Another gold rush has brought a high level of excitement to the network infrastructure producers and consumers alike.

SDN/NFV: Gold Rush or Fool’s Gold?

  • Jan. 28, 2020
  • in SDN .


The mad dash to SDN/NFV feels like déjà vu, for example, mid 1990s for ATM and late 1990s for MPLS. See Paul Parker Johnson’s “How SDN (Today) Is Like MPLS Was (Then).” There are huge expectations from all stakeholders to offer and implement infrastructures that reduce both capital and operational expenditures, in addition to opening new doors for rapid deployment of innovative and lucrative business services.

Intuitively, the SDN/NFV combination should reduce the total cost of ownership (TCO), both capex (COTS versus purpose-built hardware) and opex (cost of provisioning and network maintenance). In evaluating TCO, there are other costs that could favor one approach versus the other.

Most often, capex savings are only discussed in terms of COTS hardware versus physical or purpose-built hardware. Basically, capex includes any upfront nonrecurring cost; that includes the cost of “network roll-out” (NRO), which is the cost of integration, testing and verification of the incremental hardware into the existing infrastructure. Unlike the cost of hardware, this cost component is not usually depreciable unless the NRO is done by the hardware vendor, and the cost is negotiated in advance. Other capex costs can include the cost of the underlying transmission network (in some countries this is leased). For NFV, the transmission network (and eventually the hardware maintenance) can be leased from the owner of the data center, which turns this cost into an opex component as it becomes a recurring cost.

A major advantage of SDN/NFV is in its opex, which gives the operators the ability to rapidly provision new services. Service roll-out is reduced by an order of magnitude of months to days. Moreover, with fast service roll-out, a new service can be tested with a limited set of customers first, and then upon favorable feedback it can be introduced to the entire target market. This can save a lot of headache (and money) later if the service turns out to be not as well received as it was expected.

Today, most infrastructures that are built on purpose-built hardware are going to stay in operation for a while and in many cases even after they are fully depreciated. Therefore, while migration to function virtualization is moving forward, operators will face a period of a “double opex” cost factor. This is not lost on anyone, and it can become a factor in delaying the decision to virtualization.

The move to virtualization requires a close study of the intermediate and long-term goals of the organization: customer needs, market penetration goals, and service offering to name a few. Although cost containment is a big factor, the revenue side of the equation must be given a much higher weight to remain competitive. After all, costs cannot go below zero, but the sky is the proverbial limit for revenue generation! And this is where SDN/NFV based infrastructures shine: rapid deployment of new and potentially lucrative services.



Posted by

Got another minute? Check out:

BLOG

Aug. 11, 2020

Special-purpose networks, such as IP-VPNs, SD-WANs, and low-latency trading networks have been offered by CSPs for many years but engender large extra capex and opex costs. Re-imagining these networks as slices of a common network infrastructure promises to decrease the capex cost, but makes them more complex to plan, instantiate, and manage across the network domains. Moreover, the requirements of network slicing extend across every area of the network as well as network and business operations and the supporting OSS and BSS software systems. It will be a long journey. ACG Research is studying and following these journeys by the standards bodies, service providers, and vendors as they create the sliced network of the future.

...