Abstract
Abstract
In this research, an advanced Large Energy-Aware Fog (LEAF) computing based technique was introduced that allows for the modeling of large-scale vehicular network scenarios for executing thousands of streaming applications on a distributed, heterogeneous infrastructure. Compute nodes can be interconnected with different types of wired or wireless networking protocols, and edge devices can be mobile and join or leave the topology during the simulation. This level of realism permits research on energy-conserving fog computing architectures leading to more informed decisions in the planning of future infrastructure. Furthermore, the proposed model enables online decision making based on power usage, which can be used to implement energy-aware task placement strategies or routing policies. These algorithms can make direct use of LEAF’s ability to trace the power usage of infrastructure back to the responsible applications in order to identify and mitigate potential inefficiencies. Moreover, different kinds of energy-saving mechanisms can be integrated into simulations. What further distinguishes LEAF from existing fog computing simulators is the combination of analytical and numerical modeling approaches. Instead of modeling network traffic in detail, all data flows, and power models are represented by parameterizable, mathematical equations. This method leads to results that are easy to analyze and ensures scalability to hundreds or thousands of devices and applications with percentage of distance between 1 to 1000 was covered up to 98.75% using the LEAF technique. The research collects findings of various papers on the energy usage of different compute and networking equipment, including a detailed derivation of WAN connection parameters, to provide the reader with examples on how to model and parameterize LEAF experiments. Furthermore, the evaluation results indicate that fog computing may indeed be able to conserve energy in the future, mainly by reducing WAN usage.
Publisher
Research Square Platform LLC
Reference25 articles.
1. Smart parking sensors, technologies and applications for open parking lots: A review;Paidi V;IET Intell Transp Syst,2018
2. Alsbou N, Afify M, Ali I (2019) Cloud-Based IoT Smart Parking System for Minimum Parking Delays on Campus. In International Conference on Internet of Things; Springer: Berlin/Heidelberg, Germany, ; pp. 131–139
3. A Cloud-Based Smart-Parking System Based on Internet-of-Things Technologies;Pham TN;IEEE Access,2015
4. An Efficient Named-Data-Networking-Based IoT Cloud Framework;Wang X;IEEE Internet Things J,2020
5. From smart parking towards autonomous valet parking: A survey, challenges and future Works;Khalid M;J Netw Comput Appl,2020