Resource Allocation and Load Balancing for Fog Computing Using Artificial Intelligence Techniques

Abstract

Fog computing (FC) has emerged as a critical paradigm for enabling low-latency and efficient data processing in Internet of Things (IoT) environments. However, resource allocation and load balancing remain significant challenges due to the heterogeneous and dynamic nature of FC networks. This review systematically analyses recent advancements in artificial intelligence (AI)-driven approaches for optimizing resource allocation and load balancing in FC, with a focus on studies published between 2019 and 2024. The survey highlights the role of machine learning (ML), deep learning (DL), and reinforcement learning (RL) in intelligent resource allocation, ensuring efficient task offloading and execution. Additionally, it explores heuristic and nature-inspired meta-heuristic algorithms for dynamic load balancing, improving system throughput, energy efficiency, and Quality of Service (QoS). While these techniques have demonstrated significant improvements in FC performance, challenges such as real-world implementation complexity, scalability, and system heterogeneity persist. The review identifies future research directions, emphasizing the need for advanced AI-driven frameworks and deep reinforcement learning techniques to enhance resource management and load balancing in distributed FC environments.

Introduction

1. Background & Motivation:
Fog Computing (FC) has become essential for low-latency, efficient data processing in IoT environments, acting as a bridge between cloud data centers and edge devices. It addresses challenges of conventional cloud systems—like high latency, network congestion, and inefficiency—especially for real-time, latency-sensitive applications such as healthcare, autonomous vehicles, and smart cities.

2. Key Challenges in FC:

• Heterogeneous Infrastructure: Devices vary in computational capacity, energy, and connectivity.

• Dynamic Workloads: IoT devices generate unpredictable data loads.

• Latency Sensitivity: Applications require real-time responses.

• Energy Constraints: Many fog nodes are battery-powered.

• Scalability: The rapid growth of IoT devices requires adaptable systems.

• Security & Privacy: Sensitive data must be handled locally and securely.

3. Role of AI in FC:
To meet these challenges, AI-driven methods—including Machine Learning (ML), Deep Learning (DL), and Reinforcement Learning (RL)—are used to optimize:

• Resource Allocation: Efficient assignment of memory, CPU, and storage to fog nodes.

• Load Balancing: Distributing workloads evenly to avoid bottlenecks and improve QoS.

4. AI Techniques for Resource Allocation:

• ML-Based Task Scheduling: Predictive models allocate resources based on past workload patterns.

• DL for Real-Time Management: Neural networks analyze live data to prevent congestion.

• RL for Adaptive Allocation: Self-learning agents optimize resource distribution over time.

• Heuristic & Meta-Heuristic Methods: Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO) find near-optimal solutions efficiently.

5. AI Techniques for Load Balancing:

• Static vs. Dynamic Approaches: AI enables real-time, adaptive load distribution.

• Meta-Heuristics (GA, ACO): Reduce response time and node overload.

• RL Models: Continuously learn and improve workload strategies.

• Software-Defined Networking (SDN): AI enhances SDN-based routing and load control across fog networks.

Conclusion

Fog computing is involved into a large amount of IOT enabled system. This Provides load sharing on cloud, computing resources and enhanced system throughput. This paper investigates rapidly growing field of Fog computing and its impactful contribution to overcome the limitations of conventional cloud based architecture, especially for latency sensitive IoT applications. with the evolution of Fog, processing of data is possible at the network edge and thus improves responsiveness and efficiency. This study investigates core research areas such as resource management, load balancing, task scheduling, and security. Integration of AI based techniques in Fog Computing issues improves decision-making and resource optimization. This work presents an analysis of essential areas with respect of their effectiveness in enhancing essential performance matrix such as energy efficiency, latency, and quality of services (QoS). Comparison tables are provided to highlight scope, techniques, strength and weaknesses to deliver valuable insight for researchers.

References

[1] Z. Ma, M. Xiao, Y. Xiao, Z. Pang, H. V. Poor, and B. Vucetic, ‘‘High-reliability and low-latency wireless communication for Internet of Things: Challenges, fundamentals, and enabling technologies,’’ IEEE Internet Things J., vol. 6, no. 5, pp. 7946–7970, Oct. 2019. [2] K. Lone and S. A. Sofi, ‘‘A review on offloading in fog-based Internet of Things: Architecture, machine learning approaches, and open issues,’’ High-Confidence Comput., vol. 3, no. 2, Jun. 2023, Art. no. 100124. [3] H.K.Apat,R.Nayak,andB.Sahoo,‘‘Acomprehensivereview onInternet of Things application placement in fog computing environment,’’ Internet Things, vol. 23, Oct. 2023, Art. no. 100866. [4] G. K. Walia, M. Kumar, and S. S. Gill, ‘‘AI-empowered fog/edge resource management for IoT applications: A comprehensive review, research challenges, and future perspectives,’’ IEEE Commun. Surveys Tuts., vol. 26, no. 1, pp. 619–669, 1st Quart., 2024. [5] M. Aqib, D. Kumar, and S. Tripathi, ‘‘Machine learning for fog computing: Review, opportunities and a fog application classifier and scheduler,’’ Wireless Pers. Commun., vol. 129, no. 2, pp. 853–880, Mar. 2023. [6] S. Askar, ‘‘Deep learning and fog computing: A review,’’ Available SSRN, 2021. [7] Z. Ma, M. Xiao, Y. Xiao, Z. Pang, H. V. Poor, and B. Vucetic, ‘‘High-reliability and low-latency wireless communication for Internet of Things: Challenges, fundamentals, and enabling technologies,’’ IEEE Internet Things J., vol. 6, no. 5, pp. 7946–7970, Oct. 2019 [8] K. H. Abdulkareem, M. A. Mohammed, S. S. Gunasekaran, M. N. Al-Mhiqani, A. A. Mutlag, S. A. Mostafa, N. S. Ali, and D. A. Ibrahim, ‘‘A review of fog computing and machine learning: Concepts, applications, challenges, and open issues,’’ IEEE Access, vol. 7, pp. 153123–153140, 2019. [9] G. Kumar and A. Altalbe, ‘‘Artificial intelligence (AI) advancements for transportation security: In-depth insights into electric and aerial vehicle systems,’’ Environ., Develop. Sustainability, pp. 1–51, Apr. 2024. [10] H. Djigal, J. Xu, L. Liu, and Y. Zhang, ‘‘Machine and deep learning for resource allocation in multi-access edge computing: A survey,’’ IEEE Commun. Surveys Tuts., vol. 24, no. 4, pp. 2449–2494, 4th Quart., 2022. [11] H. Tran-Dang, S. Bhardwaj, T. Rahim, A. Musaddiq, and D.-S. Kim, ‘‘Reinforcement learning based resource management for fog computing environment: Literature review, challenges, and open issues,’’ J. Commun. Netw., vol. 24, no. 1, pp. 83–98, Feb. 2022. [12] V. Kashyap, R. Ahuja, and A. Kumar, ‘‘Nature inspired meta-heuristic algorithms based load balancing in fog computing environment,’’ in Proc. 7th Int. Conf. Parallel, Distrib. Grid Comput. (PDGC), Nov. 2022, pp. 396–401. [13] A. U. Rehman et al., “Dynamic energy efficient resource allocation strategy for load balancing in fog environment,” IEEE Access, vol. 8, 2020, doi: 10.1109/ACCESS.2020.3035181. [14] F. M. Talaat, M. S. Saraya, A. I. Saleh, H. A. Ali, and S. H. Ali, “A load balancing and optimization strategy (LBOS) using reinforcement learning in fog computing environment,” J Ambient Intell Humaniz Comput, vol. 11, no. 11, 2020, doi: 10.1007/s12652-020-01768-8. [15] D. Baburao, T. Pavankumar, and C. S. R. Prabhu, “Load balancing in the fog nodes using particle swarm optimization-based enhanced dynamic resource allocation method,” Applied Nanoscience (Switzerland), vol. 13, no. 2. 2023. doi: 10.1007/s13204-021-01970-w. [16] A. R. Hameed, S. ul Islam, I. Ahmad, and K. Munir, “Energy- and performance-aware load-balancing in vehicular fog computing,” Sustainable Computing: Informatics and Systems, vol. 30, 2021, doi: 10.1016/j.suscom.2020.100454. [17] M. Ebrahim and A. Hafid, “Resilience and load balancing in Fog networks: A Multi-Criteria Decision Analysis approach,” Microprocess Microsyst, vol. 101, 2023, doi: 10.1016/j.micpro.2023.104893. [18] L.Sun, M.Liu,J. Guo, X.Yu,andS.Wang,‘‘Deep reinforcement learning empowered resource allocation in vehicular fog computing,’’ IEEE Trans. Veh. Technol., vol. 73, no. 5, pp. 7066–7076, May 2024. [19] D. H. Abdulazeez and S. K. Askar, ‘‘Offloading mechanisms based on reinforcement learning and deep learning algorithms in the fog computing environment,’’ IEEE Access, vol. 11, pp. 12555–12586, 2023. [20] D.R.Patil, ‘‘Dynamic resource allocation and memory management using machinelearning for cloud environments,’’ Int. J. Adv. Trends Comput. Sci. Eng., vol. 9, no. 4, pp. 5921–5927,. Aug. 2020 [21] A. Lakhan, M. A. Mohammed, O. I. Obaid, C. Chakraborty, K. H. Abdulkareem, and S. Kadry, ‘‘Efficient deep-reinforcement learning aware resource allocation in SDN-enabled fog paradigm,’’ Automated Softw. Eng., vol. 29, no. 1, pp. 1–25, May 2022. [22] G. Shruthi, M. R. Mundada, B. J. Sowmya, and S. Supreeth, ‘‘Mayfly Taylor optimisation-based scheduling algorithm with deep reinforcement learning for dynamic scheduling in fog-cloud computing,’’ Appl. Comput. Intell. Soft Comput., vol. 2022, pp. 1–17, Aug. 2022. [23] M. Kumar, A. Kishor, J. K. Samariya, and A. Y. Zomaya, ‘‘An autonomic workload prediction and resource allocation framework for fog-enabled industrial IoT,’’ IEEE Internet Things J., vol. 10, no. 11, pp. 9513–9522, Jun. 2023. [24] S. Mishra, M. N. Sahoo, S. Bakshi, and J. J. P. C. Rodrigues, ‘‘Dynamic resource allocation in fog-cloud hybrid systems using multicriteria AHP techniques,’’ IEEE Internet Things J., vol. 7, no. 9, pp. 8993–9000, Sep. 2020. [25] T. Tsokov and H. Kostadinov, ‘‘Dynamic network-aware container allocation in cloud/fog computing with mobile nodes,’’ Internet Things, vol. 26, Jul. 2024, Art. no. 101211. [26] F. M. Talaat, S. H. Ali, A. I. Saleh, and H. A. Ali, “Effective Load Balancing Strategy (ELBS) for Real-Time Fog Computing Environment Using Fuzzy and Probabilistic Neural Networks,” Journal of Network and Systems Management, vol. 27, no. 4, 2019, doi: 10.1007/s10922-019-09490-3. [27] I. Z. Yakubu and M. Murali, ‘‘An efficient meta-heuristic resource allocation with load balancing in IoT-Fog-cloud computing environment,’’ J. Ambient Intell. Humanized Comput., vol. 14, no. 3, pp. 2981–2992, Mar. 2023. [28] P. Singh et al., “A Fog-Cluster Based Load-Balancing Technique,” Sustainability (Switzerland), vol. 14, no. 13, 2022, doi: 10.3390/su14137961. [29] F. Banaie, M. H. Yaghmaee, S. A. Hosseini, and F. Tashtarian, “Load-Balancing Algorithm for Multiple Gateways in Fog-Based Internet of Things,” IEEE Internet Things J, vol. 7, no. 8, 2020, doi: 10.1109/JIOT.2020.2982305. [30] S. S. Karthik and A. Kavithamani, “Fog computing-based deep learning model for optimization of microgrid connected WSN with load balancing,” Wireless Networks, vol. 27, no. 4, 2021, doi: 10.1007/s11276-021-02613-2. [31] S. P. Singh, A. Sharma, and R. Kumar, “Design and exploration of load balancers for fog computing using fuzzy logic,” Simul Model Pract Theory, vol. 101, 2020, doi: 10.1016/j.simpat.2019.102017. [32] R. Beraldi, C. Canali, R. Lancellotti, and G. P. Mattia, “Randomized Load Balancing under Loosely Correlated State Information in Fog Computing,” in MSWiM 2020 - Proceedings of the 23rd International ACM Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems, 2020. doi: 10.1145/3416010.3423244. [33] K. Cui, W. Sun, B. Lin, and W. Sun, “Load balancing mechanisms of unmanned surface vehicle cluster based on marine vehicular fog computing,” in Proceedings - 2020 16th International Conference on Mobility, Sensing and Networking, MSN 2020, Institute of Electrical and Electronics Engineers Inc., Dec. 2020, pp. 797–802. doi: 10.1109/MSN50589.2020.00136. [34] Q. Fan and N. Ansari, “Towards Workload Balancing in Fog Computing Empowered IoT,” IEEE Trans Netw Sci Eng, vol. 7, no. 1, 2020, doi: 10.1109/TNSE.2018.2852762. [35] F. Alqahtani, M. Amoon, and A. A. Nasr, “Reliable scheduling and load balancing for requests in cloud-fog computing,” Peer Peer Netw Appl, vol. 14, no. 4, 2021, doi: 10.1007/s12083-021-01125-2. [36] B. Alamri, M. A. Hossain, and M. S. Jamil Asghar, “Electric power network interconnection: A review on current status, future prospects and research direction,” Electronics (Switzerland), vol. 10, no. 17, pp. 1–29, 2021, doi: 10.3390/electronics10172179. [37] A. Lakhan, M. A. Mohammed, O. I. Obaid, C. Chakraborty, K. H. Abdulkareem, and S. Kadry, ‘‘Efficient deep-reinforcement learning aware resource allocation in SDN-enabled fog paradigm,’’ Automated Softw. Eng., vol. 29, no. 1, pp. 1–25, May 2022 [38] F. M. Talaat, S. H. Ali, A. I. Saleh, and H. A. Ali, “Effective Load Balancing Strategy (ELBS) for Real-Time Fog Computing Environment Using Fuzzy and Probabilistic Neural Networks,” Journal of Network and Systems Management, vol. 27, no. 4, 2019, doi: 10.1007/s10922-019-09490-3. [39] J. Yan, J. Wu, Y. Wu, L. Chen, and S. Liu, “Task Offloading Algorithms for Novel Load Balancing in Homogeneous Fog Network,” in Proceedings of the 2021 IEEE 24th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2021, 2021. doi: 10.1109/CSCWD49262.2021.9437748. [40] L.Sun, M.Liu,J. Guo, X.Yu,andS.Wang,‘‘Deep reinforcement learning empowered resource allocation in vehicular fog computing,’’ IEEE Trans. Veh. Technol., vol. 73, no. 5, pp. 7066–7076, May 2024. [41] D. H. Abdulazeez and S. K. Askar, ‘‘Offloading mechanisms based on reinforcement learning and deep learning algorithms in the fog computing environment,’’ IEEE Access, vol. 11, pp. 12555–12586, 2023. [42] D.R.Patil, ‘‘Dynamic resource allocation and memory management using machinelearning for cloud environments,’’ Int. J. Adv. Trends Comput. Sci. Eng., vol. 9, no. 4, pp. 5921–5927,. Aug. 2020 [43] A. Lakhan, M. A. Mohammed, O. I. Obaid, C. Chakraborty, K. H. Abdulkareem, and S. Kadry, ‘‘Efficient deep-reinforcement learning aware resource allocation in SDN-enabled fog paradigm,’’ Automated Softw. Eng., vol. 29, no. 1, pp. 1–25, May 2022. [44] G. Shruthi, M. R. Mundada, B. J. Sowmya, and S. Supreeth, ‘‘Mayfly Taylor optimisation-based scheduling algorithm with deep reinforcement learning for dynamic scheduling in fog-cloud computing,’’ Appl. Comput. Intell. Soft Comput., vol. 2022, pp. 1–17, Aug. 2022. [45] M. Kumar, A. Kishor, J. K. Samariya, and A. Y. Zomaya, ‘‘An autonomic workload prediction and resource allocation framework for fog-enabled industrial IoT,’’ IEEE Internet Things J., vol. 10, no. 11, pp. 9513–9522, Jun. 2023. [46] S. Mishra, M. N. Sahoo, S. Bakshi, and J. J. P. C. Rodrigues, ‘‘Dynamic resource allocation in fog-cloud hybrid systems using multicriteria AHP techniques,’’ IEEE Internet Things J., vol. 7, no. 9, pp. 8993–9000, Sep. 2020 [47] T. Tsokov and H. Kostadinov, ‘‘Dynamic network-aware container allocation in cloud/fog computing with mobile nodes,’’ Internet Things, vol. 26, Jul. 2024, Art. no. 101211. [48] M. Peng, Z. Zhao, Y. Sun, M. Peng, Z. Zhao, and Y. Sun, ‘‘Dynamic resource allocation in fog radio access networks,’’ in Fog Radio Access Networks (F-RAN), 2020, pp. 105–131. [49] I. Z. Yakubu and M. Murali, ‘‘An efficient meta-heuristic resource allocation with load balancing in IoT-Fog-cloud computing environment,’’ J. Ambient Intell. Humanized Comput., vol. 14, no. 3, pp. 2981–2992, Mar. 2023. [50] H.H.Esmatand B.Lorenzo, ‘‘Deep reinforcement learning based dynamic edge/fog network slicing,’’ in Proc. GLOBECOM IEEE Global Commun. Conf., Dec. 2020, pp. 1–6. [51] S. D. A. Javaheri, R. Ghaemi, and H. M. Naeen, ‘‘An autonomous architecture based on reinforcement deep neural network for resource allocation in cloud computing,’’ Computing, vol. 106, no. 2, pp. 371–403, Feb. 2024. [52] D. Baburao, T. Pavankumar, and C. S. R. Prabhu, ‘‘Load balancing in the fog nodes using particle swarm optimization-based enhanced dynamic resource allocation method,’’ Appl. Nanosci., vol. 13, no. 2, pp. 1045–1054, Feb. 2023. [53] H. Liang, X. Zhang, X. Hong, Z. Zhang, M. Li, G. Hu, and F. Hou, ‘‘Reinforcement learning enabled dynamic resource allocation in the Internet of Vehicles,’’ IEEE Trans. Ind. Informat., vol. 17, no. 7, pp. 4957–4967, Jul. 2021. [54] S. Bhandari, H. Kim, N. Ranjan, H. P. Zhao, and P. Khan, ‘‘Optimal cache resource allocation based on deep neural networks for fog radio access networks,’’ J. Internet Technol, vol. 21, pp. 967–975, 2020. [55] S. Khan, I. A. Shah, M. F. Nadeem, S. Jan, T. Whangbo, and S. Ahmad, ‘‘Optimal resource allocation and task scheduling in fog computing for Internet of Medical Things applications,’’ Tech. Rep., 2023.

Copyright

Copyright © 2025 Sadhna Yadav, Dr. Deepak Kumar Verma. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.