And a lot of the established optical players are baffled by it, judging from interviews at the show. After all, the optical networking market is still considered to be overcrowded, especially on the components side. (See OFC/NFOEC: Mergers Haven't Gone Far Enough.)
But 100-Gbit/s long-haul networking brings up technological challenges some engineers and investors see as a good entry point. Startups in this area might be hoping to get acquired by a component, module, or systems company that falls off the 100-Gbit/s pace.
Here are some names Light Reading overheard during the conference this week. Got any others? The message boards are open, below.
What might be more interesting about Acacia is that Eric Swanson, a founder of Sycamore Networks Inc. (Nasdaq: SCMR), is on the board and reportedly helped the company get started, although he might not be there in a day-to-day capacity now.
Swanson may be best known as the star of a July 2000 Wall Street Journal article describing life at the heart of the optical networking bubble. Don't recall the article? Someone kindly posted it here.
Sycamore's glory days are long gone, but Swanson hasn't lost the aura, at least according to some. "He's a force of nature. Money sort of follows him around wherever he goes," one module-vendor executive tells Light Reading.
The Australian startup is using orthogonal frequency division multiplexing (OFDM) to produce high-speed signals from multiple, slower-rate streams.
Ofidium has been quiet lately, but one source thinks the company will resurface next year, when it's produced a module faster than 100 Gbit/s. The OFDM technology would be one way to get past that barrier, and for line-side links -- which don't have to conform to standards -- a proprietary 200-Gbit/s interface could be handy for aggregating multiple 10-, 40-, and 100-Gbit/s ports. Verizon Communications Inc. (NYSE: VZ) likes the idea, the source says. Ofidium couldn't be reached for comment.
Sources couldn't provide any specific company names, but more than one claimed such a startup exists and has been funded.
Meanwhile, established companies have taken up the challenge for themselves. Huawei Technologies Co. Ltd. and Nortel Networks Ltd. developed their own ASICs for coherent detection, and Opnext Inc. (Nasdaq: OPXT) has demonstrated its DSP algorithms for such a chip. (See Opnext Makes Its 100G Move and Opnext Develops 100G ADC.)
— Craig Matsumoto, West Coast Editor, Light Reading