What would it take to make the optical division of Nokia Networks into a viable standalone company?

Marlin Equity Partners , which announced on Monday that it's buying the division, says it wants to be a consolidator in the optical sector. But sources tapped by Light Reading aren't sure that's the best idea. (See NSN to Sell Optical Business.)

Suggestions that came up included creating a broader play for software-defined networking (SDN) in the metro, or simply flipping the business to Juniper Networks Inc. (NYSE: JNPR).

At the core of some observers' concerns is the weak state of optical networking. "There's growth in the sector, but it has to find a way to grow profitably," says Larry Schwerin, CEO of components and subsystems vendor Capella Photonics Inc.

Marlin vs. the big fish
Not that consolidation is a bad idea. The top optical-networking systems vendor tends to have about 20 percent market share, with a multitude of single-digit competitors trailing the top three or four. NSN is one such. Infonetics Research Inc. pins the NSN's optical revenues at about €400 million (US$522 million) for the year, good for 4 percent market share.

The consensus has been that consolidation is in order, and Schwerin, a former venture capitalist, has noted for a couple of years that private equity has been circling the sector. He's also expected to see a move toward vertical integration among optical companies. (See Can Vendors Build Their Optical Components?.)

If Marlin wants to combine NSN with other optical properties, it's already gotten a start. In October, the firm announced plans to acquire Sycamore Networks Inc. (Nasdaq: SCMR), adding $54 million in annual revenue, based on Sycamore's last four quarters.

That's still not exactly a powerhouse, and Schwerin isn't convinced that adding an optical components company would be that much help either. What could Marlin do, then?

Rather than pile NSN together with more optical companies, Marlin should combine it with metro packet technology, argues Tom Nolle, principal analyst with CIMI Corp. .

Nolle's idea is that Marlin, or anybody targeting metro networks, for that matter, should be melding packet and optical technologies under the same control software. Yes, that brings software-defined networking (SDN) into the discussion.

"If you buy nothing but optical and you don't look at this other direction of how you're going to integrate it into a metro strategy, you're buying parts that don't add up to a whole. You're amassing failure," Nolle says. "My question for Marlin is: Are they looking ahead far enough?"

No metro vendor is adequately pursuing the path of integrating packet and optical aggregation, Nolle claims. And he's picking on the metro space because it's the best telecom sector, business-wise ("There's metro networking, and there's networking that doesn't have a hope of being financially viable," he says) and because he believes SDN would be relatively simple to implement in a metro aggregation setting.

The idea that comes closest to Nolle's vision is the Open Transport Switch (OTS), an idea being proposed by Infinera Corp. (Nasdaq: INFN) and other vendors. (See Optical Transport Gets an SDN Idea and Optical SDN Gets a Test Run.)

Marlin the flipper?
Another possibility would be for Marlin to flip its new optical company to Juniper Networks Inc. (NYSE: JNPR), according to Dana Cooperson, an analyst with Ovum Ltd. .

She doesn't know if that's in Marlin's plans. It's just that NSN has been Juniper's optical partner for a long time, and by some reckoning, Juniper needs to consider owning some optical networking.

"Tying the packet and optical accounts together is something Cisco has been doing, something Alcatel-Lucent has been doing, and something Huawei has been doing," Cooperson says. "If Juniper wants to become a full-service vendor, they might want to do something like that."

She's got two questions to go along with that theory, though. The first is whether the sale to Marlin includes the optical portion of NSN's services, a substantial part of its business. (She guesses it would be, but Marlin and NSN haven't specified that yet.)

The second is the state of developmental technology inside NSN. The company showed an R&D glimmer in October, claiming a fiber-optic speed record based on spatial multiplexing technology. But NSN, despite still having a worthy staff, has been lacking in other areas, such as OTN, she says.

"It's not clear how much real in-house technology they have, because they made themselves into -- not quite but almost -- an outsourced, buy-off-the-shelf play," Cooperson says.

— Craig Matsumoto, Managing Editor, Light Reading