Reactive search optimization for scientific workflow scheduling in cloud using clustering techniques

Abstract

With an Internet connection, the clients can rent the necessary newlineservices through web browsers. With the increase in deploying applications in newlinethe environment of the Infrastructure as a Service (IaaS) cloud computing, the newlinedistribution of the tasks workflow to instances of the cloud is critical. newlineScheduling is used for decreasing cost and runtime that has emerged to be a newlinevery important challenge. The primary aim is to build a new schedule that can newlinespecify the time and the resource for each task that is executed. Scheduling in newlinethe cloud is a Non-deterministic polynomial (NP)-hard problem. The primary newlinegoal of the scheduling algorithm within the distributed system was to spread newlineits load on the processors and further maximize their utilization at the same newlinetime, minimizing the time taken for execution of the task. The metaheuristicbased newlinetechniques are used to find near-optimal solutions for complex workflow scheduling problems.The problem workflow scheduling faced an innate issue due todiverse computing environments used, to address such problems of newlinescheduling is an ongoing research area. Features of virtualization that are newlineprovided by the cloud will make both deployments, as well as execution of newlinevarious scientific workflows relatively easy. This will result in some benefits newlinethat are offered by virtualization. Clustering is an important tool in data newlinemining, statistical data analysis, data compression, and vector quantization, newlinetargets data grouping into clusters because the data in each cluster a high newlinedegree of similarity is shared. The clustering is used to grouping data into newlineclusters such that the similarities among data members within the same cluster newlineare maximal, while similarities among data members from dissimilar clusters newlineare minimal. newline newline