LOAD BALANCING IN CLOUD COMPUTING VIA MAYFLY OPTIMIZATION ALGORITHM

Abstract

Cloud computing is a new technology that enables users to store and retrieve data via the Internet on demand rather than using their hardware. Cloud computing comprises distinct data centers (servers) and clients (users). Load unbalancing is a multi-variant, multi-constraint issue that lowers the efficacy and performance of system resources. Therefore, a load scheduling technique is needed to distribute work among the right VMs and preserve the trade-off between them. To achieve better performance, this paper presents a novel mayfly optimization algorithm for load balancing (MFO-LB), which utilizes mayfly flight behavior and mating dynamics. The proposed technique balances the load in the cloud by managing the incoming loads by allocating resources according to user requests. The proposed work intends to increase performance by uniformly dividing the workload among the virtual machines, which will decrease utilization and reaction time. The proposed MFO-LB approach is beneficial for maintaining system stability, reducing response time (RT), and maximizing resource productivity in cloud environments. Finally, the effectiveness of the proposed technique is assessed by employing several metrics, including execution cost, RT, execution time, and makespan. The proposed method achieves up to 23.4 % low RT, a 24 %decrease in makespan, and a 31.5 % decrease in completion time, respectively.