An Intelligent Green Controller for Dynamic Resource Provisioning in Heterogeneous Cloud–Edge IoT Systems

The rapid growth of Internet of Things (IoT) applications has increased the demand for efficient, scalable, and energy-aware computing architectures capable of supporting latency-sensitive services. Traditional cloud-centric and edge-only models often suffer from high latency, resource underutilization, and energy inefficiencies in dynamic environments. To address these challenges, this paper proposes an intelligent energy-aware edge–cloud collaborative framework that jointly optimizes energy consumption, task latency, and resource utilization. The proposed framework integrates adaptive workload offloading, multi-objective scheduling, and real-time feedback control to dynamically allocate tasks across heterogeneous edge and cloud resources. A mathematical energy consumption model is developed to quantify computation and communication overheads, enabling informed scheduling decisions. The framework is evaluated using extensive simulations on CloudSim and iFogSim under diverse workload and network conditions. Experimental results show that the proposed approach consistently outperforms cloud-only, edge-only, and single-objective scheduling strategies across multiple performance metrics. It achieves up to 28% reduction in total energy consumption, 35% improvement in task latency, and improved load balancing and Quality of Service (QoS), demonstrating its effectiveness for scalable and energy-efficient edge–cloud computing environments.