Comparative Analysis of Hadoop Security Ad-Ons
Khushbu Khalid Butt, Guo Hui Li, Muhammad Obaid Ur Rehman · 2019
Hadoop is a big-data processing framework which is widely used for data storage and processing. Now-a-days security is one of the major concerns in the digital world. Any system is only considered reliable when it provides proper measures to secure the valuable data of an organization. In this paper we discuss the security hadoop offers for data stored in it. Map-Reduce (MR) was introduced by Google as a batch processing implementation in their Google File System (GFS). Hadoop provides its own Hadoop Distributed File System (HDFS) for storage and it used MR at its core to process the data stored in the HDFS. The early Hadoop project established a security stance that the entire cluster of machines and all of the users accessing it are part of a trusted network. Hadoop framework has a lot of components in it and in the earlier versions it only offered unix based POSIX security for the data, it did not include authentication and authorization between clients and services. In the recent years many security solutions were offered by different enterprises in the form of open-source tools and distributions for Hadoop that had security integrated in them. The solutions for Hadoop security are evolving day by day and a-lot of man power is being invested by academia, open-source community and enterprises to further improve these solutions. Because of the work of the community, the Hadoop can now be deployed in production with sufficient security controls.