Place VNFs where needed

Is it better to distribute VNFs out to the service edge, or consolidate them in a cloud datacenter? The answer depends on the service.

Sometimes it makes sense to put all of VNFs in a cloud or CO to take advantage of economies of scale seen in data centers. At other times it makes sense to host VNFs at the customer site for security or latency. These options should be seen as complementary, not mutually exclusive.

Pure-play virtualization allows a CSP to put the resources where they are needed (centralized or distributed), based on the requirements of the customer and service and available resources.

Pure-play enables service agility

NFV makes service delivery more agile with quicker response to meet a broader array of customer demands. How does pure-play virtualization achieve that agility? Through the following:

• Mobility. With pure-play virtualization, CSPs can move VNFs based on the demands of service. A customer might start with a centrally-hosted service. If later expansion calls for functionality at the edge, the CSP can migrate VNFs to the service edge. The service topology changes but the core VNFs and their management remains the same.
• Resilience. Availability and up-time are major concerns of CSPs as they move to virtualized services. With pure-play virtualization, CSPs can take advantage of system resiliency that comes from software running on open servers, like cloud or data-center resiliency.
• On-Demand. Free trial for software-based services is a proven marketing tool. Because the incremental cost of a free trial is very low, suppliers can aggressively offer customers the chance to be sure that a given application meets their needs. With pure-play virtualization, CSPs have the same ability to offer services to customers – without deploying a physical appliance to the service edge.
• Scalability. A huge advantage of converting network functions from appliances to VNFs is that they can scale up and down, based on demand and the underlying server. Scaling on demand enables all kinds of usage-based services, both for end users as well as for network infrastructure. Scaling based on the server provides future-proofing: the investment in operational integration of a VNF is preserved while the performance increases with new and more powerful hosting servers.

Go virtual, all the way

The vision for NFV is to enable the use of standard and open compute servers to host a variety of mix and match software VNFs.

To fully achieve the NFV vision means delivering these VNFs and services in a scalable, cloud-based manner to all parts of the network. That means keeping the value-added functions in software, including layer 2 Carrier Ethernet functionality. Doing so enables complete flexibility in delivery of services at any part of the network, and ensures that operators will be able to ride the technology curve of open servers. The result is a simultaneous lowering of costs and the enablement of new services, which combine to drive agility and profitability.