<![CDATA[Insider threats currently represent one of the most critical and elusive cybersecurity risks for organizations, as perpetrators already possess legitimate authorized access to the system. Traditional security architectures focusing solely on perimeter defense have proven inadequate in containing lateral movement within database networks, potentially leading to large-scale sensitive data breaches. This research aims to design and implement a granular access control mechanism in MySQL to mitigate such threats. The study proposes a proactive mitigation strategy through the implementation of Row-Level Security (RLS) integrated with a Zero Trust framework. The methodology employs applied experimental research by constructing a controlled simulated database environment. The control mechanism is designed using View-based Access Control to filter data dynamically based on user session identity, and is validated through three specific threat scenarios: mass data exfiltration, horizontal privilege escalation, and vertical prerogative exploitation. Testing results demonstrate that this implementation significantly succeeds in restricting data visibility; in the exfiltration scenario, users could only access 1,500 out of 10,000 data rows, strictly adhering to their authorized scope. Although a computational overhead with an average latency increase of 20,7% was observed, it remains well within the acceptable performance threshold (<30%), ensuring that the principle of least privilege is absolutely enforced without compromising operational efficiency.]]>
Copyrights © 2026
