Optimizing Resource Allocation and Enhancing Security in Cloud Systems: A Data-Centric Approach

High-performance resource orchestration and robust data security represent critical, often competing, operational objectives in modern cloud computing architectures. This study presents a unified infrastructure framework designed to reconcile these requirements by integrating a Geographically Aware Placement Algorithm (GAPA) with an Artificial Intelligence-Driven Monitoring system (AIDAM). GAPA dynamically optimizes workload distribution based on regional server capacity and geographic proximity, while AIDAM leverages deep unsupervised autoencoders for real-time anomaly detection and threat mitigation. The framework was evaluated via deterministic simulation using production traces from the Google Cluster Data (2019) corpus under a systematic injection of volumetric Distributed Denial-of-Service (DDoS) anomalies. The empirical results demonstrate a 92% macro-averaged threat detection accuracy rate against low-and-slow traffic variations alongside a minimal cryptographic processing latency overhead of 3–5% relative to an unencrypted baseline scheduling configuration. Furthermore, the integrated pipeline achieved a 25% reduction in end-to-end network latency compared to traditional non-geographically aware heuristic models. These findings demonstrate that cloud infrastructure efficiency and security resilience can be simultaneously enhanced without requiring comprehensive physical re-engineering.