Optimizing Task Load Distribution in Cloud Environments via Dynamic Genetic Algorithm

In this paper, we present Dynamic-Genetic Algorithm based load-balancing technique. Dynamic genetic algorithm performs selection, crossover and mutation multiple times for a given solution. This strategy uses a memory Improved Dynamic Genetic Algorithm based Load Balancing (DGA-LB) approach to get better load balanced solution. This strategy tries to balance the load at the same time tries to minimize the makespan of task set and reduces the response time. When compared to existing methods i.e. Honey bee behavior inspired load balancing(HBB-LB) algorithm, Weighted Round Robin (WRR) and FCFS algorithms, and the testing results show a significant improvement in average execution time, minimizing the make span of task sets and proper resource use.