Proper

Abstract

Big data deals with large or complex traditional data. The term often refers to size and data. Big data presents a great challenge for database and data analytics research. It is used to get the predictive analysis from large data. It helps in decision making, and to take better decisions based on the given data. This paper consists of comparison between Hadoop Map Reduce and Apache Spark which are used for analyzing Bigdata. Even though both the frameworks are based on Bigdata, their performances differ from level to level and implementation also. In this paper we compare the performance of these both frameworks using k-means clustering algorithm.