Dynamic Resource Orchestration for Service Capability Maximization in Fog-Enabled Connected Vehicle Networks
- Authors
- Vu, Duc-Nghia; Dao, Nhu-Ngoc; Na, Woongsoo; Cho, Sungrae
- Issue Date
- Jul-2022
- Publisher
- IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- Keywords
- Fog enabled connected vehicle networks; resource orchestration; service capability; matching algorithm
- Citation
- IEEE TRANSACTIONS ON CLOUD COMPUTING, v.10, no.3, pp 1726 - 1737
- Pages
- 12
- Journal Title
- IEEE TRANSACTIONS ON CLOUD COMPUTING
- Volume
- 10
- Number
- 3
- Start Page
- 1726
- End Page
- 1737
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/58769
- DOI
- 10.1109/TCC.2020.3001289
- ISSN
- 2168-7161
- Abstract
- Technological advances in fog computing are precipitating an evolution in conventional vehicle networks to a new paradigm called fog-enabled connected vehicle networks (FCVNs). FCVNs provide communication efficiency for ensuring safe transportation through the massive Internet of vehicles. In FCVNs, massive vehicles tend to associate with roadside units and high power nodes, which act as fog nodes (FNs), when they have a good channel quality and/or popular contents. This circumstance may lead to a load imbalance among the FNs. This problem significantly decreases the resource utilization efficiency and service capability of the networks. In this article, we propose a dynamic resource orchestration (DRO) scheme to harmonize resource allocation for connected vehicles by migrating the offloaded services among FNs. A graph-theoretic approach is utilized to transform the FCVN into a directed graph model, where the maximum resource reduction obtained by service migrations is considered the weight of the link between every two FNs. Subsequently, the maximum weight matching solution is used to determine optimal pairs of FNs for migrating services to maximize network resource utilization. Our simulation results reveal that the proposed DRO scheme achieves significant improvements in terms of service capability, throughput, and resource utilization efficiency as compared with existing algorithms.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Software > School of Computer Science and Engineering > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/58769)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.