Dynamic Resource Management of Green Fog Computing for IoT Support
Publication Date
1-1-2022
Document Type
Conference Proceeding
Publication Title
2022 International Conference on Green Energy, Computing and Sustainable Technology, GECOST 2022
DOI
10.1109/GECOST55694.2022.10010417
First Page
320
Last Page
326
Abstract
The internet of things (IoT) is an integrated part of contemporary life. It includes wearable devices, such as smartwatches and cell phones, and sensors for smart cities. Fog computing can improve the support of IoT devices by allowing these devices to offloading their collected data and other tasks to the fog nodes. To build a green fog computing system, it is important to have fog nodes near the IoT devices for faster data offload; this in turn would increase energy efficiency for both the IoT devices and the fog nodes. This paper considers dynamic, on-demand, peer-to-peer fog formation using volunteering nodes. We propose multiple modifications of the Memetic Algorithm (MA) to find the best service placement from IoT to fog nodes dynamically. optimized placements are calculated in terms of task completion delay and task completion rate. We further apply a local-search heuristics and introduce additional fitness functions to provide prioritized services. Together they improve energy efficiency, service quality, and system throughput. Finally, applying a machine learning method in the last step of the placement algorithm enables the system to find the best solution from a set of Pareto optimal results. We believe this is an important work that would contribute significantly to the research of emerging green computing systems.
Keywords
Fog Computing, Fog-Cloud Architecture, IoT, Machine Learning., Memetic Algorithm, Mobile Cloud Computing, Resource Management
Department
Computer Science
Recommended Citation
Melody Moh, Teng Sheng Moh, and Mariia Surmenok. "Dynamic Resource Management of Green Fog Computing for IoT Support" 2022 International Conference on Green Energy, Computing and Sustainable Technology, GECOST 2022 (2022): 320-326. https://doi.org/10.1109/GECOST55694.2022.10010417