Parallel Algorithms for Mean, Variance, and Standard Deviation over a 2D-Mesh Network

Abstract

Abstract: Due to advanced technology and usage of internet at worldwide scale, there has always been a challenge to handle and manage the huge amount of data in an efficient way. There are so many statistical and non-statistical methods applied to data to find result and conclusions that can later be used to solve many other complex problems in an easy and fast way, which also reduces the cost. In this study, we propose algorithm to parallely implement the process of finding mean, variance, standard deviation over a 2D-mesh network. Our proposed algorithm successfully finds standard deviation in 5√N+8 times. The algorithm can be compared with the conventional method for finding standard deviation in N2 times, whereas the standard deviation with parallel implementation takes 5(N-1)+4 times for the similar networks.

Keywords: Fast All Sum, mean, variance, standard deviation etc.

Cite this Article: Gaytri Kumari Gupta, Sudhanshu Kumar Jha. Parallel Algorithms for Mean, Variance, and Standard Deviation over a 2D-Mesh Network. Recent Trends in Parallel Computing. 2019; 6(3): 17–22p.