Multi-Objective Optimization for IoT Tasks based Data Replication Placement in Fog Computing

Abstract IoT applications have recently grown, quickly influencing cloud computing and Artificial Intelligence (AI) applications. Cloud computing has drawn more and more attention, particularly replication techniques and their uses. Cloud computing also provides different spaces according to usage from users and pay-for usage. It also provides flexibility in expanding or reducing the spaces according to use. Data on cloud computing is kept in several locations due to the data's growth and size. It's essential to shorten data transmission times between nodes, reduce bandwidth usage, lighten the pressure on the network, and balance the load across different regions. The expense of maintaining the system's data availability, performance, and reliability will rise as the number of replicas is increased and distributed across more locations. In this paper, we developed an Equilibrium Optimizer (EO) with Multi-Objective Optimization (MOO) to optimize a data replication placement based on IoT on fog computing. Second, we used a levy distribution to distribute a replica between nodes in cloud computing. Also, data transfer via cloud computing with a minimum bandwidth reduction. The performance of the proposed Multi-Objective Equilibrium Optimizer (MOEO) strategy was evaluated with the selection and placement of data replication using different criteria and sizes of data. To confirm the suggested algorithm's outcomes, comparisons were made between it and other algorithms. Experimental results demonstrated the superiority of the proposed algorithm over other algorithms in terms of addressing the problem of data replication placement, improving cost, and reducing data transmission time and distribution across geographically distributed sites.