Smart Load Balancing in Cloud Environments: A Multi-Objective Hybrid SAACO Approach

With the increased adoption of cloud computing, there is an impending need for effective load management techniques for better resource utilization and scaling up of the systems. Poor load balancing is prone to excessive use of resources, delayed response times, and even crashing of servers. This paper proposes a multi-objective hybrid load balancing model using simulated annealing (SA) and ant colony optimization (ACO) algorithms, SAACO, to enhance load balancing of scientific applications in clouds. The SAACO framework enhances the search ability of SA and the path finding ability of ACO to reduce the average waiting time, average response time, and job completion time, which are the key performance indicators. We present several experiments that position SAACO against the existing hybrid approaches such as Genetic Algorithm-Ant Colony Optimization (GA-ACO) and Firefly-Ant Colony Optimization (Firefly-ACO), demonstrating that SAACO outperforms these models on all metrics reported by the evaluation and is therefore a reliable strategy for ensuring load balance in a cloud environment. This revision summarizes the specific metrics that were compared and effectively compliments the proposed method's shortcomings.