Multi objective workflow scheduling and task clustering for scientific workflow applications in cloud

Abstract

Cloud computing is a distributed computing paradigm which has revolutionized the Information and Communication Technology field by providing dynamic and highly scalable resources on demand to end users on a pay as you go model At present it is a promising and fascinating research domain studied for enhancing the performance of information systems and applications that find use in the daily life It provides an alternative to the organizations looking for the goal of building flexible low cost and scalable services which can be accessed over the internet The scalable and elastic nature of cloud platform provides many opportunities for running complex scientific and business workflow applications Many research problems in Physics Biology Astronomy Earth Sciences etc are articulated as scientific workflow applications which are complex as these applications consist of a very large number of tasks with each task being composed of multidimensional scientific and mathematical calculations Scientific community is showing increasing interest in adopting the cloud platform for deploying workflow applications in because of its attractive features like on demand resource provisioning heterogeneous resources pay per use model of resource and flexible billing models The main requirement of scientists for executing workflow applications in cloud platform is to reduce the makespan and the cost of execution Additionally it is also very important for cloud service providers to consider the power consumption of the cloud data center newline