Calculation of CPU performance, power and cost using Hadoop

摘要

Technology is progressing at a fast pace, and it is the need of the hour to scrutinize big data workloads on cloud computing. The requisite of working with large data sets and more secure platforms gives rise to the necessity of analyzing techniques and tools regarding performance and power consumption. One of the most frequently used big data cloud computing framework is Hadoop. Hadoop - an open source implementation of MapReduce - is employed for processing big data, while handling extensive workload applications for several purposes. The performance, cost and power estimations of underlying processors on Hadoop can be done by characterizing different workloads on Hadoop framework. A comprehensive study is carried out to calculate CPU performance, cost and power consumption regarding different workloads, so as to facilitate efficient usage of different processors to deal with various workloads according to the user requirements.