Publicação
Automatic service migration in a MEC-NFV environment
| Resumo: | Network Functions Virtualization (NFV) and Multi-access Edge Computing (MEC) represent significant advancements in the telecommunications industry, addressing critical issues related to cost, complexity, and service agility. MEC, in particular, enables the efficient placement of computing resources at the network Edge to support low-latency applications. However, one of the key challenges in Edge environments, stemming from the inherent mobility of user devices, consists of ensuring seamless service migration to maintain optimal system performance and adapt to network conditions. This dissertation proposes several enhancements to the European Telecommunications Standards Institute (ETSI) MEC-NFV framework, specifically designed to enable automatic metricsbased service migration. The proposed system allows clients to define precise conditions under which their applications should be migrated, based on factors such as resource utilization and network performance metrics. By continuously monitoring application performance and system resources, the system triggers migration operations dynamically to ensure high Quality of Service (QoS) and adherence to Service-Level Agreements (SLA). To validate the proposed solution, a Proof of Concept (PoC) was developed, which integrated a Kubernetes cluster with the ETSI Open Source MANO (OSM). A Metrics Collector agent tracks realtime resource usage and network latency, feeding data into an intelligent MEC Application Orchestrator (MEAO) that automates the relocation of services based on predefined SLA parameters. This PoC, which demonstrated the feasibility of the proposed system, was made open-source, providing a foundation for further research and development in Edge Computing architectures. Furthermore, an experimental evaluation was conducted using the PoC. The obtained results confirmed the effectiveness of the proposed solution, showing that it reduces both the duration and variability of service migration times and enhances overall system responsiveness when compared to traditional approaches. |
|---|---|
| Autores principais: | Pereira, Pedro Costa Jorge Dias |
| Assunto: | Multi-access edge computing Network functions virtualization Management and orchestration Service migration |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Aveiro |
| Idioma: | inglês |
| Origem: | RIA - Repositório Institucional da Universidade de Aveiro |
| Resumo: | Network Functions Virtualization (NFV) and Multi-access Edge Computing (MEC) represent significant advancements in the telecommunications industry, addressing critical issues related to cost, complexity, and service agility. MEC, in particular, enables the efficient placement of computing resources at the network Edge to support low-latency applications. However, one of the key challenges in Edge environments, stemming from the inherent mobility of user devices, consists of ensuring seamless service migration to maintain optimal system performance and adapt to network conditions. This dissertation proposes several enhancements to the European Telecommunications Standards Institute (ETSI) MEC-NFV framework, specifically designed to enable automatic metricsbased service migration. The proposed system allows clients to define precise conditions under which their applications should be migrated, based on factors such as resource utilization and network performance metrics. By continuously monitoring application performance and system resources, the system triggers migration operations dynamically to ensure high Quality of Service (QoS) and adherence to Service-Level Agreements (SLA). To validate the proposed solution, a Proof of Concept (PoC) was developed, which integrated a Kubernetes cluster with the ETSI Open Source MANO (OSM). A Metrics Collector agent tracks realtime resource usage and network latency, feeding data into an intelligent MEC Application Orchestrator (MEAO) that automates the relocation of services based on predefined SLA parameters. This PoC, which demonstrated the feasibility of the proposed system, was made open-source, providing a foundation for further research and development in Edge Computing architectures. Furthermore, an experimental evaluation was conducted using the PoC. The obtained results confirmed the effectiveness of the proposed solution, showing that it reduces both the duration and variability of service migration times and enhances overall system responsiveness when compared to traditional approaches. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.