An Enhanced Arithmetic Optimization Algorithm for Efficient Task Scheduling in Fog -Cloud Computing

Abstract

ABSTRACT

The growth of Internet of Things (IoT) applications has increased the demand for efficient task processing in cloud–fog computing environments. Traditional cloud computing often suffers from high latency, network congestion, and inefficient resource use when managing large, heterogeneous workloads. Fog computing mitigates these issues by bringing computational resources closer to end-users, enabling faster task execution and improved Quality of Service (QoS). However this research addresses the task scheduling problem in fog–cloud environments, focusing on optimal virtual machine (VM) allocation to minimize makespan and energy consumption. We propose an Enhanced Arithmetic Optimization Algorithm (EAOA) that integrates dynamic inertia weights, mutation coefficients, and a triangular mutation strategy to improve the exploration and exploitation capabilities of the standard Arithmetic Optimization Algorithm (AOA), additionally, a network model is introduced to assess and monitor various network performance metrics.. Simulation results using NASA iPSC workload traces show that EAOA consistently outperforms existing algorithms, achieving lower makespan, reduced energy consumption, and balanced resource utilization. These findings demonstrate that EAOA is an effective solution for dynamic and heterogeneous IoT-based systems, ensuring efficient task scheduling and optimal use of fog–cloud resources