Adaptive DDoS Protection Architecture for Cloud Servers: A Custom Traffic Filtering Tool

Abstract: Cloud computing architectures are highly susceptible to Distributed Denial of Service (DDoS) attacks that can exceed server processing capabilities. A volumetric DDoS attack overwhelms a targeted system with a massive volume of unwanted network traffic, causing severe CPU and memory exhaustion and preventing legitimate user requests from being processed. This research project presents a localized DDoS protection architecture designed for small to medium- size cloud environments. The proposed approach employs a custom C++ traffic filtering engine positioned at the network edge, utilizing a hash map data structure to track incoming IP addresses and their corresponding request rates. Controlled synthetic traffic is generated using an asynchronous Python script, and the protected and unprotected server states are compared using CPU utilization, memory consumption and request latency as primary evaluation metrics. The experimental results demonstrate that the filtering mechanism can reduce server CPU utilization from 98.7% to 22.4% while maintaining a legitimate request latency of 52 ms, thereby establishing the usefulness of lightweight edge-based filtering for volumetric DDoS protection.

Keywords: DDoS Mitigation, Cloud Security, Traffic Filtering, Hash Map, Edge Computing, Rate Limiting, Network Security.