FiOS ONTs: Proliferation and Challenges

Once we got the bulk of the bugs run out of the BPON product Verizon began deploying in massive numbers. These are not cell phone numbers or iPod numbers but we got to the point that a business day might see 10,000 installs a day. At that rate, we were deploying on a monthly basis what many of the small competitors had done life of company. The shear volume of this deployment created its own challenges. Networks like to have stability and really want to have no changes. Just having the network technician mucking about installing new cables and attaching more equipment led to problems. On top of that, we could tell when new region was introduced to FiOS. The initial failure rates climbed, then fell. This was all new technology and people had to learn. I think that is one thing that people miss in these big rollouts of new technology. Literally 10s of 1000s of people inside Verizon were impacted by FiOS. Imagine the amount of training and change that happens. I know people complain about Service Providers moving slowly, but watching this all unfold was a huge learning experience for me.

Well, the initial deployments were suburban for FiOS. There were many construction cost advantages to this, especially as we shook out the network. The Single Family Home ONT itself went through several revisions, but they all supported a similar number of interfaces. The reason for most of the changes was cost reduction.

We did have two other kinds of ONTs. They were targeted at specific segments: Multi-Dwelling Units (like Apartments) and Small Business Units. The former ramped up the number of phone ports and often used VDSL or VDSL2 as the link to the gateway inside the residence. The latter had T-1 ports. Both of these units failed and for different reasons.

The easier to talk about is the SBU (or MTU) ONTs. These failed because Verizon could not figure out how to manage T-1 lines off ONTs. To this day, I am not sure what the issue was. But we built several variants per our contract and they could work in the system. But making them work operationally for the field to deploy became an insurmountable challenge for Verizon. In the end, I think they gave people T-1 over Ethernet Adapters and called it done.

The MDU failure is a bit harder to explain. There was always the challenge of modularity. No matter what we built it would never match the needs of the building exactly. On top of that there was a big issue on the coax side. With an SFH ONT, there was a single home's worth of TVs hanging off of the coax port. So, there might be a few hundred feet and 1 - 5 TVs hanging off the port. For an MDU, those numbers were a LOT harder to get a handle on. In some cases the coax signal was too high and in others too low. The VDSL(2) ports were also hard to get right as the number of MODEMs available was low. In the end, Verizon would bite the bullet and run fiber to each apartment and give them a SFH ONT.

So, what is the lesson here? Really, Tellabs had almost no impact on the outcome. These problems were either inside Verizon or became a Rubik's cube of challenges. That is the problem with the big deployments. Where Verizon could provide 100% clarity things went fast. Where they couldn't they ground to a halt. One off issues are very hard to deal with in bulk deployments. Think about that every time you want to scale a business and a customer wants a "special".

Jim Sackman FocalPoint Business Coaching We Focus On Your Business - Time, Team, Money, Exit Business Coaching, Sales Training, Web Marketing, Behavioral Assessments, Financial Analysis Lunch and Learn