As OFC/NFOEC approaches once again, I find myself reflecting upon the state of optical communications. Certainly, there is much to be enthusiastic about. The never-ending bandwidth explosion shows no signs of slowing, thus creating widespread demand across all optical industries, from 100GE for the core, pluggable optics for interconnect, and PON components for FTTx, to name just a few. And there is no better place than OFC/NFOEC to get a read on the status of these important optical technologies. While I would be foolish to make any predictions here, given the unpredictable nature of technology, there are certainly things I will be looking for in technical sessions, on the showfloor, and even in poster sessions.

I am confident there will be no shortage of 100GE demos. And while showing coherent detection of a polarization multiplexed, multi-phase signal may be the pinnacle of technical achievement, I find myself struggling with whether or not it is a sign of progress. Tier1 carriers do indeed need 100GE ULH that is able to run over the existing 10G core fiber infrastructure, as this saves them the cost building out a whole new backbone.  (Here's Verizon's release) Nonetheless, the technology required to do this has priced 100GE out of reach of the majority of potential users. So, I will be scouring the conference not for 100GE hero experiments, but rather meaningful work towards bringing cost-effective 100GE to the masses.  ADVA Optical Networking maintains close relationships with its customers, and we have heard the outcry for more cost-effective 100GE solutions loud and clear.  We are currently working on various application optimized 100GE technologies - which is just a fancy way of saying you will only need to pay for the performance you need.  (You can read more on the topic in this whitepaper.)

I am also certain vendors will be demonstrating pluggable optics of all flavors and varieties, with the buzz likely to be around CFP and CXP modules. While the names may be similar, the actual products couldn’t be more different. CFP is a brick sized module necessitated by the need to make room for all the fancy electronics needed for 100G-Ethernet. CXP, on the other hand, is not burdened by Ethernet, and is able to squeeze 120G of Fibrechannel and/or Infiniband into a module the size of a pack of gum. Again, are these new optical modules truly a sign of progress, or just latest in a game of speeds and feeds? Like SETI, I will be searching the show universe for signs of intelligence… in the modules; for, believe it or not, with all the fancy electronics in these expensive optical modules, they are still not “self-aware”. I think the time is long overdue that optical modules are able to be pinged and report the status of their link. This is one of the main shortcomings in optical links, as they seek to displace their electrical counterparts, as even the most basic electrical interface reports parameters of the link itself.  ADVA Optical Networking has developed Optojack® technology that allows pluggable optical modules to report back various performance parameters of their link.

I also expect to see PON everywhere, just as it has been the last few years. Certainly, PON continues to be a hot topic, and the leading protocols have been busy bumping up their data-rates, while leading vendors have been working on lowering component costs. (The March edition of Lightwave has excellent coverage of the transition).  Once again, however, I will be looking past the latest cost tweaks and bit-rate bumps for real sign of scalability. For, if I plot the growth in bandwidth consumption by the average consumer, it is only a matter of months, not years or decades, before Time-Division-Multiplexed (TDM) PON protocols run out of steam. Time-sharing of access bandwidth made sense in the early days of PON, when geographic penetration was low. However, with the high uptake rates and bandwidth usage models of today, dedicated access bandwidth is called for, and the industry is rushing to meet this new dynamic, with WDM-PON one of the leading contenders (though not the only one). So, I will be looking for new techniques in Dynamic Bandwidth Allocation (DBA), Next Generation Access (NGA) architectures, and technologies that help bring a dedicated 1Gbit/s to home and desktop.  ADVA Optical Networking is at the forefront of WDM-PON technology and architectures, and you can read more about it in this Lightwave article.

Finally, while the transition from copper to glass hasn’t quite yet occurred, that hasn’t stopped pundits from continuing to proclaim “it is just around the corner”. The latest attempt “Lightpeak” is from Intel, and consists of an optical version of the ubiquitous USB that hopes to leapfrog USB3.0, also known as “SuperSpeed”. Marketing lingo aside, Lightpeak is rumored to be backed by the two largest names in consumer electronics (names withheld). In the end, it will all come down to how well the interface is specified, execution, and above all else, cost. After all, these days component costs are measured in pennies, not dollars.

Never before (not even pre-bubble) have I seen this level of optimism. There is a sense that now is the time for optics to really grab the opportunities that are available. If everyone can just work together, as a team, as an industry, and properly define the protocols and interfaces with a close eye on cost, perhaps glass will one day better known for communications than windows and glassware.

You can follow the latest updates, and pre-show commentary, at the OFC/NFOEC Conference Blog.  See you in San Diego!