Adaptive Load Balancing Strategies in Cloud Computing: A Survey

Paper ID : 1031-ICEEM2025 (R1)

Authors

Aya Allah Gamal Ebrahim *1, Gamal Mahrous Atia2, Ahmed Mostafa Elmahalawy2 113511 232951

Abstract

Load balancing plays a vital role in cloud computing by optimizing resource utilization alongside minimizing power use and upholding agreement performance standards. The advancement of cloud system efficiency depends on different load-balancing strategies such as adaptive threshold tuning and reinforcement learning-based migration and Gaussian model-based optimization and genetic algorithm-based strategies together with prediction-based proactive balancing. In this survey, we conduct a review of different approaches while pointing out their methods and strengths together with their drawbacks. Experimental investigations show advancements in cost management together with service level agreement fulfillment and energy system efficiency and workload balancing. The algorithms face several persistent issues because they generate excessive computing demands and operate using fixed threshold values and limited scalability in real-time scenarios and suboptimal handling of dynamic workloads. Future research should combine anticipated optimization and adaptive control approaches by using dynamic threshold because this integration will improve system performance while trimming down operational costs.

Keywords

Cloud Computing System, Virtualization, Load Balancing, Overloaded Host Detection, Virtual Machine Migration (VMM).

Status: Accepted