No one can doubt on Hadoop’s flexibility and scalability as it is a platform that supports all kinds of analytics-focused, data-centric app solutions. Though the platform was not initially intended and developed with security or data governance. Both the components – security and data governance, are highly critical to any data platform that anticipates breaking into the enterprise mainstream. The communities- open source and the vendor- have made major step in both areas in recent years, but still need lot of improvements. Big data and hadoop community joins hands to solve security challenges as these are critical issues, which will continue to prevent practitioners from using and deploying Hadoop to mission critical apps and support production-grade work pressure.
Hadoop Security Evolution :
Several transformations have made in Hadoop platform and in just a decade, it has come a long way. It was initially developed in 2004 to help Yahoo! Engineers index the WWW (World Wide Web). Since the time, Hadoop has transformed into a highly flexible, scalable big data platform that supports all kinds of data processing workloads and data-centric, analytics driven applications.
Security was never on the mind of engineers who developed the platform due to several reasons. Hadoop was also lacked in most basic security capabilities, like user authentication and permission controls. Since time, major advancements have been made to allow Hadoop to handle new kinds of workloads and applications, numerous security capabilities have been created by both the communities.
The Hadoop community joins together to resolve security issue and challenges because these issues will keep stopping developers from deploying Hadoop to support mission-critical software applications and production-grade workload.
The three A’s of Data Protection and Security :
Security and governance in Hditadoop needs similar approaches followed by traditional data management world. It comprises of three A’s of security and data protection:
1. Authentication
2. Authorization
3. Auditing
4. Data protection
Authentication – It’s a process of precisely determining the given user identity attempting to access application or Hadoop cluster based on one of a numerous factors.
Authorization – Once the user ID is determined, authorization of data occurs. Under this process, what data, data types or hadoop applications available for user access are the things determined.
Auditing – Auditing is a recording and reporting process. It is done while authorized user once gets the access to the cluster- it records and makes a report on what access, changes, and additions done to data.
Data Protection – It refers to the use of techniques such as data masking and encryption to protect highly sensitive data against unauthorized users and apps access.
