DV-DVFS: merging data variety and DVFS technique to manage the energy consumption of big data processing

Abstract Data variety is one of the most important features Big Data. result aggregating data from multiple sources and uneven distribution data. This feature causes high variation in consumption processing resources such as CPU consumption. issue has been overlooked previous works. To overcome mentioned problem, present work, we used Dynamic Voltage Frequency Scaling (DVFS) to reduce energy computation. this goal, consider two types deadlines our constraint. Before applying DVFS technique computer nodes, estimate time frequency needed meet deadline. In evaluation phase, have a set sets applications. The experimental results show that proposed approach surpasses other scenarios real datasets. Based on paper, DV-DVFS can achieve up 15% improvement