A Secure and Containerized Swarm Intelligence Real-Time Energy-Conscious Framework for Automated Fog Nodes Using Hybrid Energy Optimization Algorithms

Abstract

Cloud computing has revolutionized the way IT infrastructures are implemented, as it is scalable and affordable; however, the rapid growth of cloud infrastructures has resulted in a substantial increase in power consumption of data centers, thus requiring optimization of energy efficiency as a major concern. This study proposes a novel adaptive energy optimization framework aimed at minimizing power consumption of distributed data centers of various cloud infrastructures using Kubernetes for containerization of workloads. The proposed framework utilizes the efficiency of containers to share the operating system kernel of a machine, thus minimizing virtualization overhead. The proposed framework incorporates optimization techniques include Ant Colony Optimization (ACO), Swarm Intelligence, and Dynamic Voltage and Frequency Scaling S (DVFS). The major modules that perform the task include Adaptive Resource Adjustment Algorithm (ARAA) and Automated Resource-Aware Container Lifecycle Management (ARCLM), which together perform the task of resource allocation and automatic termination of idle containers within the Kubernetes environment. The proposed framework also integrates fog computing with cloud computing, enabling the operation of containerized fog computing nodes with the same layer for resource management. The smart contracts operate in both layers, namely cloud and fog, for secure and automated resource allocation, creating it a viable and energy-efficient solution for cloud -fog computing.

Introduction

This study focuses on developing an energy-efficient hybrid cloud–fog computing framework that reduces energy consumption, CO? emissions, and SLA violations while maintaining high performance and resource utilization. The rapid growth of cloud data centers has significantly increased energy usage and environmental concerns, creating a need for sustainable computing solutions. The proposed framework combines containerization, Kubernetes orchestration, fog computing, swarm intelligence, Ant Colony Optimization (ACO), Dynamic Voltage and Frequency Scaling (DVFS), and smart contracts to optimize resource management and energy efficiency.

Background and Motivation

Cloud computing offers flexibility and efficient resource management but consumes large amounts of energy due to increasing demand for data processing and storage. Existing research highlights the importance of:

• Energy-efficient resource management.
• Container-based virtualization.
• Intelligent workload scheduling.
• Machine learning-driven cloud optimization.
• Sustainable cloud and edge computing practices.

The proposed solution aims to minimize energy consumption and carbon emissions while ensuring reliable service quality and efficient workload execution.

Literature Review

Previous studies have shown that:

• Container consolidation reduces energy consumption in cloud data centers.
• Intelligent resource management frameworks improve performance while lowering energy use and carbon emissions.
• Machine learning and predictive autoscaling enhance resource utilization.
• Energy-efficient VM placement and scheduling algorithms reduce power consumption and SLA violations.
• Swarm intelligence and optimization algorithms improve task allocation and workload distribution in dynamic cloud environments.

These findings form the foundation of the proposed energy-aware cloud-fog framework.

Proposed Methodology

The framework consists of four major stages:

1. Cloud Container Setup

• Services are deployed in lightweight containers managed by Kubernetes.
• Containers share the same operating system kernel, reducing overhead compared to virtual machines.
• Resources are allocated only when needed, preventing unnecessary energy consumption.

2. Fog Server Deployment

• Fog servers are placed near end users to provide low-latency processing.
• They perform local computation, storage, and networking.
• Only essential data is sent to the cloud, reducing communication overhead and energy usage.

3. Dynamic Resource Allocation

• Resources are allocated dynamically based on workload demand.
• Continuous monitoring of CPU, memory, storage, and network utilization is performed.
• Additional resources are allocated during high workloads and released during low workloads.

The framework uses:

• Swarm Intelligence
• Ant Colony Optimization (ACO)

These techniques select nodes with:

• Low power consumption
• Low latency
• High availability
• Stable performance

4. Fog Servers as Containerized Nodes

• Fog servers are integrated into the same Kubernetes orchestration layer as cloud servers.
• Cloud and fog resources are managed using common scheduling and scaling policies.
• Workloads can move seamlessly between cloud and fog environments.
• Smart contracts enforce policies related to resource allocation, scalability, and energy usage.

Energy Optimization Model

The framework calculates total energy consumption by considering:

• CPU power usage
• Memory power usage
• Network transmission power

It also applies Dynamic Voltage and Frequency Scaling (DVFS), which lowers CPU frequency and voltage during low workloads, significantly reducing processor power consumption.

Adaptive Resource Adjustment Algorithm (ARAA)

The ARAA continuously monitors:

• CPU utilization
• Memory usage
• Queue length
• SLA requirements
• Node energy profiles

Based on workload intensity, it automatically:

• Scales resources up during heavy demand.
• Scales resources down during light demand.

This adaptive approach prevents both overloading and underutilization of resources.

Results and Analysis

A prototype system was implemented using:

• JSP, Servlets, and Swing for interfaces.
• MySQL for workload and energy data storage.
• Kubernetes for container orchestration.

Performance was evaluated using:

• CPU utilization
• Execution time
• Power consumption
• Resource utilization efficiency

Key Results:

• DVFS significantly reduced unnecessary processor power usage.
• Average CPU utilization reached approximately 94%, reducing wasted energy.
• ACO-based workload scheduling achieved the best execution times.
• Swarm intelligence improved workload balancing and system responsiveness.
• Multi-fog deployment reduced network traffic and latency.
• The ARCLM approach identified idle containers and deactivated them to save additional energy.
• Smart-contract-based resource allocation further enhanced energy efficiency.

Conclusion

The research presented in this paper provides an innovative approach to energy-efficient resource management for hybrid cloud-fog computing environments through the implementation of virtual container technology and advanced resource management practices. The deployment of container orchestration using Kubernetes reduces the virtualization overhead of deploying containers to various environments and facilitates the horizontal scaling of applications across both cloud and fog resources. A key component of the framework is the use of optimization techniques, including Ant Colony Optimization, genetic algorithms, and Dynamic Voltage and Frequency Scaling to dynamically schedule workloads, save energy, and optimize the use of resources. The most significant innovation in this work was the integration of containerized applications deployed on fog servers with the orchestration layer of the cloud. This integration allows for centralized management of distributed resources while maximizing the utilization of those resources. To facilitate the efficient and flexible use of resources, two new algorithms have been developed: an Adaptive Resource Adjustment Algorithm (ARAA) and Automated Resource-Aware Containers Lifecycle Management (ARCLM). The ARAA dynamically adjusts the allocation of resources while the ARCLM prevents idle resources from being wasted. After conducting experiments with JSP, Servlets, Swing and MySQL to construct and evaluate how well this prototype was built, we have demonstrated that this system functions in real time, has the ability to scale and is a viable platform to deliver applications and provide services. The results of all experiments indicate that the current generation of our framework will decrease the overall amount of energy consumed while improving application performance. This would lead to the development of a strong foundation for the creation of flexible and energy-efficient Cloud-Fog systems, which would promote the use of green computing practices and would lead to sustainable development.

References

[1] A. Katal, S. Dahiya, and T. Choudhury, “Energy efficiency in cloud computing data centers: A survey on software technologies,” Cluster Computing, vol. 26, no. 3, pp. 1845–1875, Jun. 2023. [2] T. C. Hiremath and K. S. Rekha, “Energy efficient data migration concerning interoperability using optimized deep learning in container-based heterogeneous cloud computing,” Advances in Engineering Software, vol. 183, Art. no. 103496, Sep. 2023. [3] A. A. Khan, M. Zakarya, R. Buyya, R. Khan, M. Khan, and O. Rana, “An energy and performance aware consolidation technique for containerized datacenters,” IEEE Transactions on Cloud Computing, vol. 9, no. 4, pp. 1305–1322, Oct.–Dec. 2021. [4] B. Magotra, D. Malhotra, and A. K. Dogra, “Adaptive computational solutions to energy efficiency in cloud computing environment using VM consolidation,” Archives of Computational Methods in Engineering, vol. 30, no. 3, pp. 1789–1818, Apr. 2023. [5] Z. Zhong and R. Buyya, “A cost-efficient container orchestration strategy in Kubernetes-based cloud computing infrastructures with heterogeneous resources,” ACM Transactions on Internet Technology, vol. 20, no. 2, pp. 1–24, May 2020. [6] K. Ajmera and T. K. Tewari, “Energy-efficient virtual machine scheduling in IaaS cloud environment using energy-aware green-particle swarm optimization,” International Journal of Information Technology, vol. 15, no. 4, pp. 1927–1935, Apr. 2023. [7] R. Mandal, M. K. Mondal, S. Banerjee, G. Srivastava, W. Alnumay, U. Ghosh, and U. Biswas, “MECpVmS: An SLA-aware energy-efficient virtual machine selection policy for green cloud computing,” Cluster Computing, vol. 26, no. 1, pp. 651–665, Feb. 2023. [8] M. Waseem et al., “Containerization in multi-cloud environment: Roles, strategies, challenges, and solutions for effective implementation,” arXiv preprint arXiv:2403.12980, 2024. [9] N. Gholipour, E. Arianyan, and R. Buyya, “A novel energy-aware resource management technique using joint VM and container consolidation approach for green computing in cloud data centers,” Simulation Modelling Practice and Theory, vol. 104, Art. no. 102127, Nov. 2020. [10] N. Ashry, R. Attia, H. Nashaat, and R. Rizk, “CO2 emission mitigation in container-based cloud computing by the power of resource management,” in Proc. Int. Conf. Advanced Intelligent Systems and Informatics, 2023, pp. 97–111. [11] S. B. Nath, S. K. Addya, S. Chakraborty, and S. K. Ghosh, “Green containerized service consolidation in cloud,” in Proc. IEEE Int. Conf. Communications (ICC), Jun. 2020, pp. 1–6. [12] T. Gamage and I. Perera, “Optimizing energy efficient cloud architectures for edge computing: A comprehensive review,” International Journal of Advanced Computer Science and Applications, vol. 15, no. 11, pp. 637–645, 2024. [13] I. Hamzaoui, B. Duthil, V. Courboulay, and H. Medromi, “A survey on the current challenges of energy-efficient cloud resources management,” Social Network Analysis and Mining, vol. 1, no. 2, pp. 1–28, Mar. 2020. [14] A. Saboor et al., “Containerized microservices orchestration and provisioning in cloud computing: A conceptual framework and future perspectives,” Applied Sciences, vol. 12, no. 12, Art. no. 5793, Jun. 2022. [15] B. M. Beena et al., “A green cloud-based framework for energy-efficient task scheduling using carbon intensity data for heterogeneous cloud servers,” IEEE Access, vol. 13, pp. 73916–73938, 2025 [16] S. Tiwari and M. B. Beena, “A neighborhood inspired multiverse scheduler for energy and makespan optimized task scheduling for green cloud computing systems,” IEEE Access, vol. 12, pp. 157272–157298, 2024. [17] R. Buyya, S. Ilager, and P. Arroba, “Energy-efficiency and sustainability in new generation cloud computing: A vision and directions for integrated management of data centre resources and workloads,” Software: Practice and Experience, vol. 54, no. 1, pp. 24–38, Jan. 2024. [18] T. S. Reddy and M. B. Beena, “Analysis of green computing models on AWS using machine learning algorithms,” in Proc. 1st Int. Conf. Women in Computing (InCoWoCo), Nov. 2024, pp. 1–6. [19] I. Vasireddy, R. Wankar, and R. R. Chillarige, “Improving scalability, energy efficiency, and cost-effectiveness in Kubernetes clusters using a nonlinear regression-based predictive replica model and ORLE algorithm,” Egyptian Informatics Journal, vol. 31, Art. no. 100732, Sep. 2025. [20] S. Sunil and S. Patel, “Energy-efficient virtual machine placement algorithm based on power usage,” Computing, vol. 105, no. 7, pp. 1597–1621, Jul. 2023.

Copyright

Copyright © 2026 Preethi S, Sundar U, Chitra Devi P. 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.