Article Info

Death of the All Optical Network – Gilder Telecosm 2006


Infinera took the bold and stunning risk of angering the Gilder priesthood by illustrating that an all optical network was not the future and not the best solution.

As a comm semi guy who will always have a soft spot in his heart for TDM circuit switches and electrical crossconnects, it was a message that resonated with me. But to the assembled congregation, still clutching to a vision of an all optical future, it was analogous to Galileo arguing that the sun did not revolve around the earth.

JDSU (JDSU) has made a concerted move away from the optical module business and towards subsystems like ROADMs and EDFAs. It is even rumored that JDSU has lost future 10GE module contracts at Cisco (CSCO) (this is single-sourced, any confirmation welcome) so they may have passed the point of no return. JDSU pushed hard during the panel that the future was indeed all optical, and that they were going to provide the building blocks to do it.

Intel (INTC), as I suspected, may be exiting the commodity module business, buy they have tee-ed up a big effort to do pure optical research. The UCSB/Intel announcement was just a small step, even though it received a disproportionate amount of media attention. Kevin Kahn opened my eyes to what Intel is doing outside of the commodity optical products that I used to solely associate them with. This is worth following.

Infinera laid out why O-E-O makes sense and that they have the most efficient solution as a result of their ‘optical chip’.

I remember a meeting with Infinera (Actually called Zepton before a savvy marketing guy changed the name of the company) way back where Drew Perkins laid out an array of optical modules on the conference room table. He pointed at each one,  and said that they were way too big with some very colorful language.

The vision at Infinera appears to be the same as it was then. High integration creates high reliability, high port density, and low per port cost.

I think people were mortified to find out that Infinera, the poster boy of Telecosm 2.0, is not an all-optical player This illustrates the depth to which people have gone to really understand what this company is doing, and I realized I didn’t know what sort of electrical switching function resides inside their system. I had fallen under the spell…

The best part of the discussion was the limited debate between the merits of the ROADM and O-E-O. This was a good panel, with excellent participants and should have been allotted more time simply because much time was needed to explain the differences between what Infinera does with optical-electrical-optical (O-E-O) conversion and how ROADMs work. Some carrier input, particularly from a non-Bellco perspective, would help.

This topic will be just as relevant next year and should be revisited.


(Italics are my comments. Plain Text is my transcription) If folks have changes/corrections by all means add them into the comments- I am not a reporter by trade.


  • Kevin Kahn, Senior Fellow; Director, Communications Laboratory, Intel
  • Michael Ricci, Senior Vice President, Optical Communications Group, JDSU
  • David Welch, Chief Strategy Officer, Founder, Infinera


  • Frank Galuppo, Chief Executive Officer, Amedia Networks
  • Jerry Rawls Chairman of the Board, President and Chief Executive Officer, Finisar


  • Charlie Burger, Senior Technology Analyst, Gilder Technology Report

Mike Ricci – JDSU

Things are different now (Uh oh, hide your wallet when as an investor you hear anything like this). We need an optical network now, we didn’t need one 6 years ago.

ROADMs, Tunable Optics, and EDFAs are the three critical components that provide everything you need to do the all optical network. (What about a large sale optical switch? You can’t build them out of ROADMs very well) 9000 ROADMs deployed so far.

Kevin Kahn – Intel Photonics (Sitting next to me, very interesting guy)

Runs Intel Enterprise Interconnect Research. He was the guy who oversaw the work surrounding the recent Intel announcement. I applaud his organization for garnering so much public attention for what was really a heterogeneous material bonding innovation. It was a good marketing job.

JDSU is focusing on what building blocks are needed to do silicon photonics. Five key areas of focus – Modulator, Laser, Photodetector, Light Guides, Low Cost Assembly (I.E. no active alignment, more like wire bonding from the chip packaging business). Silicon is infrared transparent, so the material could be used. But it can’t do 850nm… No real progress until in silicon photonics until 2002. Kevin then walked through a neat timeline of silicon innovation milestones. Good stuff.

Dave Welch – Infinera – Chief Strategy Officer

Infinera believes that optical integration drives out cost, primarily by integrating all of the O-E-O functions into a small module.

Endgame is eventually to integrate L2/L3 capabilities right on to the transponder. Some confusion among the audience, including the Intel photonics guy in particular (who is siting next to me). Do they actually do L2 monitoring and aggregation among lambdas? I don’t think so. Another sharp guy at my table says it is a big STS-1 crossconnect. I’m starting to realize that I don’t really know what Infinera does, and may be under a marketing spell.

The Achilles heel of the ROADM is the inability to re-color wavelengths – ROADMs only switch the current wavelength. O-E-O still needed in order to change wavelength colors and perform basic traffic switching. If you don’t do this you will strand bandwidth.  This has a rather interesting reaction from the Gilder priesthood, where wasting bandwidth is the mantra. This should be interesting to watch.

Q & A

Some confusion – Charlie the Moderator thinks WDM is dead. Infinera goes through the process of illustrating how ROADMs result in stranded wavelengths. O-E-O is still valuable. He handles the explanation well. This guy is an exceptional presenter to the laymen.

Infinera’s equipment is being used throughout Infinera’s network. Gilder appears to be surprised that carriers are still deploying OEO conversion… seems like there is a bias against O-E-O conversion among the moderator/GilderIt is clear that the concept of an all optical network is such a central Gilder theme that they are having a hard time letting go. Ricci from JDSU still pushing that all optical is the future. Perhaps Infinera has now angered the Gilder gods….

Gilder asks if optics were cheap enough, would things be all optical? I think he is trying to see if the state-of-the-art is an intermediate step to optical Valhalla. Infinera guy makes point – “Future is packet based network, so I must look at the bits.” There will be O-E-O.

Gilder trying to incite a little rivalry between Infinera and JDSU O-E-O and ROADM. Who is cheaper. Who will ship more. Who can displace the other. Panelists aren’t biting. Its clear that both have a role, but I applaud the effort to keep the debate rolling!

Question: 40G vs 4 x 10G – which one wins? Infinera: 4 x
10G. 100G will be parallel first. JDSU: Increasing pressure to supply 40G (Yes of course there is pressure but what is the volume)

Question: What about the cable companies? What are they doing? (Good question, people are sooooo Bellco focused) “Yes they are.” But no information on how they are different, if at all. More representation from cable guys is needed next year – I would bet they would side more with the JDSU vision.

Full Disclosure: I hold in Intel and have a short position (hedging) in Cisco

This post is one of a series as I blog the Gilder Telecosm 2006 conference. All posts can be found by searching for ‘Telecosm 2006’.


Comments are disallowed for this post.

  1. A Concept Company: Building a Sustainable Business or a Profitable Exit for Investors?

    Answer these two questions: (1) why would anyone start a business in the space of optical transmission products in 2001 and (2) how do you obtain a positive return on $290M of venture capital?

    The answer to the first question is that you believe that all the spending that occurred in the prior years of optical networking will in effect cause the industry to lie dormant for a few years. During this dormant time, you will be able to commercialize a next generation of breakthrough technology products that when the industry begins to emerge from a dormant state, the company will be in a position to capture the new dynamic market share. This assumes that the existing competitors do not invest in optical R&D, the number of actual competitors decline and the market for optical systems does not become commoditized. Those would be the assumptions in 2001. It is now 2006 and we can begin to assess what five (5) years and twenty (20) quarters of work has produced.

    How has the market for optical systems evolved?
    The consolidation of the service provider industry coupled with the demise or spending decline of surviving tier 2 and tier 3 service providers has actually created a more challenging market structure. Many European and Asia-Pacific service providers have returned to being large, hybrid service providers that dominant their market with little if any real competitors. The good news is that service providers are making investments (e.g. CAPEX) in their networks; the bad news is these investments are going to the large system companies (e.g. Cisco, Nortel, Lucatel, Ericsson). An additional challenge for Infinera is that it is unlikely that tier 1 service providers (e.g. at&t, Verizon, BT, VT, DT, CTC, NTT) will purchase products from Infinera due to their size, sustainability and the need for the tier 1 service providers to support their incumbent suppliers and the billions of dollars of installed equipment. This raises the question: Can a big enough ($200M+ annual run rate) profitable business be built from tier 2 and tier 3 service providers to take a company public?

    Did the other optical system suppliers hibernate as anticipated?
    Yes and no. Yes in the sense that Nortel, Lucent and Cisco all reduced R&D investment in optical systems, but no if you consider Alcatel, Marconi, Ciena, ECI, Huawei, Tellabs, ADVA and Siemens. These companies plus a host of smaller startups like Meriton, Transmode, Lumentis and others in Europe continue to innovate and keep the competitiveness of the optical system market high.

    Does the market for optical systems sustain high profit margins?
    No. There is no evidence that optical systems are returning the glory days of 70 percent GMs or even reasonable product margins in the 50s. On the eve of Ciena’s IPO, Sprint order $90M worth of WDM products at 70 points of gross margin. That was 1997, this is almost 2007. The optical market on a global basis is nearly hyper-competitive to use a 1990s bubble term. No company has a sustainable market advantage that provides a technical or cost advantage that is compelling enough for end-users to pay a higher price. All suppliers of optical systems have moved their supply chains to Asia-Pac and the emergence of e-auctions by service providers has produced a market structure closer to the Wal-Mart model then the Bell Labs/Bell System model. Market forces and competitive pressure have marginalized the business of optical systems.

    Are there fewer suppliers of optical systems?
    Yes, but the bad news is the companies that have survived will fight even harder to win the business. There is simply not enough business to provide a healthy business for all the survivors.

    How did Infinera build a business?
    Connections and relationships are the answers to the question. Infinera is a concept company. This applies to their technology and their business plan. The founders of Infinera were the founders of a service provider named OnFiber Networks. When the service provider market crashed in 2001, they started Infinera. The concept of Infinera was to create photonic integration on chip and then apply the concept of Moore’s Law to photonic integration. The whole concept sounds fantastic – but in reality Moore’s Law has little to do with photonic integration, but the extrapolation and marketing spin does provide for a fantastic branding concept for Infinera. Moore’s original concept that was the genesis for what would become his law was published in 1965 as: “Cramming more components onto integrated circuits”, Electronics Magazine 19 April 1965.

    In terms of building a business, Infinera has marketed their concepts of “integration, cost break through, following Moore’s law” to service providers competing in a price sensitive market. How do you do that? You call on your contacts to win business. One of Infinera’s initial customers was OnFiber. Yes, the same OnFiber that Jagdeep Singh and Drew Perkins, present day CEO and CTO of Infinera, founded. The next step is to put your VCs and Board of Directors to work. This is where men like Vinod Khosla and Hugh Martin become invaluable. Hugh was CEO of ONI Systems and Vinod who was the founding CEO of SUN has become a legendary VC who helped start companies like Cerent, Juniper and Siara. In 2003 Forbes magazine named Vinod Khosla “The Man with the Golden Touch.” In 1999, two of his best optical companies Cerent and Siara Systems were acquired for staggering total of $11.2 billion. Vinod is also a former board member of Qwest. In addition to deploying the contacts at the board of director level, the next step is to acquire a sales team with deep relationships within the tier 2-3 service provider market. The best place to find that is hire the sales team of one of the most successful companies to sell into that market during the tech bubble of the 1990s. A company that does not find its heritage in the Bell System, a company that is a true rogue to the state sponsored world of telecommunications. That company would be Ciena who built a legendary business within the tier 2-3 service provider market. What does this all lead to? It leads to sales and Infinera has been able to generate substantial business within the tier 2-3 service provider market. The question is where is this business and $290M of private equity leading?

    Who would buy Infinera?
    Most of Vinod’s optical startups where acquired by public companies and the current CEO of Infinera had his last optical company Lightera acquired by Ciena. The natural question is who is going to buy Infinera? This is a really a difficult question to answer. The Lucent/Alcatel merger takes two active acquirers during the tech bubble off the market. In addition, the recent disappointing sales results at Lucatel means they have bigger things to worry about then photonic integration. When you are running an annual business of $26 billion dollars, $100M contribution is nice, but is does not change the world. Nortel is still challenged with litigation issues. Cisco has downsized and reduced the emphasis of the optical business unit. Ciena is not in the market for a company like Infinera. Fujitsu does not buy US technology companies. Ericsson bought Marconi. Siemens and Nokia are out the acquisition business with their new JV. Infinera is not in the core business of Motorola. Tellabs is focused on access, but it might be a potential suitor. Sycamore would probably want to buy Infinera, but it would cost too much and they are trying to figure out how to sell pipe organs. That leaves ECI and the Chinese companies of Huawei, UT Starcom and ZTE. Infinera would be too expensive for ECI, ADVA cannot afford them and when was the last time one of the Chinese companies went on an acquisition spree for US companies? It does appear that the M&A window is closed for some time.

    Can Infinera complete an IPO?
    It is not so much a question of can Infinera go public, but rather can they build a business in the tier 2/3 service provider market that is attractive enough for them to achieve a public listing. The present conditions to achieve an IPO are difficult to overcome. The IPO process is a huge undertaking in the current market for the following reasons. (1) The NASDAQ conversion from fractions to decimals has eliminated the profits required to fund research for mid-market companies. Analysts and investors cover and trade only the large caps because no company can afford to cover or trade below the large caps. (2) The agreement that separated research and banking has hurt the small companies who want to go IPO. The small companies are suffering because profits cannot be made taking small companies public – even if they are of high quality. (3) There are 100 companies at $100M+ in revenues that could go public in the next 12 months, but the barrier is still very high and with no analyst coverage or market makers, it is hard to create demand. (4) Private companies that have been building their business over the past 4-5 years under estimate the world of SOX and VSOE compliance. SOX compliance has completely changed the culture of technology industry. The method by which a company sold products/technology in the ‘80s and ‘90s is no more. Then there is the recent news which has really tainted the world of technology. These issues are: stock option backdating, short selling scandal, Lucatel results, Nortel results and profit levels of public companies in markets that are considered stagnant. The overall health of the technology industry is good, but companies are not executing as well as expected and there is a lot of turbulence in the overall market.

    If Infinera did complete an IPO, where would that leave them?
    The good news is they will have raised somewhere between $150M and $250M of working capital. Recently they raised $110M of PE money, bringing their total to $290M. An IPO of Infinera will provide them the working capital to run their business, but it is doubtful they would raise enough money to build a big business. If the overall optical systems market does not support rapid growth, how does a public listing help? An IPO would raise capital – but it would also increase the cost of their business model. The cadence of the company would change to a quarterly reporting process, SOX compliance as well as removal of the attractive branding value of a concept company. The real objective at Infinera is to be bought by a big company. This is the best exit plan for the investors. A public listing is the best option for the employees – but this would remove the attractiveness of a concept company, open their business to public scrutiny and make them subject to onerous compliance requirements and they would probably have a 12 month lock-up on stock options.

    W. R. Koss

    Posted by W. R. Koss | October 9, 2006, 11:41 AM
  2. Andrew,
    You stated the following above: “Intel (INTC), as I suspected, may be exiting the commodity module business”

    Was this based on comments from Kevin Kahn? What external input toggled the “as I suspected” in your statement? Thanks.

    Posted by Photon Guy | October 9, 2006, 4:23 PM
  3. That’s a very important clarification. Kevin said nothing to this effect and I did not ask him (because he couldn’t tell me if he did know). This was based on talking with others… and the general consensus was that this was a likely outcome.

    I’ve written about this in the past and have a bias toward this event.

    Thanks for bringing this up.

    Posted by Andrew Schmitt | October 9, 2006, 4:32 PM
  4. In a price sensitive market there are two basic questions to ask:
    – Is a benefit to focus on O-E-O (including cost) in building a network infrastructure, for a given service?
    – and for which granularity the cost of integration is more beneficial? Established the granularity of the traffic matrix, what is the cost trade-off of ROADM vs O-E-O? We all know that the marketing message is a perception of the reality, therefore is the high integration at the stage to enable a replacement one-to-one with the current optical technology (80 chs, 50GHz), maintaining a fix cost for the manufacturing? The answer is not today. High integration with low cost per port is a good opportunity today that enables all companies to think about this new topic on the same line.
    Dr Rodolfo Di Muro, PhD, CEng IET, EMBA, MIEE

    Posted by Rodolfo Di Muro | October 10, 2006, 5:56 AM
  5. A company called Luxtera has developed a CMOS based XFP module. CMOS is much cheaper, easier to manufacture and scale compared to InP which is being used by Infinera. Assuming Luxtera has the technology, this could become a major problem for Infinera. Luxtera is a pure components company, and they do not have the high burn rate of Infinera. A number of optical system vendors like Nortel, Ciena, Cisco and others could buy the cheaper transponders from Luxtera. This would create a major problem for Infinera as their key advantage of cheap transponders will be negated. What do they do then? Maybe they should have stayed an optical component company and not become a systems company.

    Posted by lpskeptic | October 19, 2006, 12:57 AM
  6. Interesting read. I bumped into this blog just surfing around on other issues. The OEO/OOO debate, I think, was most eloquently summarized by one of our private equity backers here at Tropic many years ago (he is a PhD in Physics):
    “Bosons for transport, leptons for processing!”
    his rationale being that bosons (of which photons are one type) don’t like to interact much with each other or what is around them, (so they can go the distance for transport) whereas leptons (of which electrons are one type) love to interact (so those interactions can be used to create processing elements). It is hard to argue with mother nature, and her offspring, Physics.

    The issue really is about how much processing does one really need.
    Given all the historical efforts at doing traffic engineering for IP, and the sheer volume of transit traffic (ie traffic that needs to just pass through with little or no local processing) that exists all over the current networks, both IP and transport, it seems fair to say we need both: transport for distance and when you know where you are going, and processing, where routing decisions need to be made a granularity smaller than “a wavelength”. At Tropic we can’t make the OEO economics work when we analyze the real network problems our customers have (no matter HOW good we make those huge-die InP yields look).

    With respect to technology choices (eg InP vs CMOS Si, etc) again you hit the nail on the head. At Tropic we made a decision NOT to go vertical, precisely because of the huge number of people innovating in photonics and electro-photonics already, the highly competitive supplier market that was emerging, and the potential that anything we did internally would get leapgfrogged in short order (aside from arguments centered on the higher capital requirements for vertical integration).

    Bottom line: This blog seems to have nailed it. Infinera has a great *story*, has some great insiders helping it along, but when you look closely at His Majesty’s garb, it does look rather threadbare, where clothed at all.

    I would like to therefore coin a new acronym: AON: the “Appropriately Optical Network”.

    Check the Tropic site for the alternative to OEO everywhere. We build managed, reconfigurable and fully automated photonic networks without *ANY* OEOs, if that is what is appropriate – “Transponderless”, “alien wavelengths” etc. Photons in/photons out, no electrons between, but we will do high-quality OEO if needed and/or if the client device does not support integrated DWDM transceivers, which, as new vendors emerge for DWDM GBICs, SFPs, and XFPs, is getting to be a tenuous position to be in… roughly half our current installed base (in field operation since early 2004) has OEO, half does not, but launches DWDM directly at our “AON”…

    Thanks for shedding some of the light of truth (IMHO) on Infinera…
    Ben Bacque
    VP R+D
    Tropic Networks Inc
    613 270 9660

    Posted by Ben Bacque | October 30, 2006, 7:09 PM
  7. So, from a lame persons perspective, should we assume that Luxtera’s technology has more promis that Infinera? Thank you Jack

    Posted by jack | March 28, 2007, 1:29 PM
  8. Trackbacks / Pingbacks

  9. WDMBlog » OEO Switching vs. ROADM | October 9, 2006, 12:13 PM
  10. WDMBlog » A Contrarian View of Infinera | October 11, 2006, 8:28 PM
  11. Infinera - The Optical Component Company That Wasn’t at Nyquist Capital | June 8, 2007, 8:49 AM