A Customized Task Scheduling in Cloud Using Genetic Algorithm

Abstract

Cloud computing is an emerging technology in distributed computing which provides pay per use according to user demand and requirement. The primary aim of the Cloud computing is to provide efficient access to distributed resources. Scheduling of task is a critical issue in cloud computing, because it serves many users. The An approach for categorizing the tasks as Hard Real-Time Tasks (critical tasks that need to be completed on time with high rates of confidentiality) and Soft Real-Time Tasks (tasks that can be completed with certain delay and still can be efficient in its own way) before they are scheduled is applied. From the results observed the efficient processor for a particular combination of the tasks is determined thus producing customized results for each of the tasks. Efficient task scheduling is of high criticality for obtaining high performance in heterogeneous multiprocessor systems. Since task scheduling is a NP-hard problem, The Genetic Algorithm, an Evolutionary Algorithm which make use of techniques inspired by evolutionary biology such as inheritance, mutation, selection and crossover that is capable of producing optimal solutions.