Google File System and Hadoop Distributed File System - An Analogy

Big Data has indeed been the word which IT Industry is talking about lately. With advancement of automation and data being processed in real time, it has now become a necessity for companies to look forward to sustainable solutions to store their huge datasets and compute valuable information out of it. High performance computing heavily relies on distributed environments to process large chunks on data. With ever-increasing data, a reliable and easy to use storage solution has become a major concern for computing. Distributed File System tries to address this issue and provides means to efficiently store and process these huge datasets. This paper presents and compares two common distributed processing frameworks involved in dealing with storage of large amounts of dataGoogle File System (More commonly now known as ‘Colossus’) and Hadoop Distributed File System. Also, it gives an insight to the MAPREDUCE framework common to both GFS and HDFS.