Resource Allocation and Scheduling in Fog–Cloud Computing: A Comprehensive Survey of Metaheuristic, Learning-Based, and QoS-Aware Approaches

Abstract

Fog cloud computing has become a vital paradigm for next-generation networks and the Internet of Things (IoT), bringing computation and storage closer to the edge. This integration reduces latency, improves energy efficiency, and enhances quality of service (QoS). However, effective resource allocation and task scheduling remain challenging due to heterogeneous networks, dynamic workloads, and conflicting objectives.

This study provides a comprehensive review of resource management strategies in fog cloud environments, focusing on heuristic, metaheuristic, and learning-based approaches. Key aspects such as computation offloading, load balancing, energy optimization, and cost efficiency are analyzed, with attention to their strengths, limitations, and areas of application.

The study highlights key open challenges such as large-scale real-world validation, the incorporation of security and privacy mechanisms, cross-layer optimization, and the design of hybrid models that integrate metaheuristics with reinforcement learning. Through a structured taxonomy and critical analysis, it offers valuable guidance for advancing adaptive, secure, and energy-efficient resource management in future fog cloud ecosystems.