Dissertation (MSc Information Technology)This study sought to evaluate and propose mechanisms to secure relational databases against security vulnerabilities. The study was influenced by the fact that recently, there have been many data breach incidences reported across the world. These haveresulted from vulnerabilities found in database management systems. When these vulnerabilities are exploited, the outcome is the data breach incidences that lead to loss of customer trust for businesses and imparement of national security as a whole. The study utilized two relational databases that are PostgreSQL and Microsoft SQL Server. An emulation experiment with documentary review were employed to collect necessary study data. Identification of security features and vulnerabilities that are found to affect the data tier of the web applications were examined. These vulnerabilities were; sql injection, sensitive data exposure, weak authentication and authorization, insufficient logging and monitoring, using components with known vulnerabilities and security misconfiguration. The findings from the study have shown that Microsoft SQL Server is more security feature-rich in terms of confidentiality, integrity and availability compared to PostgreSQL. Furthermore, the examination of these relational databases against the identified vulnerabilities has revealed that Microsoft SQL server is more resilient to security attacks in its default behavior compared to PostgreSQL. Lastly, the study proposes security frameworks that have control measures to remedy against identified vulnerabilities for each relational database.