Multi objective Internet of Things (IoT) services placement in fog computing using WOA-MLP optimization (whale optimization algorithm and multi perceptron neural network)

With the continuous advancement of Internet of Things (IoT) applications, their widespread integration into various aspects of daily life has significantly enhanced the quality of human existence. Fog computing, as a distributed computational solution, has emerged to serve these applications by leveraging fog nodes proximate to IoT devices. IoT applications have evolved into multiple services with diverse Quality of Service (QoS) requirements that can be deployed on fog nodes. Consequently, devising an efficient service orchestration plan to harness the capabilities of various resources within the fog ecosystem is a challenging issue that must be addressed. In this research, an efficient service orchestration solution based on Multilayer Perceptron (MLP) neural network and Whale Optimization Algorithm (WOA) is proposed for deploying IoT applications on the fog infrastructure. The proposed solution monitors the QoS requirements of IoT services and the capabilities of available fog nodes to determine an efficient service. The orchestration plan is initially estimated using the MLP model and refined using a metaheuristic optimization algorithm. In this approach, operational power, energy consumption, and delay are employed as objective functions to find the desired service orchestration plan. Simulation results demonstrate that the proposed solution enhances resource utilization and service acceptance ratio while reducing service delay and energy consumption compared to other metaheuristic-based mechanisms.