Cloud computing environments face persistent challenges in efficiently distributing workloads across dynamic and heterogeneous resources. Load balancing is a critical mechanism to optimize resource utilization, minimize latency, and ensure service reliability. This research proposes a hybrid load balancing model that merges static resource allocation with dynamic load redistribution to overcome limitations found in traditional single-strategy approaches. The hybrid model leverages static mapping for predictable baseline workloads while integrating dynamic decision-making based on real-time system metrics such as CPU load, memory usage, and response time. This dual-layer design allows the system to adapt intelligently to fluctuating workloads and enhances overall operational stability.
Performance evaluation was conducted using CloudSim, a simulation framework tailored to model and test cloud infrastructure. Comparative analysis against standard algorithms revealed significant improvements in resource utilization, task completion rate, and response time under varying load conditions. The proposed approach shows promise for scalable and adaptive load management in both public and hybrid cloud settings. Future work includes embedding machine learning components to refine real-time decisions and extending the model for deployment in multi-cloud and edge computing ecosystems.