Optimization of load balancing in cloud computing using the bin-stretching approach

Cloud computing has revolutionized how organizations and individuals utilize computational resources by offering scalable, flexible, and cost-effective solutions. However, one of the most pressing challenges within this domain is load balancing, which entails the optimal distribution of user tasks across virtual machines (VMs) to enhance efficiency and reduce latency. Traditional load-balancing algorithms, such as Min-Min and Max-Min, primarily focus on static allocation methods and are limited by their inability to adapt to dynamic workloads and varying resource availability. This paper introduces a novel Worst-Fit Based Load Balancing Algorithm (WFBLBA) to address these limitations by integrating memory management into the load-balancing process. WFBLBA models load balancing as a bucket-pulling problem and incorporates virtual memory considerations using a worst-fit heuristic approach. This innovative algorithm aims to minimize the Makespan and improve resource utilization by considering the available RAM of VMs and the impact of virtual memory delays on task processing. The research methodology includes comprehensive simulations conducted using the CloudSim Plus framework, which facilitates a detailed analysis of the proposed algorithm's performance under various cloud computing scenarios. Our results indicate that WFBLBA significantly enhances resource efficiency by up to 20% and reduces task processing delays by 15% compared to existing algorithms. This improvement is attributed to WFBLBA's ability to optimally allocate tasks based on real-time resource availability and the dynamic requirements of cloud-based services. Furthermore, this paper discusses the implications of our findings for cloud service providers and users.