Dynamic Load Balancing in Distributed Systems

Distributed computing has revolutionized the way modern systems handle scalability, fault tolerance, and resource optimization. With the rapid growth of online users and real-time services, efficiently managing request loads across multiple servers has emerged as a critical concern. Traditional load balancing strategies often fail to adapt to dynamic resource consumption, leading to system bottlenecks and uneven utilization. This study focuses on addressing the issue of distributing real-time workloads in distributed systems by deploying a dynamic load balancing framework. The proposed system, Dynamic Load Balancing in Distributed Systems in a publication, introduces a web-based simulation platform that compares the effectiveness of Round Robin and Least CPU load balancing algorithms. Developed using the Flask web framework, the platform simulates request routing to multiple backend servers while continuously monitoring resource metrics such as CPU utilization. Round Robin allocates requests sequentially, offering simplicity and uniformity, whereas Least CPU dynamically assigns requests based on the real-time processing load on each server, thereby optimizing resource allocation. Experimental evaluations using simulated high-traffic workloads demonstrate that while Round Robin maintains consistent performance under steady loads, the Least CPU algorithm significantly outperforms it in dynamic environments, delivering reduced response times and improved load distribution. These findings validate the importance of resource-aware balancing in distributed architectures and lay the groundwork for integrating more intelligent strategies. The simulation tool not only aids in educational demonstrations but also serves as a testbed for future research involving adaptive algorithms, AI-driven predictions, and orchestration with platforms like Kubernetes. It underscores the relevance of dynamic load balancing as a foundation for building scalable, resilient cloud infrastructures