A Scalable AI-Driven Framework for Cybersecurity Training Simulator with Live Network Traffic Visualization

Abstract

As cyber threats evolve, Denial of Service (DoS) attacks remain a primary concern due to their ability to paralyze critical infrastructure by overwhelming server resources. Traditional defense mechanisms, such as static firewalls and signature-based antivirus solutions, often lack the scalability required to mitigate high-volume traffic flooding. This study proposes a proactive DoS mitigation framework leveraging real-time network traffic monitoring and distributed Big Data analytics. The system integrates a high-throughput packet inspection mechanism with a Hadoop MapReduce framework to perform parallelized traffic analysis. By evaluating request payloads against dynamic server capacity thresholds, the framework distinguishes legitimate traffic from malicious flooding attempts at the network edge. The architecture comprises a robust server listening module, a MapReduce-driven network monitor for large-scale packet processing, and a traffic simulation environment to validate system resilience. Experimental evaluations indicate that the proposed framework effectively filters volumetric attacks while maintaining seamless service availability for authorized users. This approach demonstrates that integrating distributed computing paradigms into network security significantly enhances institutional capacity to withstand large-scale cyber disruptions.