OMM-inous News

Earlier this year, everything seemed to be going right for OMM Inc.. It had filed for an IPO, launched a big expansion plan, and was happily bragging that it would be first out of the gate with a rack-based optical switching subsystem based on 3D MEMS (micro--electro-mechanical system) technology (see Agere's 3D MEMS Switch 'Not First').

Now everything has gone into reverse, Lizzie Grubman-style. The IPO filing has been withdrawn. The company has been shedding staff. And today, OMM announced that it’s discontinuing its 3D MEMS development and laying off another 100 employees (see OMM 86s 3D MEMS). Its workforce now totals 220, down from 400 at the beginning of the year.

OMM is putting a brave face on it, pointing out that its smaller 2D MEMS subsystems are still the only ones to have passed grueling environmental and reliability tests (see OMM Gets a Breather) and that these have been shipped to no fewer than 28 customers.

As for bigger 3D subsystems, “the market’s much further out than we thought –- maybe two or three years away”, says Conrad Burke, OMM’s senior VP of marketing and business development. He says this only dawned on OMM recently (an epiphany, like).

The real problem for OMM, and for all the other startups developing optical switching subsystems, is that it’s almost impossible to say when the market for any of their products will take off. And that applies not only to 3D subsystems, which are seriously challenging to develop, but also to the much simpler 2D subsystems of which OMM is so proud.

The big problem is that subsystem manufacturers like OMM are at the other end of the food chain from the folk that actually determine market requirements -– the service providers and their customers.

OMM has managed to sell a relatively small number of their 2D boxes to 28 system manufacturers, so they can integrate OMM’s boxes into their prototype kits. After that, the system manufacturers have to persuade carriers to evaluate their prototypes in their labs and conduct field trials before they stand a chance of getting a multimillion-dollar contract. Then, and only then, is OMM likely to receive any significantly sized orders (see Components, Anyone?).

In OMM’s case, Burke says that three of its 28 customers have got their switches into carrier trials, although he can’t tell whether that’s lab trials or field trials.

One of those customers, by the way, is Ilotron, the U.K. startup that got into financial trouble earlier this year. Most of Ilotron’s assets were acquired by Altamar Networks (see Altamar Buys a Bargain), and Altamar isn’t planning to continue development of Ilotron’s switch, according to Ian Wright, its senior VP of engineering and CTO.

In other words, only two of OMM’s customers are anywhere close to selling their switches to carriers, and they may not be that close if they haven’t actually started field trials.

OMM’s efforts at reducing staff levels are clearly aimed at containing costs while the agonizing wait for a big order for its 2D subsystems continues. Burke says OMM has enough cash to continue “quite well on into 2002.” The big question is whether that will be long enough.

When it comes to the large 3D subsystems, Burke says OMM made three 64x64 prototypes, but they remained ensconced in the labs.

Corning Inc. (NYSE: GLW) is widely credited with shipping the first 3D subsystems. Ciena Corp. (Nasdaq: CIEN) and Tellium Inc. (Nasdaq: TELM) both demonstrated switches incorporating Corning 256x256 port 3D MEMS modules at the Supercomm and NFOEC trade shows. Corning's "relationships with customers and potential customers continue unchanged," says Mark Newhouse, general manager of Corning's optical networking devices division, in an email to Light Reading.

A bunch of other startups are planning to ship 3D subsystems by the end of the year. One of them is Integrated Micromachines Inc. (IMMI), which plans to ship a 36x36-port prototype in October followed by a 256x256-port beta version early next year. “Our customers are beating down the door” says Steve Walker, IMMI’s 3D program manager. He says “more than several” system vendors are designing IMMI’s 3D subsystem into their equipment.

Another startup, Onix Microsytems Inc., plans to ship commercial 3D systems “some time next year” according to Meng-Hsiung Kiang, its founder and director of business development.

Of course, startups like IMMI and Onix may have equipment manufacturers beating down their doors to get early access to the latest developments in 3D subsystems, but that doesn’t mean that these equipment manufacturers have got carriers beating down their doors for monster optical crossconnects.

If OMM is in trouble getting big orders for relatively simple 2D MEMS subsystems, the chances are that the likes of IMMI and Onix will face an even longer wait for big orders for their 3D MEMS subsystems.

— Peter Heywood, Founding Editor, Light Reading
http://www.lightreading.com Want to know more? The big cheeses of the optical networking industry will be discussing this very topic in a session at Opticon 2001, Light Reading’s annual conference, being held in San Jose, California, August 13-16. Check it out at Opticon2001.

Page 1 / 4   >   >>
Peter Heywood 12/4/2012 | 8:01:38 PM
re: OMM-inous News I guess you're refering to Ilotron getting killed.

In one respect, Corvis and Ilotron were similar, in that both outfits believed that the only way you could have a genuinely all-optical network was to sell a complete package of an all-optical switch integrated with a long haul DWDM system and controlled by a management system.

That's where I was coming from by positioning Ilotron as a Corvis competitor. I have to admit I also chose this angle because I knew that a lot more folk would read the story than if I simply pitched it as a UK startup developing an optical switch.

In another respect, Corvis and Ilotron aren't similar. Ilotron was using OMM's MEMS subsystem, and all MEMS subsystems have significant insertion losses. Corvis, on the other hand, is using something secret inside its switch but I'm fairly sure it's based on amplifiers so that the light actually gets a boost of power as it passes through Corvis's switch, rather than losing power. I think this makes a big difference in terms of whether the whole system eliminates electronic interfaces.

Anyhow, that's my understanding. Anybody else want to chip in?
dc boy 12/4/2012 | 8:01:38 PM
re: OMM-inous News .....does this pretty much put a stake in the heart of this story????


JANUARY 24, 2001

Corvis Gets Some Competition


Corvis Corp. (Nasdaq: CORV - message board) has made a big thing out of being first to market with a package of technologies for building all-optical networks G- networks that eliminate transponders and thus slash carrier costs.

Now, however, Corvis has gotten some direct competition. A U.K. startup called Ilotron Ltd. has developed a similar (and possibly better) package of technologies with exactly the same goal.

Belzebutt 12/4/2012 | 8:01:32 PM
re: OMM-inous News From: http://www.lightreading.com/do...

Things get more complicated still, unfortunately, because itGs not possible to convert one wavelength to another without changing the optical signal back into an electrical one and retransmitting it. This is [...] an after-thought with the Corvis switch. Another box has to be bought to accomplish this.

I was flamed by a person on the CORV message board when I stated the above. He (romulus?) claimed that I was wrong and that Corvis could do all-optical wavelength translation, but obviously didn't say how. What do you say to that, Peter?
Peter Heywood 12/4/2012 | 8:01:25 PM
re: OMM-inous News Belzebutt, Corvis can't do all optical wavelength conversion.

Luxcore claimed it was doing all optical wavelength conversion in the demo of its switch at OFC. It's now developing subsystems rather than a switch. We ran a story on this at NFOEC. Here's the URL:


This story identifies a couple of other startups developing all-optical wavelength converters - Lightbit and Optovation.

There's some doubt whether any of these folk are doing genuine all optical wavelength conversion.

EG: Optovation says that its all-optical wavelength conversion comes courtesy of a component that Alcatel has sold for a couple of years. However, I've talked to someone who claims familiarity with this subject who says (a) the Alcatel component isn't genuinely all-optical and (b) "anybody claiming to do all optical wavelength conversion is lying".

We'll follow up on this when we get a moment to catch our breath.
Belzebutt 12/4/2012 | 8:01:23 PM
re: OMM-inous News What about the ability to switch one or a few wavelengths, separating them from the rest. Do you think that's in there? Seems like it would "only" require some filters...
lambdaswitching 12/4/2012 | 8:01:19 PM
re: OMM-inous News Peter H. said GǣGǪ, and all MEMS subsystems have significant insertion losses.Gǥ

Not trying to pick a fight here, can you quantify how much in your view is a significant amount of insertion loss?

IGm struggling a bit myself with a definition of the term significant.

Peter Heywood 12/4/2012 | 8:01:14 PM
re: OMM-inous News Just got this note from someone at Agere:

"Look at this next time you write about 3D MEMS. You can buy it already. It's
not only a datasheet or press announcement, it's a product! And it worked
during OFC 2001 already! Have not heared much from you about it...


I'm witholding the name because it was a private note.

hawkman 12/4/2012 | 8:01:12 PM
re: OMM-inous News With 6dB + loss and not telcordia qualified, you might as well still be in development.
manoflalambda 12/4/2012 | 8:01:12 PM
re: OMM-inous News Hey all,

Well, that Agere PDF (thanks LR) gives a loss of 6 dB. 1 dB of that would be in the ribbon connectors (MTP or MPX) on the side of the box (1/2 dB in and 1/2 dB out).

Any applications, of the top of your collective heads, for a 6 dB unit (does this unit do power monitoring on the inputs as well?)?

lambdaswitching 12/4/2012 | 8:01:11 PM
re: OMM-inous News hawkman said

" With 6dB + loss and not telcordia qualified, you might as well still be in development."

The loss number seems about where I would expect, on Telcorida, wonder what the time is to complete qualifications, assume there are no "issues" with passing.
Page 1 / 4   >   >>
Sign In