Abstract
dc:descriptionThe evolution of the fifth generation of networks (5G Networks) brought upon a set of new technology solutions that aim to address the increased demands of users (i.e., high data rates, ultra-low latency and reliability) as well as support diverse requirements of next-generation applications. Communication systems are going through radical transformations, incorporating emerging technologies, such as Network Function Virtualization (NFV), Software-Defined Networks (SDNs), and Cloud orchestration systems.At the same time, micro-data centers are being deployed to the corresponding unified environments, bringing virtualized services closer to the users (i.e., improving service performance and reliability, as well as reducing the communication latency). This new research area is defined as edge computing. However, legacy cloud deployments usually employ traditional virtual machines (VMs), which are resource-costly, face slow time for deployment or scaling up of virtual resources. Given the limited-resource availability as well as the dynamic characteristics of user demands or application requirements in edge clouds, the traditional virtual machines are not suitable in these environments. In this respect, the adoption of alternative lightweight virtualization approaches for edge cloud seek to fulfill such demands, given their particular features e.g. rapid manipulation of virtual resources. Thus, it is essential to seek new flexible orchestration and management solutions that utilize more efficiently the available resources i.e., compared to traditional cloud deployments. In other words, there is a need for systemic adaptations of 5G and Beyond ecosystems towards these goals.The main challenge of the dissertation is to propose novel flexible, adaptable and efficient edge-cloud orchestration solutions utilizing lightweight virtualization technologies, such as containers and Lightweight Virtual Machines (LVM). The latter technologies bring significant adaptability advantages to dynamic changes in network conditions and service requirements, along with fast responsiveness, being crucial for edge cloud deployments in 5G networks. This dissertation addresses crucial issues associated with 5G networks (i.e., elasticity, scalability, heterogeneity and resource efficiency), identifies the associated trade-offs of utilizing lightweight virtualization technologies, implements and investigates lightweight cloud orchestration platforms with orchestration mechanisms (e.g., efficient resource allocation and service elasticity), as well as proposes a new cloud orchestration strategy (i.e., Virtualization Technology Shifting) that takes into account the performance and resource demands of alternative virtualization technologies to address challenging application and resource requirements.
Degree
thesis:*- Grantor dc:publisher
- Πανεπιστήμιο Μακεδονίας
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Valsamas, Polychronis
- Βαλσαμάς, Πολυχρόνης
Subjects
dc:subject × 13- Δίκτυα πέμπτης γενιάς
- Εικονικές μηχανές ενός σκοπού
- Ενορχήστρωση
- 5G Network and Beyond
- Containers
- Unikernels
- Edge Cloud Orchestration
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Communication engineering and systems, Telecommunications
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/52015