Abstrakti
The maturing 5G network technology sees growing commercial deployments, with a shifting focus to service delivery. 5G networks, a common platform for diverse services, utilize network slicing for service isolation. Cloud-native services, composed of interdependent micro-services, are allocated to network slices spanning multiple areas, domains, and data centers. Due to mobility events caused by mobile end-users, slices with their assigned resources and services need to be re-scoped and re-provisioned. This requires slice mobility, which involves a slice moving between service areas. Slice mobility requires the inter-dependent service and resources to be migrated to reduce system overhead and to ensure low-communication latency by following end-user mobility patterns. Recent advances in computational hardware, Artificial Intelligence, and Machine Learning have attracted interest within the communication community, with increased research interest in self-managed network slices. However, migrating a service instance of a slice remains an open and challenging process given the needed coordination between inter-cloud resources, the dynamics, and the constraints of inter-data center networks. In this regard, this dissertation defines and enables smooth network slicing mobility patterns while maintaining both system and network resources stable. Specifically, we design, implement, and evaluate our proposed migration framework. Then, we design and define different network slice mobility patterns with their corresponding grouping methods and relevant mobility triggers. Next, we introduce various SFC migration strategies as an underlay technology enabler for network slice mobility patterns. After that, we propose an agent for automating the triggers selection process for enabling various network slice mobility patterns. Finally, we develop a network-aware agent capable of selecting accurate bandwidth values while ensuring fast and reliable service migration, thus enabling slice mobility while matching network and system requirements. In each section of this dissertation, the research results are evaluated and validated under different configurations in real-world settings or simulated environments. This dissertation provides recommendations for improving and extending the notion of mobility in network slices while also highlighting the various outstanding questions and suggesting future challenges and research directions.
Julkaisun otsikon käännös | Network Slice Mobility and Service Function Chain Migration across Multiple Administrative Cloud Domains |
---|---|
Alkuperäiskieli | Englanti |
Pätevyys | Tohtorintutkinto |
Myöntävä instituutio |
|
Valvoja/neuvonantaja |
|
Kustantaja | |
Painoksen ISBN | 978-952-64-1633-5 |
Sähköinen ISBN | 978-952-64-1634-2 |
Tila | Julkaistu - 2024 |
OKM-julkaisutyyppi | G5 Artikkeliväitöskirja |