TAi , Here we choose a competitive mechanism based on local selection of the method . exists a total of N process unit or combination of process units to meet the conditions of its execution. Encoding with digital Stage1 Randomly select from among a certain number of string, as shown in Figure 3. individuals (known as the marriage of scale), Stage2 Compare the relative fitness value as selection criteria, and choose the high fitness ones into the next TA1 TA2 ··· TAi ··· TAn generation. General scale selection for marriage 2, and the numerical value of the fitness value is not directly proportional P1 P2 ··· Pi ··· Xn so this avoids the effects of super-individual, to a certain extent, Figure 3. Encoding with digital string avoids premature convergence and stagnation. For TAi, there is a selection of a total of N process units e) crossover operator available. Each task can only select one process unit, so Pi Suppose the crossover rate for bad individual is C1 , and represent that task TAi will be executed by the ith of provided for good one is C2 . The formula to compute C is: N process units up for selection ,and 1 ≤ i ≤N. At the same f *C1 , f ≤ − f time, introducing the conflict measurement and counteract results to the code to ensure that each chromosome in the C ={ ( C 1 − C 2 )*( f − − f ) (1) feasible solution space. C1 − , f > −f f max − − f b) the initial population The procedure of crossover is: In this paper, the algorithm generates initial population as 1) Get two chromosomes POP[i], POP[j] from previous follows: generation and work out CFitness[i], CFitness[j]. 1) initialization: set the size of population POP as 2) Get the crossover ratio by (1) for the two popsize. chromosomes. 3) Generate a random digit c=Random(0,1); 2) begin execution: the initial number of chromosome in If C<c, generate an assembly as D composed of gene the population is ChromNum = 0. points with difference, set D={d1, d2, ∧ , dm }. 3) IF ChromNum= popsize, end algorithm, output the If D is empty, leave the chromosomes to the next initial population, otherwise execute following steps. generation directly. Else, randomly crossover two gene points with difference, 4) Chromosome number for ChromNum = ChromNum and hold the chromosome with higher fitness value in the next +1, the ChromNumth chromosome in the population is generation in order to reduce degradation. Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.
  • 4. f) mutation algorithm Suppose a formula for self-adaptive mutation rate M: REFERENCES − [1] XIAO Zhi-Jiao, CHANG Hui-You,YI Yang “An Optimization f *M 1 , f ≤ f M ={ − − (2) M ethod of W orkflow Dynamic Scheduling Based on Heuristic GA ”, Computer Science, Vo1.34 No.2 2007. (M1−M 2 )*( f − f ) [2] Indranil Gupta , G. Manimaran , and C. Siva Ram Murthy "A M1− − ,f > f New Strategy for Improving the Effectiveness of Resource f max − f Reclaiming Algorithms in Multiprocessor Real-Time Systems", The procedure of mutation is: Journal of Parallel and Distributed Computing 60, 113133 1) Work out the mutation ratio for every chromosome (2000). according to formula (2). [3] Ramamritham K.,Stankovic A.J.."Efficient scheduling algorithms for real time multiprocessor systems . IEEE 2) Select a chromosome randomly using the selection Transactions on Parallel and Distributed Systems" 1990 , algorithm mentioned above, set it as POP[i]; 1(2):184~194 3) At the mutation position i, generate a new random [4] Qiao Ying.Wang Hong An, Dai Guo Zhong. "Developing a digit for it and save as POP[new]; new dynamic scheduling algorithm for real-time multiprocessor systems" . Journal of Software , 2002 , 3(1):5l ~ 58(in 4) Compute CFitness[new]; Chinese). 5) IF CFitness[new] ≥ CFitness[i], POP[new] is [5] LeeCY,Piramuthu S.Tsai YK."Job shop scheduling with a added into the population, else, save the previous chromosome genetic algorithm and machine 1earning " Inr J. Pred g) termination algorithm Res.1997.35(4):1171--1191. [6] Wang L C, Chen H M, Liu C M."Intelligent scheduling of In this paper, the total selection process will be able to copy FMS with inductive learning capability using neural networks"[J] the chromosome that has the highest fit in the population .The lnt.J FMS 1995.7:147 一 l75. termination of genetic algorithm is to reserve the chromosome [7] Nakasuka S ,Yoshida T "Dynamic scheduling system utilizing that 20 generation consecutively keeps the highest fit value, machine learning as a knowledge acquisition tool " lnt. J Prod when the conditions meet the termination will be based on Res. 1992,30 (2): 411—431 fitness value of the size of sorted chromosomes, output the [8] L Ozdamar . A genetic algorithm approach to a general results to the decision makers to make decisions. category project scheduling problem[J].IEEE Trana Syst Man Cyber,1999, 29(1):44—59. V. EXPERIMENT AND FUTURE WORK [9] Saswati Sarkar “ Optimum Scheduling and Memory Our experiment is designed to across different platforms Management inInput Queued Switches with Finite Buffer where two tasks with distinct CRVS to be dispatched on two Space”, 2003 IEEE 1373 process units, one of which is 32-bit while the other is 64-bit. [10] Radoslaw Szymanek and Krzysztof Kuchcinski, “ Task Assignment and Scheduling under Memory Constraints ” , The two process units can be presumed as mini cloud 1089-6503/00$ 10.00 0 2000 IEEE computing system. The primary matches were first collocated, we continued our experiment by setting population size per generation: 20, crossover rate for bad individual: 0.9, crossover rate for better ones: 0.4, the largest crossover rate: 0.2, the least crossover rate: 0.1, WL: 1.0, We: 1.0. The algorithm ended at the 85th generation. Experiment sums up as follows: 1) The optimal fitness value at the 100th generation reached 2.9. 2) The utilization of resources is highly developed. 3) The ration between expected execution time and react execution time is 15/8. The algorithm in this paper synthetically takes both time utilization and resource utilization into consideration, so the result is of high signification. Researches of task scheduling in cloud computing tends to be more and more popular in the future, because the more efficient algorithm is used, the more benefits we can get. Traditional ways of Job-shop scheduling often leads to over- costed problem in cloud computing, while genetic algorithms can gain overall scheduling optimization. In the future, more attention will be paid on reducing the solution space in GA, which is of especial importance in cloud computing for its integration of innumerable resources and tasks. Authorized licensed use limited to: Arulmigu Kalasalingam College of Engineering. Downloaded on August 03,2010 at 08:42:05 UTC from IEEE Xplore. Restrictions apply.