Big Data Platform-A Review

Abstract

Hadoop is popular distributed system used for the analysis of large amount of data. Hadoop is based on distributed computing having HDFS (Hadoop Distributed File System) &Map Reduce programming paradigm. Hadoop is highly fault-tolerant due to its imitation of data transversely on multiple nodes and can be set out on low cost hardware. The file system –HDFS—written in JAVA and designed for heterogeneous hardware and software. Hadoop is very much appropriate for high volume of data & where data format is different like semi structured, unstructured. Hadoop also make available the high speed admittance to the data of the application which we want to use. Hadoop architecture is cluster based (cluster consists of racks), which is consist of nodes (data note, name node), physically separate to each other, in idyllic circumstances. In Hadoop a program known as map-reduce is used to collect data according to query. As Hadoop is used for massive amount of data therefore scheduling and way of containing data in Hadoop must be efficient for better presentation. With this feature of Hadoop the traditional system is replacing with Hadoop. The research objective is to study and explore various scheduling techniques, which are used to increase performance in Hadoop. This paper include the idea of working of Hadoop, its internal details and why Hadoop is better than the Traditional system.