Proficient, long-time networking engineers and planners comprehend that designing architectures can be really tough, and they realize the main reason Verizon wants to do home-runs from the CO to the cell towers for 5G. A centralized approach reduces the number of active nodes in the field that require power (including backup) and maintenance, which will become part of the total lifetime cost of sustaining the service. In fact, of all of the principal necessities for a small cell, the priciest is power. So, in not introducing a large number of intermediate points, just the expense saved in not having to replace the batteries every several years can be substantial. Another big advantage with placing all of the Baseband Units (BBUs) at the CO is that there is the most efficient use of the bandwidth, as there will be variations on the number of users at each of the towers. Other benefits of the centralized strategy include: a simpler network to manage, greater control over security matters, lower administrative costs, reconfiguration/capacity upgrades accomplished at one network point -- and the greater unlikelihood of not winding up with a noticeable amount of unused capacity, which could otherwise, theoretically be unbundled to competitors by regulators. Conversely, the upfront fiber cost is substantially higher with this approach of providing fronthaul capacity, especially in ensuring the necessary latency. Most critically, the biggest disadvantage with this method is the predisposition to a devastating failure, including fiber cuts, which also could take out any unused strands held in reserve.
In the initial stages of the CATV industry, the entire infrastructure design was based on the home-run model, until the challenge with costs from the standpoints of fiber management and service reliability were figured out. The most important benefit of a decentralized architecture is that there is not a potential, single point of a collapse, while offering the assurance of route diversity. Therefore, there is a greatly lower threshold of pain with a fiber cut, along with avoiding anything harmful happening to the service riding on that optical transport.
We have no doubt that Verizon realizes this problem and is beginning to ask itself questions, such as, what is the likelihood of getting the service up in, say, two hours? An intermediate point facilitating capacity diversity may even result in a better cost equation, which considers the number of nines of reliability that can be supported. Perhaps Verizon determines the number of minutes a year that can be tolerated in case of an outage, and then works backward from that assessment, compelling the operator to include some strategically-placed, midway nodes for protection purposes.
Intriguingly, it appears that AT&T has adopted the decentralized game plan for 5G fronthaul because as always, keeping its first-cost to a minimum, trumps every other consideration, and also, fibeReality’s continues to believe that the service provider is not truly serious about making the jump to the next generation of wireless. Thus, its plan is to put the BBUs in a central drain site, and then connect up with the CO, using passive gear to reach the targeted latency. Unlike Verizon, it evidently chose to avoid the high cost of installing a lot of fiber to achieve the desired gain, in which CPRI especially becomes a limiting factor.
With Verizon’s actual concentration of 5G on the enterprise space, and although it has been way off on its initial projections for fiber expenditures, the payback period could still turn out to be reasonably short. Given the various levels of mission-critical applications, and the cost of network downtime being particularly high, the company will be able to demand an impressive premium from its customers.
As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.
To follow us on our totally separate, quick update company blog, which is exclusively on fibeReality’s LinkedIn page, please click here.
[written by Mark Lutkowitz]