Modified Genetic Algorithms (GA) for Load balancing in Cloud Computing

Abstract

The efficiency of the upcoming cloud computing generation will be determined by how rapidly the infrastructure is constructed and how dynamically the resources are utilized. To prevent any one resource from being overworked or underutilised, load balancing divides the fluctuating workload among several nodes. This is one of the primary issues with cloud computing. A skilled load balancer should adjust its strategy in accordance with shifting tasks and conditions. This can be viewed as an optimisation challenge. This research uses a Genetic Algorithm (GA) to suggest a novel load balancing approach. The technique seeks to shorten the length of a specific operation by distributing the load on the cloud infrastructure. The Cloud Analyst simulator has been used to model the suggested load balancing approach.  According to simulation findings for a typical example application, the suggested methodology performed better than the current approaches, such as Round Robing (RR), First Come First Serve (FCFS), and a local search algorithm called Stochastic Hill Climbing (SHC).