Ethernet services

Nortel: There's More to PBT Than BT

And so the fallout begins. (See PBT Sidelined at BT.)

As the data transport and carrier Ethernet fraternities consider the impact of BT Group plc (NYSE: BT; London: BTA)'s decision to favor Multiprotocol Label Switching (MPLS) over PBT in its 21CN next generation network, Nortel Networks Ltd. is licking its wounds. (See A Guide to PBT/PBB-TE.)

Nortel has been the main driving force -- along with BT, until now -– behind PBT, a new, still pre-standards version of Ethernet that, its supporters say, replicates many of the carrier-friendly attributes associated with SDH/Sonet, in terms of service management and scale. (See BT Likes Nortel's New Ethernet Flavor, PBT Means What?, and PBT's Ethernet Appeal.)

And when BT announced in January 2007 that it was awarding PBT equipment deals to Nortel and Siemens Communications, now part of Nokia Networks , it kick-started a lot of interest in the technology from carriers and other vendors, and gave Nortel a confidence boost. (See Nortel, Siemens Win PBT Deals at BT, Nortel on PBT: Today BT, Tomorrow the World!, and PBT Gathers Support.)

Now, though, everything has changed: Whereas BT a year ago was showing 21CN network diagrams with PBT in nearly every link, now MPLS is favored. (See PBT Stars at Ethernet Expo .) Even earlier this month, it was clear that PBT was losing a little bit of the heat it had in the past, according to our LRTV report on the subject:

Nokia Siemens has an MPLS story to fall back on -– it has a strong relationship with Juniper Networks Inc. (NYSE: JNPR) and now has its own MPLS capabilities in-house following its acquisition of Atrica. (See Nokia Siemens Clings to Juniper and Nokia Siemens to Acquire Atrica.)

But Nortel doesn't have that: It had an MPLS router, codenamed Neptune, in development a few years back, but that strategy was abandoned in late 2005. (See Neptune Changes Orbit.)

So how is Nortel reacting to BT's decision?

Down but not out
"Obviously we're disappointed with the BT decision," Mervyn Kelly, Nortel's director of carrier Ethernet, EMEA, told Light Reading Friday.

BT's decision is a blow to Nortel for a number of reasons. First of all, of course, there's the money. What impact will this decision have on Nortel's revenues from BT? "Reduced," remarks Kelly matter-of-factly. "But we don't comment on individual customer revenues."

Then there's the stated reason for BT's change of heart –- that the carrier's customers need multipoint service capabilities right now, and MPLS is the best technology to deliver that.

"It's disappointing that we weren't able to persuade them of the total merits of our solution. We have been doing multipoint with PBB [Provider Backbone Bridges] for years, and now we're enhancing PBB with PLSB (Provider Link State Bridging) to give it better control," and multipoint service capabilities, says Kelly.

"We can now also do guaranteed point-to-point with PBT with end-to-end performance measurements. Altogether that's a replacement for SDH, so we're disappointed that BT has chosen to make a decision based on a smallish area."

A smallish area?

"BT has made a decision based only on the multipoint needs of customers. A lot of industry analysts have pointed out that point-to-point accounts for a lot of services. If you consider multipoint in isolation, then you're likely to look at what you have in place already," which for BT would be MPLS, of course. "But you shouldn't consider anything in isolation -– most transport services are going to be point-to-point," states Kelly.

And then there's the scalability of MPLS to consider, claims the Nortel man. While it's possible to have MPLS end-to-end (core and metro) for data services, "we don't think that's the right way to go because of the scalability issues that MPLS has. You have to take a view about building a transport network that can scale, rather than just delivering a particular service initially. There have been a lot of changes in the past few years –- a realization that MPLS is not the only option. A lot of carriers are wary of building a 1,000-node MPLS network for data transport."

BT obviously isn't one of them, though.

Blow to PBT's market perception
And there's the inevitable blow to PBT's reputation as a potential provider of efficient data transport, now that its main carrier supporter has largely spurned it. Though, of course, Kelly is quick to point out that BT is not the only operator looking to build out new Ethernet capabilities.

"One customer decision doesn't break a market," says the Nortel man. "The market has moved on. It's not down to an individual customer endorsement. There are more customers and there are more vendors. In the past 12 months we have signed more than 30 carrier Ethernet customers, not all of which have been made public."

But how many of those are major carriers? BT was PBT's lead operator supporter and an influential Tier 1 operator too, while many, but not all, of Nortel's other carrier Ethernet accounts are with much smaller players. (See Nortel Wins in China, Nortel Pushes More PBT, Nortel Wins in Colombia, and Nortel Lands More PBT Action, for example.)

"A lot of the existing carrier Ethernet customers are Tier 2 operators, that's true, but that's the nature of the market. Tier 2 carriers make decisions quicker than Tier 1s, which can take a long time to make a decision. Are we engaging with a lot of Tier 1s other than BT? Yes -– watch this space. We expect to sign up some Tier 1 carrier Ethernet customers in the next few weeks and months."

And, of course, BT is still a customer -– it may have sidelined PBT but it hasn't thrown it out altogether. "We are still a 21CN vendor, and they're continuing to look at PBT going forward. It's not like we've had a falling out with BT –- we still do a lot of work with them."

And what of these other Tier 1s? A number have already been identified as being interested in PBT and of conducting trials. Is Verizon Communications Inc. (NYSE: VZ) one of those carriers? "We're engaging with a lot of Tier 1s. Verizon has been a Nortel customer for years. It's confidential," says Kelly. (See Telefónica Checks Out PBT and NTT Joins PBT Roll Call.)

What's next? MetroNext?
So what's Nortel's next move? Will it reinvigorate its PBT marketing possibilities with the immediate launch of its new carrier Ethernet switch, dubbed MetroNext? (See Nortel Preps New PBT Switch.)

" We are still very much committed to carrier Ethernet and its co-existence in networks with MPLS, with PBT in the metro and MPLS in the core... [But] I can't comment on a launch date. We are continuing to invest in the carrier Ethernet portfolio, and MetroNext is one of those investments."

— Ray Le Maistre, International News Editor, Light Reading

douaibei 12/5/2012 | 3:39:49 PM
re: Nortel: There's More to PBT Than BT Nortel is right in preaching this concept. but the real world need to be proofed by number, low risk, new revenue not just new technology.

PBT is a very good try to reuse the ethernet frame format, unfortunately it's too late, the technical world need something complate to proof they value like MPLS or BGP.

business investor do not care what sort of technology they use as long as there is money on the table.

hungnv 12/5/2012 | 3:39:46 PM
re: Nortel: There's More to PBT Than BT why MPLS and MPLS? that is old story and very complicated technology too many protocol to improve it also it is too expensive. Use a technology or box make all service you want and very cheap. Money spend for other not for big Capex and Opex, a good money is net money go to investor pocket. The more you use PBT/PBB/PLSB the more you love this, so simple and so reasonable price. If using MPLS in a core layer then PBT in access layer that concept like Host and remote for local switch
Sign In