But to take advantage of all that, Calient needs new ways of selling the product. That's what it's announcing today: a goal of finding new partners and developing new ways to sell its FiberConnect optical switch.
One option is to develop a version that can be integrated into other types of equipment.
The company is also changing its name to Calient Technologies, because nothing says "dawn of a new era" like a new name.
The plan is the creation of new CEO Gregory Koss -- a name you might remember. He's spent the past two years running BTI Systems Inc. , but his real claim to fame comes as the CEO who sold Internet Photonics to Ciena Corp. (NYSE: CIEN) and Sonoma Systems to Nortel. (See Ciena Buys More Than Catena and Nortel to Buy Sonoma Systems.)
Like other optical switch vendors, Calient has found some small success in niche markets, with revenues of $3 million to $5 million a year for the past few years, Koss says. Its switches get used for protection switching or fiber management, or at the land/sea junction of an undersea long-haul connection.
Those sorts of applications have been the bread and butter of Glimmerglass , an optical switch vendor that emerged a few years after Calient. It claims the Amsterdam Internet Exchange B.V. (AMS-IX) and Cable & Wireless Communications as protection-switching customers.
Polatis Inc. has also kept the all-optical torch lit, albeit with smaller devices. (See Polatis Takes the White House.)
Koss sees potential for Calient to grow, but his assessment is that Calient wasn't flexible enough to grab that business. "What it came down to was, they built one big box. Take it or leave it," he says.
Here's what he wants to do differently:
- Find partners to help Calient expand into other markets. Koss's main target is the government sector, an area where Glimmerglass has found some success.
- Let customers alter FiberConnect. Calient gets requests for more variable optical attenuators or the addition of arrayed waveguide gratings. But Koss doesn't want Calient to spend time on multiple variants of its system. Instead, he's enlisting a partner company in Wuhan, China, to build FiberConnect systems for customers to alter.
"What I was trying to figure out was, what's the next right product after the '320? After working on it for months with Arjun [Gupta, Calient's chairman], it turns out I couldn't, because all the markets that can take advantage of photonic technology -- they're all different," Koss says.
- Put FiberConnect in other people's boxes. That's going to require three pieces. The first, naturally, would be a linecard containing the actual switch -- the array of Micro-Electro-Mechanical Systems (MEMS) Mirrors. But there's also a configuration board that's necessary (configuration will eventually be handled all in software, but not yet) and an optical monitoring module.
The monitoring piece is a big deal all by itself, because it's there where most of the expense of FiberConnect lies. Some customers are sticklers for measuring insertion loss at every port, but Calient does that today by using a whole bunch of optical taps. The monitoring module would be a much cheaper way to do it.
"If you look at what AT&T Inc. (NYSE: T), Verizon Communications Inc. (NYSE: VZ), and others are doing, they've been asking for a colorless, directionless, contentionless network, and the key to that is to build an optical switch."
The concept of the 1,000-port all-optical switch appeared to have died with the dotcom boom, but the new monster switches are a different breed, Koss says. "In Calient's early days, people were talking core capacity. This ROADM application is just lots of add/drops."
If you're wondering who's been willing to fund this all this time, it's Gupta, founder of TeleSoft Partners . Telesoft and some new investors are even putting another $10 million into Calient to fuel Koss's new business plan. Gupta is being named Calient's chairman today, taking over for George Schmitt, another partner at the same venture firm.
It's taken more than $375 million to get Calient to this point, but Koss contends that the company's found a good spot in the industry, partly because so few other companies will bother to build this kind of equipment.
"It's not only very capital-intensive, it's a very long process. The core team that developed the technology and the patents is all together," Koss says.
— Craig Matsumoto, West Coast Editor, Light Reading