Network Function Virtualization (NFV) offers network operators a new way of building complex IT applications and taking them to market faster.
The two key aspects of NFV are:
- Service chaining: multiple virtualized network functions (VNFs) are used in sequence to build more complex network services.
- Orchestrating: to support multiple VNFs, the network must be able to characterise VNF instances, monitor and repair them and bill for the services provided. An orchestration layer is responsible for this complex task.
NFV enables operators to:
- Reduce the capital expenditure of building networks (CapEx) by supporting a pay-as-you-grow model that eliminates unnecessary overprovisioning.
- Reduce operational costs (OpEx) by cutting down the number of equipment needed, and network services management.
- Address evolving demands by delivering flexibility and agility by quick scaling up or down services.
- Easy trial and change of services, as well as accelerate time-to-market.
What is Network Function Virtualization?
NFV stands for Network Function Virtualization, and it is virtualizing all existing OSS (Operations Support Systems), in the case of telecoms, or enterprises legacy and dedicated hardware in the case of organisations, to make them software driven on standardised hardware.
To date, these network services and functions have been complex, costly and time consuming, requiring specialised teams, and has also presented problems around scale, elasticity and lock-in.
This static model is rapidly becoming outdated in the cloud-based world that humanity now lives in, and this offers an opportunity to the telecoms industry as a new service to the market.
In the days before NFV and SDN (Software Defined Networking), these new services could take more than a year before they were ready to be introduced, but NFV acts as a growth enabler by making these new services just about instantly available in minutes.
To say NFV is a game changer in the industry would be an understatement.
Examples of NFV include virtualized firewalls, intrusion detection devices, load balancers, and WAN accelerators.
How does Network Function Virtualization work?
Typically speaking, network components usually require significant code changes to turn an existing network function into a virtualized network function, to make it run from virtualized hardware and software (the creation of a virtual machine that performs like a real computer with an operating system) to the cloud. It is difficult to accomplish in a standardised manner, and without losing control.
Enter NFV, which reduces expenses and accelerates deployment by moving functions (e.g. encryption and firewalls) from dedicated hardware to commodity servers.
Using NFV, general-purpose servers replace dedicated, hardware-based network appliances; the purpose being for the Management and Orchestration (MANO) system to administer and coordinate the functions that execute on these host servers. The VNFs that are created are founded on the telecom operator’s and/or organisation’s needs, to be deployed when and wherever.
NFV network architecture and components
The NFV architecture helps define standards for NFV implementation, and each component within the architecture is based on these standards for enhanced stability and interoperability.
The architecture includes:
- VNFs (Virtualized Network Functions); software applications that deliver network functions such as directory services, file sharing, and IP configuration.
- NFVi (Network Functions Virtualization infrastructure) includes the infrastructure components (compute, storage and networking) to support software that is needed for network apps to run.
- MANO (Network Management, Automation and Network Orchestration) gives the framework to manage NFV infrastructure and the provisioning of new VNFs.
Benefits of NFV
- NFV allows service providers to run network functions on standard hardware instead of dedicated hardware.
- Multiple functions can be run on a single server, meaning that you don’t need as much physical hardware, resulting in consolidation of resources, less physical space needed, and a reduced amount of power, and therefore reduced cost.
- Flexibility, as VNFs can be run across different servers or moved around. What this translates into is that operators can launch new services and apps a lot faster.
- Provides a low-risk method of testing the value of potential new services.
NFV offers operators a new and faster way of building complex technical applications and deploying them, bringing unprecedented agility and flexibility.
Further pros of using NFV include running network functions on standard hardware and a single server, needing less physical space, a reduction in the amount of power needed, and therefore reduced cost, a low-risk way to test potential new services, and take new services and apps to market a lot faster.