The MPLS donnybrook: How long will it last?

Telecom industry consolidation is a fact, and the prospect of totally anti-competitive attitudes by mega-carriers remains latent. But none of this changes the reality on the ground of a real shootout between AT&T and Verizon for MPLS customers. Competitive RFPs for enterprises looking to get past legacy networks, primarily frame relay and ATM, and into the IP/MPLS space have shown amazing price competition from the two giant carriers during the past two years.

The question is: Is this an ongoing phenomenon or will it peter out? We are watching several factors to evaluate this question.

MPLS competition really accelerated once the two top carriers acknowledged that MPLS didn't have to be deployed only in a "managed" guise, meaning carrier control right into the customer premise device. Nor was it a service that all customers had to be baby-stepped into with a frame relay interface. Today's match-up was set once AT&T, with its AVPN service, and Verizon, with the Private IP service it assumed from MCI, had built mature platforms for true, unmanaged MPLS services.

In particular, a key point was reached when AT&T settled on a way to deploy and price out AVPN, which involves a required choice of a Class of Service package for every site by enabling either some real-time traffic or entirely non-real-time traffic according to the customer's specs. Verizon's PIP does not require this additional rate element for sites that do not anticipate running voice-grade traffic, but AT&T's per-site CoS pricing turned out to be economical enough in truly competitive procurements to produce a real donnybrook for the business.

Or at least it did so when the final, crucial rate element was added: access. There's no question that AT&T and Verizon have stolen a march on potential competitors with their access footprints. That's true both in terms of their national set of interexchange carrier POPs (from the legacy AT&T and MCI) and the benefit they appear to get from adding local networks for large swaths of the U.S. under the same corporate roof via the mergers.

But much depends on your going-forward assumptions. Does the combination of economical T-1 access, a fair MPLS port price, and properly compared CoS charges only last as long as AT&T and Verizon think they are locking in national customers for a generational change in technology? Will access pricing ultimately seize up under the economic principles of an emerging duopoly? Can the varied challenges being faced by Sprint, Qwest and Level 3 which we are exploring in this blog eventually be resolved, so that at least one can fulfill the vigorous No. 3 role in highly granular corporate IP networks for medium and large enterprises?

This clearly is the underlying subtext of even many non-IP, non-MPLS-related issues examined here. We will continue to follow it closely and we welcome your input and experiences as well.