The following 3 tips should offer a simple approach to mitigating the security risks of big data:
Managing risk 1: unauthorised access
Unauthorised access could lead to a loss of data, or even a data security leak. That is why it is important to have tight management on who can access big data and when. Typically, only the employees that need access privileges in order to do their jobs should be granted those privileges.
Managing risk 2: privilege escalation
As with risk 1, granting unneeded access privileges increases the risk of insider threats to the data security of your organisation. Even administrators should not have complete access to Hadoop clusters, only the access needed to do their jobs.
Managing risk 3: lack of visibility
Without session recording, it becomes almost impossible to identify and solve potential data security issues.
The solution: centralised identity management
Centralised identity management helps making big data environments more secure. This is done by implementing the necessary per-user access control and privilege management.
This should allow organisations to grant the necessary access privileges based on the employee’s role or job function. Furthermore, the solution allows administrators to access certain data, functions, and nodes, w ithout needing complete root access.
In short, by taking advantage of centralised identity management, the organisation effectively decreases the risk of an insider data security leak, and can detect suspicious activity while demonstrating compliance.