Comprehend the performance of mapreduce programming model for K-means algorithm on hadoop cluster

Abstract

MapReduce is a programming model used for processing Big Data. There are had been considerable research in improvement of performance of MapReduce model. This paper examines performance of MapReduce model based on K Means algorithm inside the Hadoop cluster. Different input size had been taken on various configurations to discover the impact of CPU cores and primary memory size. Results of this evaluation had been shown that the number of cores had maximum impact of the performance of MapReduce model.