AlcaLu's Alwan: PBT Will Lose Its Shine
That Alwan would say that comes as no surprise: AlcaLu is one of the major IP/MPLS router players that stands to lose out if PBT, also known as PBB-TE in the Institute of Electrical and Electronics Engineers Inc. (IEEE) standards process, picks up commercial momentum and is deployed in any scale by major operators. (See MPLS: Metro a No-Go?, Vendors Clash Over PBT, and PBT vs MPLS: Round VII.)
But while many views in the industry go unnoticed, the AlcaLu man is a respected and influential figure whose opinions carry weight in the vendor and carrier community. (See AlcaLu Expands Alwan's Role and AlcaLu Exec Wins Popularity Contest.)
Chatting to Light Reading on the phone after his recent presentation at the MPLS & Ethernet World Congress in Paris, Alwan admitted that PBT has attracted a lot of attention among major carriers, but he has his doubts that the technology will live up to the hype created by the technology’s supporters. (See PBT Key to Ciena Acquisition, BT Goes Live With PBT, BT Counters PBT Claims, Nortel Pushes PBT Pact, Nortel on PBT: Today BT, Tomorrow the World!, Nortel, Siemens Win PBT Deals at BT, Deutsche Telekom Flirts With PBT, and Huawei Joins PBT Fan Club.)
”The industry loves debate -- technologies go through a hype cycle and then settle into a role. The main thing about PBT at the moment is the provisioning and control plane functionality. It fits in with the traditional OSS that is designed for transport networks. The question for the industry is: Can we do these things in MPLS? Of course we can,” claims Alwan.
”The same market forces that drive MPLS will drive any carrier Ethernet product. It’s about plugging in with the OSS and carrier mindsets. That’s why we [Alcatel-Lucent] are agnostic."
Agnostic? Really? (See PBT Cost Claims Questioned.)
"PBT isn’t bad –- it’s an approach. But what problem is it solving?” asks Alwan.
”Some of the arguments [in favor of PBT] are short-term. They may be real today, but they don’t stand up in the long term. That’s my view based on 20 years of building switches.”
MPLS is 'hitting its stride'
Alwan recalls there were similar questions around MPLS when it was in the early stages of development. “When I first started working on this technology at TiMetra, we debated whether MPLS would become a mainstream technology. MPLS is just hitting its stride -- it went through its hype cycle, found its place, and now it has hit the market in a very major way, and has contributed to a renaissance in router design.
”Now there’s a huge amount of adoption by carriers such as Belgacom SA (Euronext: BELG), BT Group plc (NYSE: BT; London: BTA), Wind Telecomunicazioni SpA , and others." (See Wind Launches IPTV, Belgacom Uses Alcatel, and Alcatel Signs BT 21CN Deal.)
And just as this is happening, so another technology comes along, claiming to do many of the same things better and more cost-effectively. “This industry has the tendency to do the same thing but in different ways, and that adds complexity.”
Alwan adds that PBT has emerged as the telecom world is in flux. “We’re at a point where two worlds are colliding -- transport and packet Ethernet, and the carrier Ethernet debate has the headlines. But it’s not just about the technology. It’s about organizations, and politics within equipment companies and service providers.
“It’s an important debate, for sure, but we believe the way forward is to enhance MPLS. Hopefully [the industry] will take an MPLS approach to transport requirements. Let’s use the technology that’s already widely deployed. We have to stay focused on what our customers need.”
So won’t technology based on an eventual PBB-TE standard have a role to play at all? “I won’t comment on where PBB-TE might end up. There’s a lot of evaluation at the moment, and that’s good. But we believe that enhancing MPLS is the right direction. There’s a lot of overlap, and MPLS has a huge headstart.”
What does PBT 'bring to the party?'
Alwan is clearly irritated by the way some of the PBT debates are evolving, particularly around relative cost and functionality. PBT’s supporters stress how its functionality is based around “MAC address encapsulation -- but what does that bring to the party? That’s the big question on the table... If you go the chip level there is zero difference between the address lookup in MPLS addresses compared with MAC addresses. The big difference in cost at the moment is linked to scale. I don’t expect a cost differential to exist [in the future] -- there is no fundamental basis for it.”
He continues: “2008 is going to be a pivotal year in this particular debate, and the goal should be to enhance MPLS to further support OA&M [operations, administration, and maintenance]. MPLS has real contracts, real networks –- there are hundreds of networks being built based on that technology.”
MPLS set to improve
MPLS is going to improve, Alwan believes. “Network performance monitoring and performance management -- TDM networks have this, but MPLS still has to bake this in and have a standardized approach. Part of the tension around the transport-versus-packet debate is around this area. That’s a debate that will be resolved in time,” says the AlcaLu man.
In the meantime, Alwan is bullish about the future of Ethernet as a carrier-class service technology. “2008 will be a big year for carrier Ethernet. Ethernet as a service -- that’s what is really going well right now. It’s a really strong market.” (See Ethernet Services Hit $9.4B.)
— Ray Le Maistre, International News Editor, Light Reading