BIG DATA MANAGEMENT USING MAP REDUCE TECHNIQUE

Abstract

Most data are unstructured, quasi-structured or semi-structured and homogeneous. The mix of data volume and speed they generate makes it difficult for Big Data to work on the current computing infrastructure. Traditional management, storage and research systems do not have the tools to analyze this data. Big data is precisely stored in the distributed file system architecture because of its precise nature. Apache's Hadoop and HDFS are widely used for storing and managing large data. Big data analysis has a large allocation of file systems, so it must be tolerated with errors. Map Reduce uses the Big Data DBMS Technique for efficient analysis, as well as techniques such as mergers and indexing, as well as other methods, such as graphical searching, for Big Data classification and clustering. Map Reduce is a method of mapping files by indexing files, such as sorting, categorisation, and shrinking. This paper seeks to analyze the state of the art and develop a brief overview of the Big Data clustering using Map Reduce Technique.