Implementation of distributed processing using hadoop

Abstract

“Implementation of Distributed Processing using Hadoop” is a project which is going to implement Distributed processing and explore its applications over centralized computing that inculcates Dictatorship, administrative system, under-utilization of resources and higher initial setup. Distributed processing, a phrase that includes parallel processing where single node utilizes one or more processors and resources of other nodes. Its is also known as distributed computing. Distributed processing give more performance than single system. If any one node in this environment goes down another will take care of the application and services that is available to cater the business customers. There are no limitations on the resources, more resources can be added easily. The goal of the project is to implement distributed processing using Hadoop and then to compare the results from distributed computing with that of result from centralized computing. Hadoop is a Java based framework for running applications on large clusters of hardware. Hadoop has a famous component - “HDFS” i.e, “Hadoop Distributed File System”, is a highly fault-tolerant distributed system. Hadoop provides high throughput access to application data and is suitable for application that have large data sets. Hadoop is one of many ways Distributed Computing concept can be implemented, which is used in this project. Hadoop environment is a trending technology for distributed storage and distributed data processing of very large datasets on computer cluster. And so it is solving a big problem of industry - “BIG DATA”.