An Efficient Approach for Task Scheduling Based on Multi-Objective Genetic Algorithm in Cloud Computing Environment

IJCSC VOLUME 5 NUMBER 2 JULY-SEPT 2014 PP. 110-115 ISSN-0973-7391 An Efficient Approach for Task Scheduling Based on Multi-Objective Genetic Algorithm in Cloud Computing Environment 1 Sourabh Budhiraja, 2 Dr. Dheerendra Singh 1 Student, M.Tech CSE, SUSCET, Tangori, Mohali, Punjab, India 2 Professor, Dept. of CSE, SUSCET, Tangori, Mohali, Punjab, India 1 er.sourabh.cse@gmail.com, 2 professordsingh@gmail.com Abstract : Cloud computing is recently a booming area and has been emerging as a commercial reality in the information technology domain. Cloud computing represents supplement, consumption and delivery model for IT services that are based on internet on pay as per usage basis. Its ability to reduce cost associated with computing while increasing flexibility and scalability for computer process has proved to be a great advantage. The scheduling of the cloud services to the consumers by service providers influences the cost benefit of this computing paradigm. In such a scenario, Tasks should be scheduled efficiently such that the execution cost and time can be reduced. In this paper, we proposed an efficient approach for task scheduling based on Multi- Objective Genetic Algorithm (MOGA) which minimizes execution time and execution cost as well. For task scheduling, a Multi-Objective genetic algorithm is implemented and the research is focused on crossover operators, mutation operators, selection operators and the Pareto solutions method. The experimental results show that the proposed algorithm can obtain a better solution. Index Terms- Task Scheduling, Cloud computing, Multi-objective Genetic Algorithm, CloudSim. I. Introduction Cloud computing is the latest buzzword in the IT industry. It is an emerging computing paradigm with the foundations of grid computing, utility computing, service oriented architecture, virtualization and web 2.0. The user can access all required hardware, software, platform, applications, infrastructure and storage with the ownership of just an Internet connection. A cloud is a type of parallel and distributed system a collection of interconnected and virtualized computer that are dynamically provisioned and presented as one or more unified computing resources based on service level agreements established through negotiation between the service providers and consumers. In this information technology oriented growing market of businesses and organizations, cloud computing is an emerging and attractive alternative to satisfy their day by day increasing needs. It provides virtual resources that are dynamically scalable. It describes virtualized resources, software, platforms, applications, computations and storage to be scalable and provided to users instantly on payment for only what they use [1]. 1.1 Cloud Computing Service Models In Cloud Computing the term Cloud is used for the service provider, which holds all types of resources for storage, computing etc. Mainly three types of services are provided by the cloud. First is Infrastructure as a Service (IaaS), which provides cloud users the infrastructure for various purposes like the storage system and computation resources. Second is Platform as a Service (PaaS), which provides the platform to the clients so that they can make their applications on this platform. Third is Software as a Service (SaaS), which provides the software to the users; so users don t need to install the software on their own machines and they can use the software directly from the cloud. Due to the wide range of facilities provided by the cloud computing, the Cloud Computing is becoming the need of the IT industries. The services of the Cloud are provided through the Internet. The devices that want to access the services of the Cloud should have the Internet accessing capability. Devices need to have very less memory, a very light operating system and browser. Cloud Computing provides many benefits: it results in cost savings because there is no need of initial installation of much resource; it provides scalability and flexibility, the users can increase or decrease the number of services as per requirement; maintenance cost is very less because all the resources are managed by the Cloud providers [2]. 1.2 Problem Areas in Cloud Computing Security: Security issues faced by cloud providers (organizations providing software, platform, or infrastructure as a service via the cloud) and security issues faced by their customers [3]. Fault Tolerance: The increasing popularity of Cloud computing as an attractive alternative to classic information processing systems has increased the importance of its correct and continuous operation even in the presence of faulty components [4]. Resource Discovery: Cloud computing is an emerging field in computer science. Users are utilizing less of their own existing resources, while increasing usage of cloud resources. With the emergence of new technologies such as mobile devices, these devices are usually under-utilized, and can provide similar functionality to a cloud provided they are properly configured and managed. Resource

discovery and allocation is critical in designing an efficient and practical distributed cloud [5]. Load Balancing: Cloud Computing is an emerging computing paradigm. It aims to share data, calculations, and service transparently over a scalable network of nodes. Since Cloud computing stores the data and disseminated resources in the open environment. So, the amount of data storage increases quickly [6]. Task Scheduling: Task scheduling and provision of resources are main problem areas in both Grid as well as in cloud computing. Cloud computing is emerging technology in IT domain. The scheduling of the cloud services to the consumers by service providers influences the cost benefit of these computing paradigms [2]. 1.3 Multi-objective Optimization Optimization deals with the problems of seeking solutions over a set of possible choices to optimize certain criteria. If there is only one criterion to be taken into consideration, they become single objective optimization problems, which have been extensively studied for the past 50 years. If there are more than one criterion which must be treated simultaneously, we have multi-objective optimization problems [14]. Multiple objective problems arise in the design, modeling and planning of complex real systems in area of industrial production, urban transportation, capital budgeting, forest management, reservoir management, layout and landscaping of new cities, energy distribution, etc. It is easy to see that almost every important real-world decision problems involves multiple and conflicting objectives which need to be tackled while respecting various constraints, leading to overwhelming problem complexity. The multiple objective problems have been receiving growing interest from researchers with various backgrounds since early 1960. There are a number of scholars who have made significant contributions to the problem. Among them Pareto is perhaps one of the most recognized pioneers in the field [14]. 1.3.1 Non-dominated or Pareto optimal Solutions In principle, multiple objective optimization problems are very different from single objective optimization problems. For a single objective case, one attempts to obtain the best solution, which is absolutely superior to all other alternatives. In the case of multiple objectives, there does not necessarily exist such a solution that is the best with respect to all other objectives because of incommensurability and conflict among objectives. A solution may be best in one objective but worst in the other objectives. Therefore, there usually exist a set of solutions for the multiple objective case which cannot simply be compared with each other. Such kind of solutions are called non-dominated solutions or Pareto optimal solutions, for which no improvement in any objective function is possible without sacrificing at least one of the other objective functions [14]. Fig. 1 Pareto-optimal solutions (maximization case)(adopted from[14]) II. Related Work In 2007, Yin H., Wu H., Zhou J gave an improved genetic algorithm with limited number of iteration to schedule the independent tasks onto Grid computing resources. The evolutionary process was modified to speed up convergence as a result of shortening the search time, at the same time obtaining a feasible scheduling solution [7]. In 2009, Zhao C., Zhang S., Liu Q., Xie J., Hu J provided an approach for an optimized algorithm based on genetic algorithm to scheduled independent and divisible tasks adapting to different computation and memory requirements. The algorithm in heterogeneous systems, where resources(including CPUs) were of computational and communication heterogeneity[8]. In 2010, Huang Q.Y., Huang T.L presented the conventional job scheduling strategies which focus on efficiency does not meet user requirements and market demand. In this paper, a job scheduling strategy was proposed and algorithm based on QoS, which could meet user requirements on time and cost [9]. In 2011, Verma R., Dhingra S implemented a genetic algorithm for MPTS in permutation flow shop scheduling environment, where the processing order of the jobs was same on all the processors. Genetic Algorithms was applied for the solution of this problem [10]. In 2012, Kumar P., Verma A. designed and tested an algorithm which was made by combining Min-Min and Max-Min in Genetic Algorithm. It was able to schedule multiple jobs on multiple machines in an efficient manner such that the jobs take the minimum time for completion [11]. In 2012, Kaur S.,Verma A proposed a modified genetic algorithm for single user jobs in which the fitness was developed to encourage the formation of solutions to achieved the time minimization and compared it with existing heuristics [2]. In 2013, Liu J., Luo X., Zhang X., Zhang F., Li B., describes a solving method based on multi-objective genetic algorithm (MO-GA) is designed and the research is focused on encoding rules, crossover operators, selection operators and the method of sorting Pareto solutions[12]. III. Problem Formulation Task scheduling and provision of resources are main problem areas in both Grid as well as in cloud computing. From the study of related work, we concluded that the existing scheduling strategies in clouds are based on the

approaches developed in related areas such as distributed systems and Grids. Scheduling in these areas is mainly tailored toward ensuring single application Service Level Agreement (SLA) objectives [2]. In cloud environment on the other hand require guarantying numerous SLA objectives and quality of service. There are many algorithms like Min-Min, Max- Min, Suffrage, Shortest Cloudlet to Fastest Processor (SCFP), Longest Cloudlet to Fastest Processor (LCFP) and some meta-heuristics like Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant-Colony Optimization (ACO) and Simulated Annealing (SA) already existing for task scheduling [2]. Our propped work focuses on optimizing the task scheduling algorithm, a Multi-Objective Genetic algorithm is implemented and the research is focused on crossover operators, mutation operators, selection operators and the Pareto solutions method to achieve the minimization of both Cost and Makespan for better scheduling of Jobs to the resources. IV. Algorithm Description Our main purpose is to schedule tasks to the adaptable resources in accordance with adaptable time, which involves finding out a proper sequence in which all the tasks can be executed such that execution time and execution cost can be minimized. Cost is also an important parameter as the cloud computing services by service providers to service consumers are provided on internet on pay as per usage basis. The Genetic Algorithm is a flexible approach enabling, for the task scheduling problem, different individual representations and algorithm implementations to select individuals and perform crossover and mutation. A Genetic algorithm (GA) is a search heuristic that mimics the process of natural evolution. This heuristic is routinely used to generate useful solutions to optimization and search problems. Genetic algorithms belong to the larger class of evolutionary algorithms (EA), which generate solutions to optimization problems using techniques inspired by natural evolution, such as inheritance, mutation, selection, and crossover. However, the appropriate representation of potential solutions is crucial to ensure that the mutation of any pair of individual (i.e. chromosome) will result in new valid and meaningful individual for the problem. An output schedule of tasks is an array list of population (called chromosomes or the genotype of the genome), which encode candidate solutions to an optimization problem, evolves toward better solutions. Time minimization will give profit to service provider and less maintenance cost to the resources. It will also provide benefit to cloud s service users as their application will be executed at reduced cost [2]. 4.1Existing Algorithm Standard Genetic Algorithm (SGA)(taken from[2]) Produce an initial population by randomly generated individuals Evaluate the fitness of all individuals while termination condition not met do o select fitter individuals for reproduction o crossover between individuals o mutate individuals o evaluate the fitness of the modified individuals o Generate a new population End while 4.2 Proposed Algorithm Modified Genetic Algorithm (MGA) Generate an initial population of individuals with output schedules of algorithm MOGA(Multiobjective genetic algorithm). Evaluate the fitness of all individuals. Call Pareto optimal algorithm. Archive building. While termination condition not met do o Select fitter individuals for reproduction with minimum execution time. o Crossover between individuals by twopoint crossover. o Mutate individuals by simple swap operator. o Evaluate the fitness of the modified individuals having relevant fitness. o Generate a new population. o Best schedule End while Fig.2: Flow Chart-Modified Genetic Algorithm (MGA)

V. Implementation and Results In our proposed work, the objective to analyze the performance of genetic algorithm in minimizing the Makespan and Cost of the processors and to find the best scheduling of Jobs on the Resources. MGA(Modified genetic algorithm) is implemented on Intel core i3 machine with 500 GB HDD, 4 GB RAM on Windows 8 OS, Java language version 1.6 and simulation work is performed using CloudSim3.0[13] toolkit on Netbeans 7.3. Table 1: GA Parameters Parameter Value No. of Resources 9 No. of Jobs 13 Population size 5 Number of Iterations 30 Crossover Type One-point Crossover Crossover Probability 0.5 Mutation Type Simple swap Mutation Probability 0.015 Tournament Size 4 Termination Condition Number of Iterations A good scheduling algorithm is that which leads to better resource utilization, less average Make-span and better system throughput. Make-span refers to the completion time of all cloudlets in the list. To formulate the problem we considered Jobs ( J1, J2, J3..Jn) run on Resources(, R2, R3..Rn).. Our objective is to minimize the Make-span and Cost. The speed of processors is expressed in MIPS (Million instructions per second) and length of job can be expressed as number of instructions to be executed. Each processor is assigned varying processing power and respective cost in Indian rupees. We have computed the Make-span (completion time of tasks) and the corresponding Cost of output schedules from the proposed algorithm. Fig. 4: Makespan v/s Number of Population Generations. The Fig. 3 and Fig. 4 shows the Makespan refers to execution time calculated in seconds and Cost with the Population Generations in the Multi-Objective Genetic Algorithm. Experimental resulting values show that our proposed algorithm takes less execution time and less Cost based on the random generation of schedules. By modifying the SGA with stochastic operators we got the better results and better resource utilization as task load is shared equally on all resources. Jobs J1 J2 J3 J4 J5 J6 J7 J8 J9 J10 J11 J12 J13 Resources R7 R2 R8 R3 R5 R2 R6 Table 2: Best scheduling of Jobs and Resources when generation=30 and Population size=5. Fig. 3: Cost v/s Number of Population Generations. After getting the best scheduling from the algorithm the cloud simulation is started. The tasks are executed on the virtual machines.

VI. Conclusion and Future Scope This thesis have concluded that the proposed approach leads to the better results in a modified genetic algorithm approach in which MOGA(multi-objective genetic algorithm) is implemented on cloud computing environment for scheduling of jobs on the processors. The fitness is developed to encourage the formation of solutions to achieve both the cost and makespan minimization. The performance of GA(on minimize makespan and cost of the processor) with the variation of its control parameters is evaluated. Increasing the Population generation and making population size constant enables the genetic algorithm to obtain minimum cost and makespan which result in a best scheduling. In future, we will be further enhancing our algorithm by supporting runtime scheduling and also considering the user s quality of service and priority of jobs for multiple users. REFERENCES 1. Kaur, P.D., Chana, I. Unfolding the distributed computing paradigm,in: International Conference on Advances in Computer Engineering, pp. 339-342 (2010) 2. Kaur S.,Verma A., An Efficient approach to genetic algorithm for task scheduling in cloud computing environment, I.J. Information Technology and Computer Science, 2012, 10, 74-79. 3. Wikipedia:http://en.wikipedia.org/wiki/Cloud_co mputing_security 4. V.Piuri, Design of fault-tolerant distributed control systems,instrumentation and Measurement, IEEE Transactions on (Volume:43, Issue: 2 ),pp.257-264,1994. 5. Khethavath, P., Thomas, J.,Chan-Tin, E.,Hong Liu, Introducing a distributed cloud architecture with efficient resource discovery and optimal resource allocation, Services (SERVICES), 2013 IEEE Ninth World Congress on 10.1109/SERVICES.2013.68. 6. Sureshbabu G.N.K, Srivatsa S.K., A Review of Load Balancing Algorithms for Cloud Computing, International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume -3 Issue -9 September, 2014 Page No. 8297-8302 Algorithm in Cloud Computing, IEEE 5th International Conference on Wireless Communications, Networking and Mobile Computing WiCom '09, Beijing, pp.1-4, 2009 9. Huang Q.Y., Huang T.L., An Optimistic Job Scheduling Strategy based on QoS for Cloud Computing, IEEE International Conference on Intelligent Computing and Integrated Systems (ICISS), 2010, Guilin, pp. 673-675, 2010 10. Verma R., Dhingra S., Genetic Algorithm for Multiprocessor Task Scheduling, IJCSMS International Journal of Computer Science and Management Studies, Vol.1, Issue 02, pp. 181-185, 2011 11. Kumar P., Verma A., Independent Task Scheduling in Cloud Computing by Improved Genetic Algorithm, International Journal of Advanced Research in Computer Science and Software Engineering(Volume 2, Issue 5, May 2012) 12. Liu J., Luo X., Zhang X., Zhang F., Li B., Job Scheduling Model for Cloud Computing Based on Multi-Objective Genetic Algorithm,1, 2, 3, 4, 5 National Digital Switching System Engineering & Technology Research Center, Zhengzhou 450002, China. 13. Rodrigo N. Calheiros, Rajiv Ranjan, Anton Beloglazov, Cesar A. F. De Rose, and Rajkumar Buyya, CloudSim: A Toolkit for Modeling and Simulation of Cloud Computing Environments and Evaluation of Resource Provisioning Algorithms, Software: Practice and Experience (SPE), Volume 41, Number 1, Pages: 23-50, ISSN: 0038-0644, Wiley Press, New York, USA, January, 2011. 14. Network Models and Optimization. Multiobjective Genetic Algorithm Approach. Series: Decision Engineering. Gen, Mitsuo, Cheng, Runwei, Lin, Lin. 2008. 7. Yin H., Wu H., Zhou J., An Improved Genetic Algorithm with Limited Iteration for Grid Scheduling, IEEE Sixth International Conference on Grid and Cooperative Computing, 2007. GCC 2007, Los Alamitos, CA, pp. 221-227, 2007 8. Zhao C., Zhang S., Liu Q., Xie J., Hu J., Independent Tasks Scheduling Based on Genetic