On Open Platform for Multi-Access Edge Computing

Susani, Nahid (2018) On Open Platform for Multi-Access Edge Computing. [Laurea magistrale], Università di Bologna, Corso di Studio in Telecommunications engineering / ingegneria delle telecomunicazioni [LM - DM270], Documento full-text non disponibile
Il full-text non è disponibile per scelta dell'autore. (Contatta l'autore)

Abstract

The Idea of moving the functions of centralized cloud computing to the edge devices of the network, brought several advantages such as low latency, higher spectral efficiency, supporting machine to machine communication. The idea of decoupling network function from infrastructure well applied for this revolution. Therefor, the concept of software defined network (SDN) and network function virtualization considered as emerging solutions for the future network. With applying NFV in the edge devices, computing services and also using network functions would be possible through using virtual machines provided by NFV. In the first chapter, there are introductions of cloud, Edge cloud computing and SDN/NFV due to the importance of these topics. Containerization of application is an another proposed concept as lightweight virtualization solution which besides NFV can be proper solution for delivering, implementing and running application in the edge network and using advantages. In fact, developer encapsulate applications inside of the containers and containerization will allows to delivery applications across different environments. In second chapter, centralized cloud and Edge Cloud had been explained and it had been moved to introducing kubernetes as Edge cloud orchestrator in third chapter. In this project, Kubernetes has been used as an open source platform for multi access edge computing which works as container cluster orchestrator. Forth chapter has been dedicated to Implementation and Evaluation of the work. Main targets of this real implementation will be extending platform of container cluster from single platform architecture to multi-platform architecture made by different physical infrastructure of AMD64 on Intel and ARM on Raspberry pi using docker containers. The second target of implementation is checking the feasibility of autoscalling number of pods based on CPU utilization of pods, across nodes with heterogeneous platform inside the same cluster.

Abstract
Tipologia del documento
Tesi di laurea (Laurea magistrale)
Autore della tesi
Susani, Nahid
Relatore della tesi
Correlatore della tesi
Scuola
Corso di studio
Ordinamento Cds
DM270
Parole chiave
Cloud Computing,Edge-cloud,network virtualization,software define network,centralized cloud,edge cloud platform,container orchestrator,resource management,Kubernetes,pod autoscalling
Data di discussione della Tesi
16 Marzo 2018
URI

Altri metadati

Gestione del documento: Visualizza il documento

^