4 March 2011 3:31:32 PM
Twitter | LinkedIn | Facebook | Reddit
With OFC/NFOEC just around the corner, I thought it might be fun to guess at what the big news might be on the show floor. While predicting buzz can be a hazardous endeavor, the steady ramp up in press releases leading up to the big event, plus the program guide give pretty good hints.
Big again this year appears to be coherent detection, though not necessarily for the same reasons as last year. Coherent was a buzz word at last year’s OFC/NFOEC, primarily because it enabled 100GE transmission, and neither was quite ready for primetime yet. Fast forward a year, and now coherent is once again in the limelight, but not just because of 100GE. Over the last year, creative people have started to realize what the extra link budget means to system architecture. More recently, they have realized that the digital filter that allows recovery of the PM-QPSK signal also can be used for other digital filtering purposes. Consider 2500km with no dispersion compensation whatsoever, and how it would change the network landscape.
Also big this year will be “future proof” network designs, which is a euphemism for Terabit networks. While 1TE isn’t quite ready for main hall discussions, how to design today’s 100GE networks to support tomorrow’s 400GE and 1TE networks will be on everyone’s minds. It turns out that to make the leap to the types of modulation formats needed in the future, while still efficiently using the available fiber optic bandwidth, fixed wavelength grids will need to be abandoned. Fortunately, a new wave of gridless ROADMs are appearing on the scene that will allow just that.
Other news at the show will be partnership between router and transport vendors. The latest bandwidth trends show exploding consumption, falling average revenue per user, and increasing traffic volatility. To meet future network scaling, at lower costs, with greater peak to average traffic patterns, the individual network layers of the OSI stack can no longer be considered islands unto themselves. There must intelligent communication between all layers, most importantly between the routers and transport equipment. Recent partnership announcements are a great first step in improving network performance, and continuing to scale to meet future demands.
Finally, I fully expect “module wars” to continue, with the latest Betamax versus VHS (if you are old enough to remember, or iPhone vs Droid if you are not) debate being centered on what is best for 40/100GE client connections: 4x25G versus 10x10G for 100GE, and QSFP versus CFP4 for 40GE, respectively. In reality, it probably doesn’t matter, but we engineers love to endlessly debate topics such as these.
See you at the show.
Director of Technical Marketing
ADVA Optical Networking