The advent of the software defined networking (SDN) paradigm with its associated open programmable device interfaces and controllers has pushed virtualisation forward toward the goal of mainstream adoption within the Telco domain. The term network functions virtualization (NFV) has emerged in recent year’s which defines the process of virtualizing network functions such as gateways, proxies, firewalls and transcoders, traditionally carried out by specialized hardware devices, and migrating those functions to software-based appliances, deployed onto commodity IT infrastructure. NFV along with SDN are one of the most promising paradigms for revolutionising the way in which networks are traditionally manage and controlled. The NFV concept is based on technologies that have proven their validity in IT, and it is the result of careful experimenting and evaluation by players both in the industry and academia in recent years.
Adoption of a NFV approach to the rollout network functions leads to various benefits including: (i) efficient management of hardware resources, (ii) rapid introduction of new network functions and services to the market, (iii) easy of upgrades and maintenance, (iv) exploitation of existing virtualization and cloud management technologies for the NFVs, (v) significant CAPEX and OPEX reduction, (vi) enabling increasing the breath and speed of innovation within the ecosystem; and (vii) encouraging openness within the ecosystem.
However various challenges needed to be effectively addressed within the virtualisation area. One such challenge is how to effectively deploy NFV in the carrier domain to leverage the advantages of the IT ecosystem while at the same time minimising any of the associated disadvantages. Standard high volume servers and software must be modified to meet the specific reliability requirements of carrier grade telecoms environments. This mission critical level of reliability is a key requirement and differentiates traditional IT from telecom environments, where downtime or poor performance are not acceptable. To meet the required design goals without sacrificing performance, software applications must be specifically designed or rewritten to run optimally in virtualised telecom environments to meet carrier grade requirements. Otherwise, applications ported to virtualised environments may experience significant performance issues and may not scale appropriately to meet the requireme
nts of dynamically varying network traffic workloads.
Deploying NFV also incurs other well-defined risks such as scalability, joint management of network and IT resources, handling network and software failures, backwards compatibility with existing Operational and Business Support Systems (OSS/BSS), or even interoperability and Quality of Service management.
Therefore, in order to address the aforementioned challenges and risks, the T-NOVA project proposes an open architecture to provide virtual network functions as a service (VNFaaS), together with a dynamic, and flexible platform for the management of network services (NS) composed by those virtual network functions (VNFs). The proposed architecture allows operators to deploy distinct virtualised network functions, not only for their internal needs, but also to offer them to their customers, as value-added services. Virtual network appliances (e.g. gateways, proxies, or even traffic analysers) can be provided on-demand, eliminating the need to acquire, install, and maintain specialised hardware at customer premises.