The Lightwave Blog

Posted by Meghan Fuller

In his last post, Stephen talked about the mounting evidence to suggest that the 40G market is real and viable. I just finished an article about tunable optical dispersion compensation (TODC) for our November issue, and if ever there was a market that would reflect an upsurge of activity in 40G deployment, it's the TODC market. By all accounts, the market is more active today than it was just six months ago, and the vendors say they are readying, if not already shipping, their 40G devices.

Guy Martin, vice president of technology and business development at Teraxion, told me that of the 2,000 TODC units Teraxion has shipped this year, 80% are being used for 40G applications.

Civcom, meanwhile, has a single-channel 40G TODC device in the final stages of development. Yair Itzhar, Civcom's vice president of sales and marketing told me that the product has already been shipped to Tier 1 and Tier 2 customers for evaluation.

ANDevices has a 40G TODC based on PLC technology, but Wenhua Lin, vice president of technology and new product development, noted that her company's customers are now making more stringent demands in terms of the tuning range. Two years ago, 200-300 psec/nm would have been sufficient, but today they are looking for something closer to 400 psec/nm and higher. Lin confirmed that ANDevices is now working on an AWG-based device with a wider pass band and dispersion window.

From my perspective, this round of interviews yielded far more concrete answers to questions about 40G than a series of interviews on the same topic elicited back in March (See TODC vendors ready inline and integrated devices from our April issue), lending credence to Stephen's assertion that there will be a market for 40G.

Incidentally, I also asked the TODC vendors about the market for 100G and was surprised to hear Giovanni Barbarossa, CTO of Avanex, admit that he does not see much activity in this realm. That's news to me, as I feel like four out of every five phone calls I field has something to do with 100-GbE. But as Barbarossa pointed out, there's a big difference between 100G speeds and 100G capacity. If, as most people believe, the first generation of 100-GbE uses a parallel architecture, then the highest bit rate would be either 10-, 20-, 40-, or possibly 50-Gbits/sec, multiplexed for an aggregate of 100 GbE. And, as Barbarossa reminded me, Chromatic Dispersion becomes more problematic as bit rates scale, not capacity.

While that's probably an obvious distinction, I've been a little slow on the uptake lately. A few too many long nights spent watching the Red Sox come up with new and innovative ways to lose to the Indians? Perhaps. Or maybe--just maybe--I might be a little preoccupied because I'M GETTING MARRIED IN NINE DAYS.

Speaking of which, Stephen has threatened to bring his laptop to the wedding to provide Lightwave readers with a running commentary on the festivities. He may or may not have been kidding, so stay tuned!

--MJF