Title:	FRAMEWORK FOR MODEL-DRIVEN SEMANTIC-ENABLED DEPLOYMENT OF CONTAINER-BASED VIRTUAL NETWORK FUNCTIONS TO SUPPORT EDGE COMPUTING AND FUTURE INTERNET SERVICES
Author(s):	Nenad Petrovic
ISBN:	978-989-8533-82-1
Editors:	Pedro Isaías and Hans Weghorn
Year:	2018
Edition:	Single
Keywords:	VNF, SDN, Edge Computing, Container-Based Virtualization, Model-Driven Engineering, Semantic Analysis
Type:	Full Paper
First Page:	90
Last Page:	100
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Virtualization of computational hardware resource has been around for years and it has proved itself as an effective solution, enabling more flexible resource management which supports Cloud computing. Beside the traditional hypervisor-based virtualization, another concept co-exists – container-based virtualization. In container-based virtualization, the virtualization layer runs as an application within the operating system, which is much lighter compared to hypervisor. However, in recent times, there is also tendency to apply a virtualization approach of network resources, due to growth of quantity of data that has to be transmitted and increasing number of mobile and wearable devices, which requires more robust and flexible network management. In networking, two important new concepts have emerged: Software Defined Networking (SDN) and Network Functions Virtualization (NFV). Both of them enable novel ways of design, implementation, configuration and management of computer networks and show potential of network performance improvement and cost reduction in previously mentioned conditions. In this paper, we focus on Network Functions Virtualization and present a model-driven, semantics-enabled framework for automated, (re-)deployment of containerized Virtual Network Functions (VNFs) and generation of SDN OpenFlow and network overlay rules as a result. We decide to use container-based virtualization approach in order to decrease the overhead induced by the hypervisor layer, speed-up the deployment and increase the overall flexibility which is crucial in many novel use cases, especially when it comes to Edge Computing.
	Go Back