OIF Steps In to Fill the 100G Void

We were grateful to have the Optical Internetworking Forum (OIF) as a workshop sponsor at this year's Optical Expo in early October. Listening to the OIF representatives speak, I realized the critical role that the OIF has to play in advancing 100-Gbit/s technology forward.

Readers may recall that the OIF was formed 10 years ago with a mission to advance the cause of optical switching technology – specifically, developing a user-to-network interface (UNI) for control-plane signaling interoperability. It was very successful in its control-plane interoperability efforts, but with its major UNI work completed, the OIF stepped into the background a bit. The group remained active in the control plane, and branched out as well, but for the most part, progress was routine.

Now, as vendors are hard at work building 100G transponders, it seems the OIF is badly needed once again to fill the significant gaps that are being left by the major 100G standards bodies. On the one hand, the Institute of Electrical and Electronics Engineers Inc. (IEEE) is building the 100G Ethernet client-side interface, which is the interface to wide-area network equipment. On the other side, the International Telecommunication Union, Standardization Sector (ITU-T) Study Group 15 is standardizing the wide-area transport of that 100G traffic, which will be carried by a new version of Optical Transport Network (OTN), called OTN4.

The IEEE and ITU-T standards efforts are critical, certainly, but they are also incomplete. Neither the IEEE nor the ITU-T is chartered with standardizing things such as modulation formats or transponder form factors. The problem with 100G is that operator demand took the industry by surprise, leaving vendors and component suppliers to scramble on their own to build products quickly and resulting in a number of uncoordinated efforts. Uncoordinated efforts lead to higher product costs, which slows and stalls technology adoption. Fortunately for all, the OIF has stepped in to fill the 100G standards gap and drive an industry consensus that would otherwise be lacking.

Specifically, the OIF has launched three major 100G initiatives to watch:

  • 100G Long Distance DWDM Transmission Framework: Specifies dual-polarization quadrature phase-shift keying (DP-QPSK) as the sole modulation format, among other things
  • 100G Long Distance DWDM Integrated Photonics Components: Provides the multisource agreement (MSA) foundation for transmit-and-receive photonic components
  • Forward Error Correction for 100G DP-QPSK Long Distance Communication: "Standardizes" FEC specification for 100G using the DP-QPSK modulation format

The OIF 100G initiatives – like 100G itself – are in early days, but the OIF's early recognition of the need for standards beyond the standards is encouraging; and, if these efforts proceed on schedule, they will be good news for components and chip suppliers, equipment makers, and carriers wishing to deploy 100G.

— Sterling Perrin, Senior Analyst, Heavy Reading

^Eagle^ 12/5/2012 | 3:28:10 PM
re: OIF Steps In to Fill the 100G Void after all, without the tunable MSA, where would Bookham be now? I recall that the LIghtreading web site recently posted some good financial news for Bookham.. wow.. finally a profitable quarter or 2 with a backlog of orders and demand for tunable lasers. Without the tunable MSA, does anyone believe Emcore would have purchased Intel's optical group? There are other examples.

^Eagle^ 12/5/2012 | 3:28:10 PM
re: OIF Steps In to Fill the 100G Void Note: while OIF in some of the sub-committees may have "slowed down" or "faded", I must point out that OIF remained active in several areas during the last several years. One area that could be pointed to is the OIF sub-group who have developed the various tunable laser MSA's: the original TLM, then the ITLA, and now the ITTA. I believe this standards work made a significant impact on the industry. Without this work, tunable lasers would not have taken off or been as straightforward for systems or transponder houses to utilize. Today, tunable lasers are the defacto choice, in fact requirement, in every RFP or RFQ from major carriers for optical transport systems in the core including metro, long haul and ultra long haul. Without the tunable laser MSA, switching, lambda routing and ROADM's would not work as effectively and the basic network design used by nearly everyone would be much more complex.

Maybe OIF has not made many splashy headlines, but they continued to do important work over the last few years prior to the recent 100G effort.


Sterling Perrin 12/5/2012 | 3:28:09 PM
re: OIF Steps In to Fill the 100G Void Sailboat,
Thanks for this post.
I agree with your comments. The point I was making was that they faded from the spotlight (ie, fewer flashy headlines), which does not indicate at all that they slowed down. I believe the 100G work puts them back in the spotlight, and for good reason. The question I wanted to answer/clarify in this column was: Given that two global standards bodies are cooperating on building 100G standards, why does the industry need an industry group like the OIF?


teleliar 12/5/2012 | 3:28:09 PM
re: OIF Steps In to Fill the 100G Void The article missed the 4th OIF initiative related to the 100G.

100G Long-Haul MSA: Specifies an Implementation Agreement including definition of mechanical dimensions of an optical line interface module and its mounting holes, definition of the electrical connector, maximum power consumption and address communication interface.

^Eagle^ 12/5/2012 | 3:28:04 PM
re: OIF Steps In to Fill the 100G Void Thanks Sterling,

I agree with your view. 100G is coming faster than 40G and will be rolled out into various applications far faster than 40G was. Lots of reasons and lots of confirming information from my sources. I think it is actually good that OIF recognized the missing parts of an end to end 100G spec and stepped up. It will help everyone I think.

However, one thing that catches my attention is that there is no single body that will own the end to end specs. So, while better than previous standards processes (all nominally 100G....ethernet and so in theory the same framing and bit rate.. the reality is each section of the span will use somewhat different ways of implementing 100G...), so there will still will be some level of incompatibility across the 3 domains that IEEE, ITU, and OIF are looking at, so still expensive silicon.

sigh. Sure would be a wonderful world if we could ever get it really straight forward and make a clean end to end spec. I know that is dreaming, but a person can wish can't they?
keithr 12/5/2012 | 3:27:54 PM
re: OIF Steps In to Fill the 100G Void OIF "was very successful in its control-plane interoperability efforts". Has any operator ever deployed this ?
Sign In