As an industry, we are navigating uncharted territory. For the first time in the history of data connectivity, societies in the developed world are operating in a period of bandwidth abundance. From the inception of the internet to recent times, the connectivity from the home or business to the internet peering points or transit backbones has always been constrained by the capacity of the Wide Area Network (WAN). Originally dial-up, then various flavours of DOCSIS, DSL, and Microwave Wireless, the shift has finally occurred where fibre is being properly recognised as the only suitable foundation for the gigabit societies that governments around the world are scrambling to build.

Why is (or rather was) Point to Point Fibre preferred?

Those in the know have always considered fibre the limitless medium. With terabits per second achievable over a single fibre strand, it is difficult to argue with this; nonetheless, an architectural debate has consumed the fibre industry over the past 20 years. Point-to-Point (P2P) fibre connectivity delivers on that true vision of “limitless.” In contrast, Passive Optical Network (PON) fibre, a point-to-multipoint architecture, has always straddled the balance between resource efficiency and the ability to stay ahead of consumer demand.

The P2P proponents have long argued that the unlimited potential, particularly the symmetric capabilities of P2P networks, means they are the only credible way to build a “real” Fibre-to-the-Home (FTTH) network. For a long time, it was difficult to counter these arguments. Moreover, with uncontended 1 Gbps symmetric services on offer, it was difficult for those aligned to PON architectures to counter anything more than the more efficient use of resources.

During the last two years, this situation has largely been turned on its head.

Points of differentiation for P2P networks have become points of parity, and the benefits of PON networks have become exponentially more pronounced. As a result, for the first time in history, we are witnessing P2P-focused operators turning their back on the technology they often fanatically defended in favour of PON technologies.

 

A shift in thinking

When we look at the motivations behind this arguably religious shift in position, we see a complex interplay between the desire to control operational costs, the use of symmetric bandwidth as a means of differentiation, the rising cost of power, and the implications that come with the loss of economies of scale.

P2P fibre has historically held a distinct advantage over PON fibre networks – the ability to provide native symmetric bandwidth. With all the original PON technologies being asymmetric in their design, it was easy for P2P fibre deployments to stand out as being superior. However, this changed in 2016 when the ITU-T standardised on a new PON technology, XGS-PON, significantly increased the capacity that telcos could deliver over PON architectures. Beyond sheer capacity gains, this new technology delivered two highly prized ingredients. The first was a compelling cost structure that meant that these higher capacities were within the budget capabilities for those operators seeking to offer true gigabit and multigigabit service offers on their networks. The second was the ability to offer all services on a symmetric basis, neutralising one of the long-standing differentiators of P2P fibre networks.

Point to Point Fibre or PON, or can it be both?

The emergence of multi-gigabit service offers across most global fibre-enriched societies is triggering an inflexion point for those operators who have built their networks based on P2P architectures and technologies. Do they now upgrade their OLT estates to 10G Ethernet to preserve their relevance in a multigigabit market, or do they reconsider their stance regarding PON and P2P technologies?

In considering this question, it is important to remember that PON technology and PON fibre architectures are not inextricably linked.

It is possible to run P2P technologies over PON fibre networks, and it is equally likely to run PON technologies over P2P fibre networks. The net impact of this is that P2P fibre operators can credibly consider leveraging both 10G Ethernet upgrades and the introduction of XGS-PON as their next step in becoming multi-gigabit relevant without having to rearchitect their P2P fibre architectures.

With both approaches capable of delivering multi-gigabit symmetric services, what are the drivers that an operator should consider when assessing whether to stay the course with P2P or cut over to XGS-PON? Fundamentally these boil down to resource consumption and scale economies.

PON is ten times better – less space, power, and fibre consumption.

When we measure a P2P network on a resource consumption perspective, the 10:1 ratio frequently crops up amongst operators. Ten times the amount of power, ten times the amount of equipment and space, and at least ten times the amount of fibre consumption is attributed to P2P.

The assertion P2P has an added fibre utilization and/or deployment cost to pass a home has historically been dismissed given the arguably little uplift in cost to lay additional fibres while a trench is open; however, this dismissal does not consider the increasing demand and usage of dark fibre assets within wholesale open-access markets. Offerings like PIA from the UK's Openreach, or equivalent dark fibre offers from Italy's Open Fiber, or Germany's Deutsche Glasfaser mean that challenger operators can quite cost-effectively utilise individual dark fibre strands to extend their networks into a region, onto which they extend to many end-users leveraging PON architectures and technologies.

Beyond fibre consumption, the space requirements of P2P equipment are typically ten times that of PON solutions, which has not been a critical concern for many incumbent operators since they often have the space required in existing buildings to house the equipment and cable management needed. However, new entrant challenger operators must acquire the necessary space to accommodate their active equipment and cable management infrastructure. These costs directly erode their bottom line and impede their ability to secure permissions to deploy street cabinets due to their size.

The most pronounced motivator behind the accelerated exodus from P2P technologies we are seeing is the exploding cost of energy. This is most pronounced in Europe and is predominantly driven by the skyrocketing cost of natural gas used for much of Europe’s electricity generation. The fivefold increase in the cost of natural gas has driven an unprecedented number of bankruptcies across the retail energy resale landscape in Europe. This has resulted in surging electrical costs for both consumers and enterprises. There has been more than a decade of continuous Year-over-Year price increases, with many anticipating the pace of enterprise electricity cost increases to accelerate moving forward. Regardless of the religious position, the engineering teams may have regarding P2P or PON technologies, the hard questions on whether operators can afford in the long term to persist with P2P technology are increasing in volume.

PON terminal devices economies of scale

The final and most recent area for consideration is the erosion of the scale economies once enjoyed by P2P fibre active equipment. We must not forget that P2P was the first FTTH technology and architecture deployed at scale, and for some time, P2P as the dominant design held both the OLT and ONT lion’s share of the FTTH equipment market. As global FTTH volumes have increased, the balance of power has shifted. ONT volumes are now dominated by PON ONTs, with P2P ONTs now representing low single-digit percentages of the overall global ONT market – a declining trajectory that continues today. On the OLT side, the port numbers remain a little higher. However, one must recall that each P2P OLT port serves only a single customer versus PON OLT ports that service tens of customers.

As a result of this decline and increasing challenges in the global supply chain, we see some vendors exit the P2P sector completely. In contrast, others divert components over to what they view to be their more strategic growth markets, namely PON markets. This rapidly evolving situation presents the operators deploying P2P solutions with an accelerated decision point on how they proceed in a multigigabit world.

P2P to PON: A Strategic Migration

For operators deciding to transition from P2P to PON technologies, regardless of their fibre architecture, consideration is needed on how this strategic migration can occur. Performing a rip-and-replace strategy is not practical for operators as the cost and service disruption is simply not tolerable.

For most, a cap and grow strategy where all new customers are connected to the network with PON technology, be that GPON or XGS-PON is the only viable option. In scenarios where existing P2P equipment vendors cannot provide equipment in a time frame that aligns with the operator’s business continuity needs, P2P equipment spares can be recovered from the network by leveraging all on-premises customer engagements as an opportunity to swap a customer over to PON and remove them from the P2P system. By continuously consolidating and grooming the ever-reducing P2P subscribers, operators can free up line cards for spares across their networks while tackling their spiraling energy costs.

Arguably technologies and scale economies may shift in balance in the years ahead. Today P2P network architectures can exploit PON technologies by leveraging a PON splitter in the central office/exchange or cabinet location, enabling P2P fibre assets to continue to have a role to play in this world where power and P2P equipment is becoming increasingly scarce. For those that want to remain risk-free, the option to continue to build P2P passive fibre infrastructures while lighting those P2P fibres with PON technologies could offer the lowest risk option for some.