Adtran’s TA 5000 Likely to be a Big Star

Implications: 1.      There is a high probability that Adtran’s TA 5000 will have a strong run for the next five to eight years.   2.      Its latest product design that can provide a whole host of services could substantially dwarf the impressive sales Adtran accumulated in delivering T1s over HDSL. 3.      For the carrier that wants a bunch of new access platforms that can deliver just about everything, the TA 5000 is going to be among the top of the list.

Analysis:  Adtran’s market opportunity for T1s over HDSL has decreased and the average price of the equipment has gone down.  However, it would be wrong to overemphasize the cannibalization of this market by the TA 5000 – or to only look for the opportunities on an individual application basis.  It has to be examined in terms of the totality of the capabilities that can be provided.  And Adtran with its newest TA model is that much further out on the learning curve – and the development people have gotten even smarter – and they have cleverly designed (with a minimal upfront cost penalty at worst) a product that can deliver T1s, ADSL, GPON, DS-3s, RPR, etc.  It also has SONET prospects on the back that can apparently go as high as OC-192.   

In building such a platform, Adtran ran into tradeoffs of how much of any particular   application it wanted to do that would obviously have cost ramifications.  Thorny issues included how high to make a slot, how much bandwidth is brought to it, how to protect module to module, and how many copper pairs are taken to a slot.  Generically speaking, in developing such a platform, it could wind up being as big as a refrigerator – and an empty shelf with the common equipment could cost $20,000 alone.  But this does not seem to be the case with Adtran’s new mid-plane design.  Apparently, the supplier has shrewdly figured out the redundancy issues and can evolve the platform to take it to higher and higher bandwidths.   

Of course, on the negative side, in talking about applications that are not just done at one out of every 20 COs, but perhaps even at each CO – where many more shelves might be required – that matters like density and cost per port can look unattractive.  In addition, it can also be a nightmare to regression test additional feature releases – no matter how large or small. The vendor can get so bound up in the process that it is almost immobilized after three or four releases.  

Nevertheless, with the admirable job that Adtran has done, the older products from Tellabs and Calix – and even Entrisphere just kind of starting up with Ericsson – will have a tough time competing against the TA 5000.