Exploring Microservices as the Architecture of Choice for Network Function Virtualization Platforms
NFV is an emerging key technology that overcomes many challenges facing network service providers, such as reducing the capital and the operating expenses and satisfying the growing demand for mobile services. Integrating NFV with MEC and cloud environments requires an architecture that enables effi...
Gespeichert in:
Veröffentlicht in: | IEEE network 2019-03, Vol.33 (2), p.202-210 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | NFV is an emerging key technology that overcomes many challenges facing network service providers, such as reducing the capital and the operating expenses and satisfying the growing demand for mobile services. Integrating NFV with MEC and cloud environments requires an architecture that enables efficient implementations and deployments of NFV entities. Microservices architecture is a promising implementation of service-oriented architecture with recognized advantages in terms of modularity and continuous delivery. This article envisions microservices architecture as the solution of choice for building NFV platforms that are hosted in a dynamic environment ranging from MEC to cloud environments. This article addresses the major challenges and requirements of the microservices architecture to fully-exploit the potential of its adoption in NFV. It also proposes potential solutions that alleviate these issues. The article also discusses the need for agile and modular NFV entities along with MEC to realize various applications. To this end, the article discusses explicitly a novel NFV microservices entities scheduler optimization model. The proposed scheduler aims at minimizing network delays while taking into consideration various functional and non-functional constraints. The evaluation of the simulation results demonstrates that the proposed model minimizes the computational paths' latencies and improves the performance and availability of the NFV service chains. |
---|---|
ISSN: | 0890-8044 1558-156X |
DOI: | 10.1109/MNET.2019.1800023 |