For the past several years, we've been hearing about the great promise of tunable lasers for applications like provisioning "on the fly." Is this just wishful thinking on the part of tunable-laser vendors or are the carriers really considering such applications? What is the status of broadly tunable laser deployments? Unfortunately, no one wants to go on record—certainly not the tunable-laser vendors, which hesitate to reveal their customers, much less their customers' customers. Even the carriers themselves remain quiet.
Iolon (San Jose, CA), for example, contends that it has a number of tier one design wins, which could not be revealed at press time. However, the company last year signed a sourcing agreement with Lucent Technologies (Murray Hill, NJ). Lucent incorporated the broadly tunable Apollo laser in its LambdaXtreme Transport DWDM system, which has since been deployed by Deutsche Telekom.
Last July, Agility Communications (Santa Barbara, CA) announced that Siemens Information and Communications Networks (Siemens ICN—Munich, Germany) had selected the Agility 3105/3106 continuous wave widely tunable laser assembly for use in a new DWDM transponder card, part of Siemens's SURPASS hiT 75xx next-generation optical-networking system. A few days later, Deutsche Telekom and Siemens announced the field trial for a 1,000-km optical network that includes the SUPASS hiT 7500.
In November, AT&T signed a contractual agreement to begin testing the same Siemens system on high-capacity routes in its network. In a press release announcing the contract, the folks at AT&T reveal that Siemens's system "features transponders that are tunable over the full 80 channels of light, resulting in quicker wavelength provisioning for customers, because there is no need to custom order 'colors' of light, and less inventory of transponders means lower costs."
If tunable lasers are indeed part of these contracts, they still seem to be used primarily for sparing and inventory control applications. Manufacturers of broadly tunable lasers have long touted their products as key enablers of dynamically reconfigurable networks, but just when will such applications appear?
According to iolon chief executive John Clarke, there are currently "a number of carrier RFPs that specifically require things like automated provisioning where broad tunability really matters. Carriers have to figure out a way to reduce the cost of not just installing the network but running the network. Quite frankly, that means a new class of equipment needs to be installed that has dramatically lower operating expenses." He believes this new class of equipment includes broadly tunable lasers.
Again, due to nondisclosure agreements, details of any RFP activity come mostly second-hand. In a recent "Telecom Equipment Update," Merrill Lynch cited several pending RFPs, including a long-haul WDM deal with Verizon Communications, which could be worth $100 million. Among those vendors rumored to be in contention are Lucent, Siemens, Ciena, Corvis, and startup Innovance (partnered with Alcatel).
Merrill Lynch also believes that SBC has or will soon issue a metro WDM RFP that specifically asks for reconfigurable optical add/drop multiplexers, which, of course, implies tunability. Merrill Lynch cites Fujitsu, Photuris, Tropic Networks, and Movaz Networks as possible contract recipients.
Even if RFPs are circulating for automatically provisioned and dynamically reconfigured networks, that doesn't mean such networks will appear any time soon, cautions Tom Hausken, director of the Optical Communications Components Practice at Strategies Unlimited (Mountain View, CA). He contends that the broadly tunable laser market is caught in a classic "Which came first, the chicken or the egg?" scenario. Tunable-laser manufacturers have to sell their lasers to system houses, which then have to sell their systems to the carriers. But risk-averse carriers will be hesitant to buy systems that haven't been proven in live carrier networks. "The poor component guys at the beginning of the chain are the ones who have to develop these products to start the cycle," says Hausken, "and yet they have to wait forever for the whole cycle to kick in."
In the meantime, what's really out there now in terms of tunability? According to Hausken, the bulk of deployments thus far feature thermally tuned or narrowly tunable lasers.
Narrowly tunable lasers may not have received the publicity of their broadly tunable counterparts, but several companies have quietly found success in the market, including Avanex, JDS Uniphase, Triquint, Fujitsu, and Furukawa. Startup Santur (Fremont, CA) made waves recently when it declared that its primary competitors are thermally tuned or narrowband lasers and not "specialty lasers" like those made by Agility and iolon. "Agility and iolon must cringe when they hear that," muses Hausken. "What Santur did was put Agility and iolon into the specialty camp, which is certainly not where they want to be relegated."
In an industry known for being relatively slow to change, narrowly tunable lasers may be the best option for now, he says. If narrowly tunable lasers are being deployed in huge quantities, the carriers have already achieved key improvements; their systems require fewer parts, thereby reducing cost and system complexity. Deploying broadly tunable lasers will give them further improvement but at what cost? "I think it's fair to say when they do these little incremental improvements, there's very little incentive to make the big jump, because that big jump could cost them a lot," reasons Hausken. "The industry likes little step-wise improvements."
Iolon's Clarke disagrees. Looking at the true applications for tunability, he says, "narrowly tunable doesn't really get you there...Increasingly, carriers are now actually requiring their vendors to supply them with equipment that provides automated provisioning, which can only be achieved through broad tunability."
For his part, Hausken remains unconvinced. In the short term, he predicts more of the same for the tunable-laser market. "It's not like the carriers will come to a crashing halt if they don't have [a dynamic] architecture in their system," he explains. "Capacity is one thing; there is a blatant economic argument for using WDM, and that swept the world very quickly—the same for higher data rates. But moving to these [dynamic] networks is a different thing. If the carriers don't do this, what is really going to happen?"
Meghan Fuller is the news editor at Lightwave.