Cloud computing has revolutionized modern computing by providing scalable, on-demand access to computing resources through a pay-as-you-use model, accessible from any location. As businesses increasingly migrate their operations to cloud platforms, cloud service providers continue to expand their data center infrastructures to support a growing number of concurrent users. However, one persistent challenge identified in the literature is the issue of uneven load distribution, which often leads to performance degradation, increased latency, and reduced resource availability. While several load-balancing techniques have been proposed to address this issue, many still suffer from inefficiencies, such as delayed redirection of loads from overloaded to underutilized servers, and the inability to meet key load-balancing metrics, including scalability, makespan, support for heterogeneous environments, and fault tolerance. To address these limitations, this study proposes a demand-responsive dynamic load-balancing scheme that optimizes cloud data center performance by incorporating real-time server metrics and modeling user request arrivals using a Poisson distribution. The scheme dynamically assigns incoming loads based on server capacity, current workload, and estimated request arrival rates, thereby reducing server overload and minimizing response delays. The simulation was implemented using load-balancing data extracted from the Google Cloud Jobs (GoCJ) dataset and evaluated using Python. Comparative performance analysis with existing techniques such as round-robin, least connection, and static threshold-based methods demonstrated the superiority of the proposed scheme. Quantitative results revealed that the proposed method achieved an average response time of 10–19 ms, compared to 18–30 ms for round-robin and 15–25 ms for least connection algorithms. Furthermore, the proposed scheme reduced the number of failed or delayed user requests by over 40%, improved server utilization balance, and maintained robust performance under high-load scenarios. These findings confirm the effectiveness of the proposed demand-based load-balancing strategy in enhancing the responsiveness, scalability, and efficiency of cloud data centers, thereby supporting the growing demands of cloud users and applications.
Hammed et al. (Tue,) studied this question.