What is Network Function Virtualization?

What is Network Function Virtualization?

With fast moving, technology and services innovation, network devices hardware lifecycle are reducing exponentially. Launching a new service often requires the variety of challenges including space, power, and significant investments. This becomes a daunting task for large Service providers and very large enterprises that get locked with proprietary hardware appliances and struggle to generate RoI on used capex. 

Network Functions Virtualization (NFV) aims to transform the way networks can evolve standard IT virtualization technology to consolidate many network equipment types onto industry standard high volume servers, switches, and storage, which could be located in data centers or end customer premises.

Why towards Network Functions Virtualization

  • Reducing Cost by consolidating network equipment’s
  • Reducing time to market cycle by moving towards software-based approach
  • Standardization and Open interface adoption
  • Support large scale multi-tenancy allowing shared resources and services for Operators.
  • Increased automation, resulting in operations simplification & business agility

NFV can be a compliment to SDN adoption by aligning closely with Commodity switches and servers. In a majority of cases, service providers will benefit from virtualizing network functions, even functions that perform real-time packet processing and whose current implementation uses network processors or ASICs.

Virtualization is the primary stage towards NFV evolution in which network function like router, firewall, load balancers etc decouple from their dedicated hardware devices allowing them to be hosted as VM over hypervisors. This capability is important because it means that network administrators will no longer need to purchase dedicated hardware devices in order to build a service chain hence reducing the cost incurred in data centers due to over provisioning.

Potential Functions which can be virtualized


NFV key pre-requisites before implementations

Hardware appliances are specially designed to meet the reliability and performance criteria for the respective network functions running over them. However, in a virtualized environment, network software/function is decoupled from hardware and hypervisor software. The evolution to NFV will, therefore, have profound implications on the way reliability and SLAs are managed. There are a number of challenges to implementing Network Functions Virtualization which need to be addressed by the community interested in accelerating progress, few of them are mentioned below:

  • Reliability & Stability: End-to-End Service availability becomes the heart of the solution with NFV as service is no more specific to individual islands of specific network devices however it is in software. New monitoring and analyzing tools will be required to manage end to end infrastructure.
  • Security: Manually defining and managing security zones can become potential gray areas as defining them in VM world requires significant planning. The introduction of hypervisors open areas for cyber attacks as different components like X86 hardware, hypervisors, and applications can be from different vendors and may lead to a risk of security holes.
  • Interoperability: The challenge is to define a unified interface which clearly decouples the software instances from the underlying hardware, as represented by virtual machines and their hypervisors.
  • Migration from Legacy Network: Migration from current legacy proprietary hardware to open based virtualized application will require significant planning and resources. Therefore NFV must go hand in hand with existing deployment in the hybrid environment to make sure minimum services disruption.
  • Management & Orchestration: consistent management and orchestration architecture is required.
  • Automated Lifecycle Management: every virtualized network function could come with its own proprietary lifecycle management, but service providers would obviously benefit from a consistent approach across all vNFs. The use of tools for automated lifecycle management and convergence on one or a small set of compatible tools will provide significant benefits to both, service providers and vendors.


The Use Cases

NFV framework has many uses cases for Service providers; few of them are as below.

  • Infrastructure as a Service (IAAS)
  • Platform as a Service (PAAS)
  • Virtualization of CPE (v-CPE), PE (v-PE) from Service Provider as part of virtualized solutions
  • Virtualization of Mobile Core Network, Base Stations

Key NFV Market Players

Add to Favorites