LAUSR

In recent years, the rapid growth of IoT devices has led to an increase significantly the amount of data generated. Transferring a huge amount of datasets from IoT devices to remote cloud servers will result in high latency and bandwidth usage. Fog computing has emerged as an Internet-based distributed computing model to store datasets generated by IoT devices near the user. Since IoT devices generate continuously massive amounts of datasets, placing them on the storage fog nodes with various capabilities to reduce latency and costs of data access and increase reliability and availability of data datasets while satisfying the QoS requirements as one of the challenging tasks to be considered. This paper proposes a metaheuristic-based data replica placement mechanism using biogeography-based optimization (BBO) for data-intensive IoT applications on the fog ecosystem. Besides, we design an autonomous framework to illustrate transferring data replicas between IoT devices and storage fog nodes for data replica placement problem in the fog ecosystem. The obtained simulation results by varying the number of data replicas and fog nodes demonstrate that the proposed mechanism is a cost-effective solution and it increases the average reliability and availability by up 13% and 15% and reduces the total cost and the latency 25% and 3%, respectively, compared with the other baseline mechanisms.