Software-Defined Resource Management for Industrial Internet of Things
Thesis event information
Date and time of the thesis defence
Place of the thesis defence
Linnanmaa, L10 Hall, remote connection: https://oulu.zoom.us/s/67032017708
Topic of the dissertation
Software-Defined Resource Management for Industrial Internet of Things
Doctoral candidate
Master of Science (Tech) Jude Okwuibe
Faculty and unit
University of Oulu Graduate School, Faculty of Information Technology and Electrical Engineering, CWC - Networks and Systems
Subject of study
Communications Engineering
Opponent
Professor Pekka Toivanen, University of Eastern Finland
Custos
Associate professor Mika Ylianttila, University of Oulu
Software-Defined Resource Management for Industrial Internet of Things
The Industrial Internet of Things (IIoT) and Industry 4.0 aim to streamline production processes and keep manufacturing viable and profitable. This presents enterprises with the opportunity to boost productivity while improving efficiency and safety and reducing costs. With heightened interest from both researchers and industry experts, IIoT has witnessed remarkable advances over the recent years thanks to developments in related technologies such as Industrial Wireless Networks (IWNs), Software-Defined Networking (SDN), cloud computing, and Multi-Access Edge Computing (MEC). Despite the proven ability of these technologies to advance the course of IIoT and Industry 4.0, an equally important but less investigated problem is ensuring that the resources upon which these technologies depend are optimally allocated and efficiently utilized.
This doctoral dissertation proposes a software-defined approach towards improving resource management and efficiency in IIoT systems. First, an SDN-based data offloading scheme is designed to coordinate data offloading for IIoT applications. This will enable constrained IIoT devices to relay their more demanding operations for energy and resource optimization. Second, a system model is developed to leverage the synergy between SDN, MEC, and containerization technologies in advancing IIoT applications for better resource management, more specifically for containerized edge microservices. Third, a novel SDN-enabled Resource Management (SDRM) scheme is developed based on Satisfiability Modulo Theory (SMT) constraint programming. With this scheme, SDRM will be able to automatically compute the optimal resource allocation for different IIoT network models and dynamically adjust assigned resources based on predefined constraints to ensure Service Level Agreements (SLAs). Lastly, the effects of collaborative edge-cloud computing for such SDN-based IIoT implementations are examined.
The results from our implementation models demonstrate the feasibility, efficiency, and performance improvements of utilizing SDN-based solutions for resource optimization in IIoT implementations. Hence, the outcome of this dissertation will help both researchers and system designers gravitate towards more resource-efficient IIoT solutions.
This doctoral dissertation proposes a software-defined approach towards improving resource management and efficiency in IIoT systems. First, an SDN-based data offloading scheme is designed to coordinate data offloading for IIoT applications. This will enable constrained IIoT devices to relay their more demanding operations for energy and resource optimization. Second, a system model is developed to leverage the synergy between SDN, MEC, and containerization technologies in advancing IIoT applications for better resource management, more specifically for containerized edge microservices. Third, a novel SDN-enabled Resource Management (SDRM) scheme is developed based on Satisfiability Modulo Theory (SMT) constraint programming. With this scheme, SDRM will be able to automatically compute the optimal resource allocation for different IIoT network models and dynamically adjust assigned resources based on predefined constraints to ensure Service Level Agreements (SLAs). Lastly, the effects of collaborative edge-cloud computing for such SDN-based IIoT implementations are examined.
The results from our implementation models demonstrate the feasibility, efficiency, and performance improvements of utilizing SDN-based solutions for resource optimization in IIoT implementations. Hence, the outcome of this dissertation will help both researchers and system designers gravitate towards more resource-efficient IIoT solutions.
Last updated: 1.3.2023