Ciena (NASDAQ:CIEN, news, filings) announced today that they successfully demonstrated a 100Gbit/s connection at the SC08 show a week or two ago, informally joining a rapidly growing 100G club. As expected, much was made of the single wavelength, and indeed this was a great achievement! The industry as a whole is making great strides toward a commercializable 100G technology, although that goal remains pretty far off yet. Right now it’s all at the ‘can we do it?’ stage rather than the ‘can we do it affordably in a production network?’ stage. And the answer there is obviously ‘YES’! Now all we need is some standards and we can talk products.
In this difficult economic environment it is encouraging that R&D like this continues to proceed steadily. Everyone is taking a different approach to this. Infinera (NASDAQ:INFN, news, filings) has concentrated on using 10x10Gbps to do 100G. Others have looked at two 40G waves and making up the difference with a few 10G waves. But of course, purists will always prefer the single wavelength solution that Ciena demonstrated, which is just cleaner: more science and less engineering, with fewer but more expensive parts. It still seems clear that the multiple wavelength approach will be economical first just because it has fewer technological hurdles to jump. But the time will come quickly when 100G on a single wave will be needed to meet the demands of the future.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Telecom Equipment
Join Our LinkedIn Group
Have you ever noticed how incrementalism leads to greater and greater levels of bulk and engineering complexity for the sake of not having to bite the bullet once? In the premises LAN and data center cabling arenas we see the same thing occurring.
Twenty some-odd years ago LANs were relegated to using multimode fibers, some of them with twenty-year warranty terms. Nevertheless, every six or seven years even MMF needs to be torn out and upgraded, each time the next power of ten is achieved in Ethernet (first to OM-2, then OM-3), and today to the point where today OM-4 grade MMF is being prescribed for 40 and 100 G in the LAN and data center. All the while the MMF designs may still demand using multiple fibers, or inverse multiplexing, meaning ten times as many electronic drivers and detectors.
The reasoning, by now impertinent, but nevertheless still ruling, is that, each of those MMF drivers is still individually cheaper than a single set of electronics to satisfy a SMF design. And so it shall forever be until someone with enough clout decides it’s time to break the nut and begin developing and leading SMF solutions down the path to mass production and the ensuing economies of scale required for COTS sourcing.