ARCHIVES

Database systems today are deployed in relational, NoSQL and cloud environments, yielding a variety of security needs for unauthorized access, data leakage, malicious application interactions, safeguarding sensitive data, etc. In this review, the key database security mechanisms are discussed, focusing on authentication and access control, encryption and data protection, and SQL injection prevention. The literature reviewed is compared and contrasted to assess the security roles, benefits, drawbacks, and operational considerations of these mechanisms in various database settings. The results suggest that there is no single security solution that can offer complete protection. Authentication and authorization control access to resources in the database; fine-grained and attribute-based approaches are more selective, but add to policy management complexity. Encryption mechanisms such as Transparent Data Encryption (TDE), column-level encryption, and encrypted database approaches provide various levels of data protection, but with varying levels of granularity, query capabilities, key-management needs, and processing overhead. SQL injection prevention offers an extra level of security at the application–database interface. The review also identifies that relational, NoSQL, and cloud databases have different security needs, and that more granular or higher levels of security can cause increased administrative and computational complexity. In general, using complementary mechanisms in combination, depending on the threats, data sensitivity, database architecture and the requirements of the target environment, is more effective in securing a database.