A Scalable AI-Driven Resource Allocation Model for Sustainable Green Cloud Computing Infrastructures

Abstract

The rapid expansion of cloud computing technologies, artificial intelligence applications, Internet of Things (IoT) ecosystems, big data analytics platforms, and distributed enterprise services has significantly increased the demand for scalable computational infrastructures and intelligent resource management systems. Modern cloud data centers continuously process massive computational workloads generated from heterogeneous applications requiring adaptive resource allocation, workload balancing, energy-efficient scheduling, and sustainable infrastructure optimization. However, traditional cloud resource allocation mechanisms frequently suffer from inefficient workload distribution, excessive energy consumption, poor scalability, underutilized computational resources, and increased carbon emissions. As global data center energy demand continues to rise, sustainable green cloud computing has emerged as a critical research challenge in next-generation distributed computing infrastructures. This research proposes a Scalable AI-Driven Resource Allocation Model for Sustainable Green Cloud Computing Infrastructures. The proposed framework integrates artificial intelligence-driven scheduling, deep reinforcement learning, transformer-based workload analytics, graph neural infrastructure coordination, adaptive energy-aware optimization, and explainable cloud intelligence to support scalable and sustainable cloud resource management. The architecture dynamically optimizes virtual machine allocation, computational workload balancing, communication overhead, energy consumption, and infrastructure utilization across distributed green cloud environments while maintaining high computational performance and low operational latency.