A Systematic Investigation into Hybridized Meta-Heuristic Optimization and Reinforcement Learning Frameworks for Adaptive Load Balancing in Heterogeneous Cloud-Edge Environments

Kat Whitefield

Authors

Kat Whitefield Department of Computer Science and Applied Mathematics University of Melbourne Australia

Keywords:

Cloud Computing, Load Balancing, Meta-Heuristics, Reinforcement Learning

Abstract

The rapid proliferation of cloud computing services and the subsequent emergence of the edge-fog-cloud continuum have introduced unprecedented complexities in resource management and task distribution. Efficient load balancing remains a primary challenge, as the stochastic nature of task arrival and the inherent heterogeneity of virtualized resources can lead to significant performance bottlenecks, increased latency, and excessive energy consumption. This research article provides a comprehensive exploration of modern load balancing strategies, focusing specifically on the integration of hybrid meta-heuristic algorithms and deep reinforcement learning (DRL) techniques. By synthesizing contemporary advancements in parallel programming patterns, bio-inspired optimization, and intelligent feedback controllers, this study delineates a conceptual framework for achieving multi-objective optimization in dynamic environments. The analysis considers the transition from traditional centralized cloud architectures to multi-level parallel scheduling over edge-cloud infrastructures. Through a detailed theoretical elaboration of various algorithmic approaches, including Grey Wolf, Whale Optimization, and Markov process modeling, the article investigates how these methods mitigate the limitations of local optima and slow convergence. The findings suggest that hybridizing swarm intelligence with adaptive learning mechanisms significantly enhances the reliability and scalability of Infrastructure-as-a-Service (IaaS) platforms, particularly in high-stakes sectors such as healthcare and large-scale industrial flexible job shop scheduling.

References

Al Reshan, MS Syed, D Islam, N Shaikh, A Hamdi, M Elmagzoub, MA Talpur, KH (2023) A fast converging and globally optimized approach for load balancing in cloud computing. IEEE Access, 11, 11390–11404.

Asghari A. et al. (2023) Multi-objective edge server placement using the whale optimization algorithm and game theory. Soft Comput.

Galante G, da Rosa Righi R, de Andrade C (2024) Extending parallel programming patterns with adaptability features. Cluster Comput 1–22

Ghafir S, Alam MA, Siddiqui F, Naaz S (2024) Load balancing in cloud computing via intelligent PSO-based feedback controller. Sustainable Computing: Inf Syst 41:100948

Jangra A, Mangla N (2023) An efficient load balancing framework for deploying resource scheduling in cloud based communication in healthcare. Measurement: Sens 25:100584

Jena UK, Das PK, Kabat MR (2022) Hybridization of meta-heuristic algorithm for load balancing in cloud computing environment. J King Saud University-Computer Inform Sci 34(6):2332–2342

Kaur M, Kadam S, Hannoon N (2022) Multi-level parallel scheduling of dependent-tasks using graph-partitioning and hybrid approaches over edge-cloud. Soft Comput 26(11):5347–5362

Khan AR (2024) Dynamic load balancing in cloud computing: optimized RL-Based clustering with Multi-Objective optimized task scheduling. Processes 12(3):519

H. K. Krishnamurthy Sukumar, "A Novel Hybrid Grey Wolf Whale Optimization for Effectual Job Scheduling and Resource Distribution in Dynamic Cloud Computing," 2025 International Conference on Sustainability, Innovation & Technology (ICSIT), Nagpur, India, 2025, pp. 1-6, doi: 10.1109/ICSIT65336.2025.11293898.

Mishra K, Pati J, Majhi SK (2022) A dynamic load scheduling in IaaS cloud using binary JAYA algorithm. J King Saud University-Computer Inform Sci 34(8):4914–4930

Muthusamy A, Dhanaraj RK (2023) Dynamic Q-Learning‐Based optimized load balancing technique in cloud. Mob Inform Syst 2023(1):7250267

Pradhan A, Bisoy SK, Kautish S, Jasser MB, Mohamed AW (2022) Intelligent decision-making of load balancing using deep reinforcement learning and parallel PSO in cloud environment. IEEE Access 10:76939–76952

Ramezani Shahidani F, Ghasemi A, Toroghi Haghighat A, Keshavarzi A (2023) Task scheduling in edge-fog-cloud architecture: a multi-objective load balancing approach using reinforcement learning algorithm. Computing 105(6):1337–1359

Skackauskas J. et al. (2022) Dynamic impact for ant colony optimization algorithm. Swarm Evol. Comput.

Souravlas S, Anastasiadou SD, Tantalaki N, Katsavounis S (2022) A fair, dynamic load balanced task distribution strategy for heterogeneous cloud platforms based on Markov process modeling. IEEE Access 10:26149–26162

Thakur A. et al. (2022) RAFL: A hybrid metaheuristic based resource allocation framework for load balancing in cloud computing environment. Simul. Model. Pract. Theory

Wang J. et al. (2023) Load balancing for heterogeneous traffic in datacenter networks. J. Netw. Comput. Appl.

Wei P. et al. (2023) VMP-A3C: Virtual machines placement in cloud computing based on asynchronous advantage actor-critic algorithm. J. King Saud Univ. Comput. Inf. Sci.

Xie T. et al. (2022) Multi-objective optimization of data deployment and scheduling based on the minimum cost in geo-distributed cloud. Comput. Commun.

Yang D. et al. (2022) Dynamic opposite learning enhanced dragonfly algorithm for solving large-scale flexible job shop scheduling problem. Knowl.-Based Syst.