Aegis Hadoop developers will explain the concept of Hadoop Distribution File System (HDFS) to entire development community across the globe. HDFS is intended to adhere to the conventional distributed file systems and Google File System advantages. The Apache Hadoop framework has its own file system- HDFS. It is a fault-tolerant, self-healing system that is developed in Java to store excessive data (in petabytes or terabytes). Regardless of format and schema, HDFS offers high scalability, reliability, and throughput while running on vast clusters of a commodity machine.
HDFS is designed and intended to address most complications and problems associated with conventional file distributed system. Major characteristics of HDFS:
1. Massive storage of data- HDFS supports zeta bytes or petabytes, or more data.
2. Commodity Machine usage – User does not require highly expensive machine to run HDFS. User can run HDFS on easily available commodity hardware. It also helps in managing data and addressing issues without interrupting the client/user.
3. Single writer/ Multiple reader model- HDFS follows the principle ‘Write once-read many’ to resolve issues of data coherency. HDFS is intended for batch processing and not for interactive usage.
4. File system management- Along with conventional file systems, file management system is also supported by HDFS. It includes write, read, rename, delete, relocate, and modify directories or files.
Read More Related This :
Apache hadoop development tools (HDT) aims at promoting plugins in Eclipse to assist developers and make their development task simple on Hadoop platform. In this blog, we are going to overview of few features offered by HDT.