Duplication based performance effective scheduling

Abstract

In cloud computing (CC), mainly list scheduling algorithms are widely used in task scheduling. The existing list scheduling algorithms are generally not efficient in reducing overall execution time (makespan). So in this paper, we have presented a list scheduling algorithm namely, Performance Effective Task Scheduling (PETS) which is merged with the task duplication method named as Duplication based Performance Effective Scheduling (DPES). Most of the duplication algorithms mainly focus on obtaining high performance by minimizing the makespan without reviewing the energy consumed by an application. But DPES algorithm not only reduces the makespan but also examines the energy consumption. Duplication strategy is used in which the parent tasks have been replicated in order to minimize the makespan while to lower the energy consumption, Dynamic Voltage and Frequency Scaling (DVFS) technique has been applied. In this paper, the DPES algorithm is compared with the PETS algorithm on various performance metrics and DPES algorithm proves to be better in each metrics comparison.