FiOS: The GPON RFP Arrives

Well here we are at one of the two points that destroyed the Access Business within Tellabs. If we look forward in time, one can see this is the point at which Tellabs tipped inevitably downward into oblivion and is about to be part of Coriant.

As we had predicted, Verizon wanted 2.4 Gbps downstream for their GPON and we had plans to use our own OLT MAC. As I have previously written, we started that project when Broadlight turned us down. We went off and built the silicon with an FPGA and were working on bringing up GPON Cards.

We had started building them on the UMC-1000. The next step back will be on bandwidth. If you recall, the UMC now had 2.5 Gbps capability and we knew we were going to go up against platforms that supported 10 Gbps. We had plans for how to deal with that. My friends at Tellabs Ashburn (aka Vinci) began building ONTs. What is humorous is that for the longest time our Manufacturing folks used the 1000 GPON blade in test fixtures as it was the first one that worked.

The problem with the UMC in this situation is that Verizon disqualified it out of hand as it was only a 622 Mbps platform. Yes that Ethernet limitation stuck in the Executive brains in Verizon even though the lower levels knew that this limitation was removed. We also had a plan to run an external 1 Gbps port to each GPON blade and use an Ethernet Switch as a concentrator for this. Each GPON blad would support 2 GPON ports (and take 2 slots) so an entire chassis would support 12 Gbps total throughput. Not the most elegant way of getting there from here. But given the installed base and work we had done, we thought it was enough.

Another problem with Tellabs was timing. Given all the work to not do anything with GPON - The original design had been cancelled then we promoted enhanced BPON - we were going to be late.

So we were stuck. We could have a product ready that they wouldn't take or we could figure out something else. We went back to our original prototypes and started work on an 8800 based design. The 8800 was being deployed by Verizon in small volume. On top of that the platform had more than enough capability to meet the demands of the RFP. It had two problems: cost and size.

To deal with this, we proposed taking on significant amount of the burden from the Juniper Routers within the 8800 based chassis (it became known as the 8865). We were changing the product's form factor and cost reducing the hardware at the same time as well. Verizon eventually accepted this plan but we were going to end up going into the labs 3rd behind Motorola and Alcatel.

Now Alcatel had stolen a page out of the AFC BPON playbook and been very aggressive with its GPON bid. They became the primary for GPON because of this. Motorola played bridesmaid once again. That must have been very tough on the former Quantum Bridge people as they were the darlings of the Verizon Technical Leadership. Our understanding was that there product had more field problems than one would have wanted and much of the luster was gone by the time GPON rolled around.

This was one of the 3 original legs that the Access Business that was AFC was built on. And the unit went from having over a 90% market share inside of Verizon for BPON to being 3rd in GPON. Leadership and market understanding do mean something.

Again, not sure how aggressive my posting will be the rest of the week given my travel to the Focalpoint Annual Conference.

Jim Sackman
FocalPoint Business Coaching
Change Your Business - Change Your Life!
Business Coaching, Sales Training, Web Marketing, Behavioral Assessments, Financial Analysis
Systemize Your Business Webinar - Free - Register HERE!