Monday, February 22, 2016

In U.K., FTTH Payback Remains a Key Issue

As always, financial payback remains a key problem for service providers building fiber-to-home networks, in the United Kingdom as elsewhere.

Aside from prosaic issues such as construction labor availability, the bigger issues remain the business model. Some wonder whether U.K. consumers will gladly pay the higher prices fiber to home networks might require.

The business model is complicated even more in competitive markets where there might be two or even three fixed network providers of high speed access, at least two of which offer hundreds of megabits per second to gigabit speeds.

In such instances, any single provider can reasonably expect take rates in the 20 percent to 40 percent range. That obviously means there is a danger of stranding (network is built but there are no paying customers using that infrastructure) as much as 60 percent to 80 percent of core distribution assets.  

That is why the build only in some neighborhoods approach is so important. That method avoids stranding so much capital in facilities that will never earn a return.

But progress will continue.  

Core communications infrastructure costs have improved over several decades, meaning it now costs less to deploy a modern, Internet Protocol based application, access or transport network.

Mobile networks and cable TV hybrid fiber coax networks cost less than fiber-to-home networks, for example. But all networks have increased their capacity on a cost-per-bit basis over the last few decades.  

Different network architectures, physical media, lower cost computing and storage, open source code, virtualized networks and Internet Protocol itself have helped. Additional spectrum has been a key enabler as well.

But greater cost reductions are expected. One new initiative, the “Telecom Infra Project,”  aims to to “develop new technologies and approaches to building and deploying telecom network infrastructure,” according to Jay Parikh, Facebook global head of engineering and infrastructure.

Facebook, Intel, and Nokia have pledged to contribute an initial suite of reference designs, while other members such as operators Deutsche Telekom and SK Telecom will help define and deploy the technology as it fits their needs, said Parikh.

Telecom Infra Project  members will work together to contribute designs in three areas including  access, backhaul, and core and management.

Significantly, the effort will apply Open Compute Project models of openness and disaggregation as methods of spurring innovation. In other words, in addition to relying on open source, the Project also will rely on use of standard, “commodity” hardware.

“In what is a traditionally closed system, component pieces will be unbundled, affording operators more flexibility in building networks,” Parikh says.

Facebook, in collaboration with Globe, recently launched a pilot deployment based on Telecom Infra Project principles to connect a small village in the Philippines that previously did not have mobile coverage.

In addition, EE is planning to work as part of TIP to pilot a community-run 4G coverage solution that can withstand the challenges presented by the remote environment of the Scottish Highlands to connect unconnected communities.

Testing new technologies and approaches and sharing what we learn with the rest of the industry will enable operators to adopt new models with full confidence that they will be sustainable.

Without much doubt, the Project will help service providers build networks at lower cost. The Project also continues to show the importance new providers and suppliers represent in the communications market.

As Google Fiber represented both a challenge to thinking about how to build fiber to home networks, and a challenge to conventional thinking about access provider market share dynamics, so too investment in unmanned aerial vehicles and unmanned balloons for Internet access are bringing new possibilities to Internet access infrastructure.

Now Facebook will try and leverage the principles behind its Open Compute Project for data center infrastructure to the problem of access and transport networks.

As always, in competitive markets, the low cost provider tends to win. Just as certainly, overall access and transport costs are going to keep dropping.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...