Optical/IP Networks

PhotonEx Set to Demo 40 Gig

Startup PhotonEx Corp. is about to make a giant leap of faith. Today, the company plans to announce that it is working on an optical transport system based on dense wavelength-division multiplexing (DWDM) that can handle 80 wavelengths each carrying 40 Gbit/s of capacity, providing a total capacity of 3.2 Tbit/s. It plans to ship the product by the fourth quarter of this year.

If that isn’t enough to grab your attention, the company is planning a follow-up announcement for next week’s Supercomm 2001, a telecommunications trade show in Atlanta. At that time, PhotonEx is expected to announce what it claims to be a record for the length and bulk of a 40-Gbit/s optical transimission by sending 16 channels at 40 Gbit/s over an unregenerated span of 1,500 kilometers. (The test had 100GHz channel spacing and 0.4 bit/s/Hz spectral efficiency.)

Normally, lab trials are interesting — with the caveat that they don’t necessarily portend commercial deployment. PhotonEx officials note, however, that they have achieved such results with commercially available components and plan to ship the product by the end of the year. The company has not disclosed its customers’ names.

Larger companies, including Alcatel SA (NYSE: ALA; Paris: CGEP:PA), Lucent Technologies Inc. (NYSE: LU), and Nortel Networks Corp. (NYSE/Toronto: NT), have all tested 40-Gbit/s products at major carriers, the longest of which was Nortel's demonstration of a 40-Gbit/s transmission over a span of 700km (see 40-Gig Forecast). The fact that PhotonEx, a startup, is promising results and a delivery schedule that appear to outpace the larger companies is certainly impressive. But there’s one major question for the company: Is the market ready for 40 Gbit/s?

Conventional wisdom indicates that carriers won't be ready for 40 Gbit/s for at least a year — and it will probably take longer for the market to develop (see 40-Gig Forecast). Telecommunications carriers are currently in a world of pain, much of it brought about by an aggressive investment in core optical transport that has yet to generate the revenue to pay for this investment. PhotonEx’s technology debut comes as carriers are scaling back on capital spending and focusing on metropolitan-area and access networks, where they can link up to real customers.

PhotonEx officials have an argument: They say that part and parcel of their product architecture, called the Ultrafast Dynamic Core, is flexible provisioning, in which individual wavelengths can be split up and parceled out to customers, thus making more efficient use of the pipes. Currently, the limited capability of carriers to provision small pieces of their high-capacity systems discounts much of the economic gain of adding capacity to the system.

“As more and more capacity gets built into the network, carriers are not seeing the high-margin network,” says Kristin Rauschenbach, president, CEO, and cofounder of PhotonEx. “They’re forced into trading the economics of long haul with the revenue potential of high capacity. Optical lambda mesh architectures are going to suffer from a lack of flexibility.”

PhotonEx says its architecture will be more efficient and “dynamic” by splitting the wavelengths into “flexible streams." How? Company officials aren’t saying, other than the that it will somehow employ technology based on the emerging generalized multiprotocol label switching (GMPLS) standard.

PhotonEx officials also won’t say how “granular” their flexible streams are — that is, the size of the smallest unit of bandwidth than can be handled by the system. This is important because the success of many recent optical switches, most notably the CoreDirector from Ciena Corp. (Nasdaq: CIEN), has been guided by granularity of provisioning. A switch that handles smaller units of bandwidth can pack more customers into individual transport pipes.

At any rate, PhotonEx’s initial technology debut makes for some interesting spectating. The company is not short on resources or expectations. It has already drawn $88 million from notable venture capitalists (see Photonex Has a 40-Gbit/s Idea). Much of the expectations are provided by PhotonEx’s core scientific talent, which includes Rauschenbach and fellow cofounder and CTO Katherine Hall. Both Rauschenbach and Hall come from the Massachusetts Institute of Technology’s Lincoln Labs, a breeding ground for cutting-edge networking technology.

— R. Scott Raynovich, Executive Editor, Light Reading
http://www.lightreading.com For more information on Supercomm 2001, please visit the Light Reading Supercomm 2001 Preview Site.

realguy 12/4/2012 | 8:21:08 PM
re: PhotonEx Set to Demo 40 Gig While claim of 1500 km (in the privacy of your own lab)is indeed impressive, the distance may in fact be misleading. was this over old high dispersive fiber (like old corning smf-28) or over modern non linear tolerent and dispersion slope corrected fibers? And of course a rule of thumb says, divide the number you got in the lab by at least 5 to get the better idea how far you really can send it in real deployed fibers.

glenda 12/4/2012 | 8:20:55 PM
re: PhotonEx Set to Demo 40 Gig Lucent's experiment where they sent seventy seven channels of 40Gb/s (total > 3Tb/s) over a distance of 1200km sounds more impressive than this claim. (See OFC post deadline submissions.)

Can someone explain the hoopla about PhotonEx's experiment?
sundevil 12/4/2012 | 8:20:48 PM
re: PhotonEx Set to Demo 40 Gig glenda,

I can't seem to find the Lucent paper. Do you have a link?
minivan 12/4/2012 | 8:20:45 PM
re: PhotonEx Set to Demo 40 Gig ***

I can't seem to find the Lucent paper. Do you have a link?


Post deadline paper PD23

BBBoa 12/4/2012 | 8:20:43 PM
re: PhotonEx Set to Demo 40 Gig Certainly, their marketing department did a good job of catching our attention. I wonder what the realities are behind this lab test. I seriously doubt they'll be able to directly translate their results to real-world network deployment.

In the real world, carriers are not going to deploy this technology for several years if it means they have to replace the fiber they currently have in the ground. Several carrier networks (including mine) are not ready for 40G simply because we're not at capacity with our 2.5/10G networks. Even when we hit our peak traffic loads later this year, there are some bandwidth efficient alternatives that don't require fiber replacement. With the moratoriums on digging up streets and neighborhoods, we need to look at other alternatives that enable bandwidth growth while utilizing our current infrastructure.
Sign In