Dynamic Load Balancing with Task Migration: A Genetic Algorithm Approach for Optimizing Cloud Computing Infrastructure

The advancement of cloud computing is contingent upon the effective establishment of infrastructure and the flexible use of accessible resources. An essential aspect of this evolutionary process is load balancing, a critical task involving fair allocation of dynamic workloads among different nodes. The primary aim is to mitigate the issues of resource overload or underutilization issues, which presents a complex challenge in optimization. A proficient load balancer must be able to adapt its method to the dynamic environment and work demands. This research presents a novel load-balancing technique incorporating task migration using a Genetic Algorithm (GA). The objective is to achieve optimal load distribution within the cloud infrastructure while minimizing the makespan for a given set of tasks. By utilizing the inherent optimization capabilities of genetic algorithms and the flexibility of task migration, the suggested approach demonstrates dynamic adaptation to the constantly evolving demands of cloud computing. In order to evaluate the effectiveness of the plan, it is implemented and assessed using simulations carried out on the Cloud Analyst simulator. The findings from the simulation outcomes, as demonstrated in a representative application scenario, provide evidence for the effectiveness of the task migration-integrated genetic algorithm-based approach compared to existing methods such as First Come, First Serve (FCFS), Round Robin (RR), and the local search algorithm Stochastic Hill Climbing (SHC). This comprehensive method not only improves resource utilization efficiency but also showcases the capacity to mitigate performance limitations through intelligent task reallocation. The proposed technique, which incorporates task migration capabilities, shows potential in facilitating the development of a more efficient and adaptable cloud infrastructure as the cloud landscape undergoes ongoing changes.