Standard Deviation Based Modified Cuckoo Optimization Algorithm for Task Scheduling to Efficient Resource Allocation in Cloud Computing Mahendra Bhatu Gawali IT Dept, Thadomal Shahani Engineering College, Bandra (W), University of Mumbai, Mumbai, MS, India Email: gawali.mahen@gmail.com Subhash K. Shinde Lokmanya Tilak College of Engineering, Kopar Khairane, Navi Mumbai, University of Mumbai, Mumbai, MS, India Email: skshinde@rediffmail.com Abstract—The Cloud Computing has an epochal technology now a day. Managing the incoming request (tasks) to avail- able resources is a challenge for scientist and researchers. This paper proposes a Standard Deviation based Modified Cuckoo Optimization Algorithm (SDMCOA) for task scheduling to efficiently manage the resources. The proposed sys-tem works, in two phases. In the first phase, the sample initial population have been calculated among the available number of task’s population. Rather to take the sample randomly, if an appropriate population’s sample for an experiment are chosen then there are more chances to get optimal result. In second phase, the Cuckoo Optimization Algorithm has been modified with respect to immigration and laying stage. This helps to improve the performance of the system. The experimental results using Cybershake Scientific Workflow shows that the proposed SDMCOA performs better than existing methods BATS, COA in terms of finish time and response time. Index Terms—Cloud Computing, task scheduling, modified cuckoo optimization, resource utilization I. INTRODUCTION Now days, industry and academia both are shifting their traditional way to utilize services offline to online. The technology which can make it possible is known as Cloud Computing. The Cloud Computing centres are responsible to hosts the applications and services such as Software as a Service, Platform as a Service, and Infrastructure as a Service. The Cloud Computing centres are builds of various specification computers or servers which are connected together. The Cloud Computing is the next paradigm of parallel and distributed computing to provide the resources. Utilization of the services by the service user and service provider both collectively Manuscript received July 16, 2017; revised November 28, 2017. formed Service Level Agreement [1]. The Cloud Computing system which is based on “pay-as-you-go” model makes it more powerful than others. The Virtualization [2] is the technology which adds strong corner to Cloud Computing. The Virtualization is actually abstracts the computing resources such as CPU, memory and other physical devices. Whenever, user submits a request to the cloud computing then such virtualization generates virtual machines to fulfil it. The Cloud Computing is basically providing any type of service (software or hardware) over Internet. To provide soft-ware or hardware services to the service user the Cloud Computing should balance thencoming request load with avail-able infrastructure. The 1and1 [3] developed a load balance system where user has the provision to shift one server to another manually. The Amazon Web Service (AWS) cloud provider [4] implemented task placement strategies by Bin-pack algorithm. This Binpack algorithm placed tasks based on the demanding percentage of computing resources such as CPU and memory. It randomly placed the tasks for execution. The Microsoft Azure Scheduler [5] schedules the jobs by kept the job execution result history. The scheduler REST (Representational State Transfer) API is responsible to man-age the interactions between scheduling activities. Round Robin and Least Connection Algorithms are developed by the Century Link [6] Cloud Service Provider. The Rackspace has utilized the Random, Round Robin or Least Connection algorithm to manage the incoming traffic over the avail-able computing infrastructure. If computing resources such as CPU or RAM were not sufficient to incoming tasks demanded then a weighted algorithm has used to handle such situation [7]. In this paper, we focused on task scheduling and resource allocation in Cloud Computing. From the 210 © 2017 J. Adv. Inf. Technol. Journal of Advances in Information Technology Vol. 8, No. 4, November 2017 doi: 10.12720/jait.8.4.210-218