The evolution of multi-tier cloud-edge computing systems has been driven by the proliferation of Internet of Things (IoT) devices and the urgent need for real-time data processing. In these systems, the low latency benefits of edge nodes are combined with centralized cloud servers to achieve a solution that addresses scalability, resource consumption, and network performance problems. In this paper, we explore state-of-the-art load balancing strategies for such multi-tier systems that involve dynamic task offloading, decentralized resource management, and collaboration of cloud and edge infrastructures. Then we study how to ensure that performance is optimal under varying workloads by considering key approaches such as energy efficient algorithms, UAV-assisted edge computing, and hybrid deployment strategies. Using a review of recent research, we demonstrate the benefits and challenges to applying these strategies in IoT, UAV, and mobile edge environments. Additionally, the paper draws some useful conclusions on possible directions for designing adaptive, scalable, and sustainable load balancing solutions for cloud-edge systems, providing a road map for future improvements in this active area of research.