Optical/IP Networks

Cisco's Optical Switching Fabric?

Cisco Systems Inc. (Nasdaq: CSCO) appears to have picked one of the candidates for the switching fabric of its future all-optical switches.

It will use MEMS (micro-electro-mechanical systems) based subsystems from Integrated Micromachines Inc. (IMMI), judging by an announcement that IMMI plans to make today, Feb 12.

IMMI will unveil Cisco as the mystery “major telecom equipment vendor” that participated in the $45 million round of funding that it announced last November (see Dark Horse Joins Optical Switch Race). In fact, Cisco contributed around $25 million to the round, Light Reading has learned.

IMMI has been developing MEMS-based optical switches for more than five years but has kept a low profile until recently. Its big claim to fame is that its switching subsystems have much lower loss than competing products. Steve Walker, IMMI’s 3D program manager, cites a loss of 3 decibels regardless of switch size. Other vendors such as OMM Inc. cite losses of 6 decibels or more for a 32x32 2D switch and say losses increase considerably for larger-scale 3D switches.

IMMI also claims that its switches consume far less power than competing products. A maximum of 10 volts is needed to turn the mirrors that deflect streams of light pulses from input to output port, according to Walker. That’s because IMMI uses electro-magnetic rather than much weaker electro-static turning forces, he says. The upshot of this is that IMMI’s subsystems are less prone to absorbing moisture and going wrong, Walker contends.

IMMI plans to make a wide range of switching subsystems -- from 1x1 devices that simply act like a door, blocking light in a single fiber, to 1000x1000-port monsters. It expects to start shipping “a large port-count crossconnect” -- one with more than 100x100 ports -- in the third or fourth quarter of this year. It plans to sell its subsystems to vendors developing products for carriers.

At the time of IMMI’s previous announcement, Walker told Light Reading that IMMI was “already making boxes together” with the mystery backer that now turns out to be Cisco. So, what might those boxes be?

One might be the ONS 15900 Wavelength Router, the switch that Cisco got when it acquired Monterey Networks. The 15900 is designed so that it can accommodate a future optical core as a replacement or possible addition to its current electrical core, according to John Adler, Director of Marketing for Cisco's Wavelength Routing Business Unit. This is reflected in the switch’s architecture and in such things as internal connections, which are all monomode fiber, Adler says.

It’s also possible that IMMI’s subsystems could form a future core of the smaller, metro optical switch, the ONS 15232, which Cisco is developing (see Cisco Preps New Optical Switch ).

The chances are that IMMI won't be alone in providing Cisco's optical switching fabric. In fact, Cisco has already bought MEMS-based subsystems from OMM, according to OMM's S-1 filing to the Securities and Exchange Commission.

OMM is the only company currently shipping commercial MEMS-based optical switching subsystems. Its investors include Alcatel SA (NYSE: ALA; Paris: CGEP:PA), Siemens AG (Frankfurt: SIE), and Sycamore Networks Inc. (Nasdaq: SCMR).

-- Peter Heywood, international editor, Light Reading http://www.lightreading.com

Page 1 / 2   >   >>
Peter Heywood 12/4/2012 | 8:54:42 PM
re: Cisco's Optical Switching Fabric? A recap on who's using what supplier of optical switching fabric:

Cisco - IMMI (investor) and OMM (in S1)
Corvis - OMM (in S1)
Nortel - Xros (acquired) and OMM (in S1)
Village - OMM (in S1)
Sycamore - OMM (investor)
Alcatel - OMM (investor) and Agilent bubbles
Siemens - OMM (investor)
Tellium - Astarte (acquired), Analog (foundry)
Lucent - home grown
Calient - Kionix (foundry)

Who have I left out?


droid 12/4/2012 | 8:54:41 PM
re: Cisco's Optical Switching Fabric? By the way, will you provide a complete list somewhere on the internet?

droid 12/4/2012 | 8:54:41 PM
re: Cisco's Optical Switching Fabric? You left out Fujitsu -> Agilent bubbles

rafaelg 12/4/2012 | 8:54:40 PM
re: Cisco's Optical Switching Fabric? Is anyone pursuing the lowest loss available MEMS from Xerox? There was an article about their production of metalic MEMS instead of silicon. They claimed to have 100% yield on theirs.
kiz100 12/4/2012 | 8:54:38 PM
re: Cisco's Optical Switching Fabric? I am not quite sure that MEMS devices would work as an optical switch. There are too many inside stories about the problems and drawbacks associated with MEMS optical switches. The claim of 3D switch desihn is also a bubble.

Up to today the most successful applications of MEMS are airbag sensor and DLP, both have much lower demand on performance than optical switch.

JW 12/4/2012 | 8:54:32 PM
re: Cisco's Optical Switching Fabric? Is OpticNet (spun out from BEI Technologies [BEIQ] last fall) on anybody's MEMS radar?

desade 12/4/2012 | 8:54:31 PM
re: Cisco's Optical Switching Fabric?
Cisco is desperate to keep their stock from dropping below ten bucks, hence these kind of no-news "announcements." How very sad.

Has the Optical Bloom faded completely? Was it all so much vaporware? Looks like it from an investment standpoint. Maybe a few are worth keeping, but the majority are dogs. Ho-hum. Bonds are looking pretty good right now.

--Marquis Mark
mess 12/4/2012 | 8:54:20 PM
re: Cisco's Optical Switching Fabric? Corning - Intellisense
Kymata - TMP
jon 12/4/2012 | 8:54:16 PM
re: Cisco's Optical Switching Fabric? Desperate?

Seems Cisco has a clear strategy to incorporate optical cross connects and has had this strategy for quite some time from the looks of things. Its easy to drag a companies name through the mud when you dont have all the inside information. Maybe its time to listen more and speak less....

Cheers - Jon
D&K 12/4/2012 | 8:54:15 PM
re: Cisco's Optical Switching Fabric? Cisco has always had a clear strategy; that is one of the great strengths it has had as the "leader" of whatever they do, in their terms, all of it.

Dragging Cisco through the mud is what most of us have to do to compete with it. Fortunately, they have begun to handle the job quite nicely themselves as of late.

For those of us that do not have all of the inside information, (most) I would say, GÇ£conjectureGÇ¥, for that serves us well in tracking and attacking the companies of the internet.

Thus, for a company that has to re-invent itself outside of the mammoth "IOS" interface into the hearts and minds of the network building populace as a whole, how do they accomplish that task while maintaining leadership in areas where their technology grows long in the tooth?

Re-announce previous announcements of previous re-announcements, thus making it appear on a weekly, monthly, or yearly basis that they have done something monumental; spoon feeding this information to the newest members of the news world to provide buzz around what they do.

What they do is continue to provide the youngest members of the networking world with "new" technology that resembles the old. Unfortunately, the folks that saw it last time have moved on.

"You can fool some of the people all of the time and all of the people some of the time, but not all of the people all of the time..." roughly remembered.

Cisco is not primed for the future, yet. They could be, if stockholder value and return were not the primary agenda. Granted the value fuels the advancement of the technology, but it also thwarts one of the most important aspects in advancing the state of the art. It is the willingness to take a chance on things beyond the grasp of those that watch idly.

So Cisco will re-invent itself as an all-optical switch vendor, while stringing customers along with things that may negatively impact their business because of their own blind faith in a company like Cisco. TheyGÇÖll probably succeed.

But the news is full of net-based companies going casters-up, seemingly because of unsound ideas, or poor management or perhaps just the absence of funds. I feel that investment in products from Cisco is much akin to buying a Timex at a Rolex price point. The product will work, but how well? Thus companies fail beyond the hype for other reasons. Most of these folks have never seen how truly well an IP network can function because all they have to compare it to is Cisco.

Now is not the time to speak less and listen more. It is time for those that have left their blind faith in Cisco and been ruined, to speak up.

So many new startups built by the Cisco expatriates with sound ideas. We should thank Cisco for giving us these insiders that have ideas that were squashed within the machine and consider carefully how Cisco could affect the outcome of pervasive broadband to the home, and the ways to make money from it.

That, Jon, is my clear strategy, developed over the years of watching Cisco giving "good enough" when "more" was needed.

Sorry to be so long winded and full of sour grapes, but again, now is not the time to listen, it is time for the curtain to come down on the wizard. Look to the others and consider that Cisco is not the greatest, only the biggest and the loudest. And watch how well that stock has performed in the last six months in comparison to the previous five years. The economy may be slowing them down but my inside information comes from folks in all areas of this miasma; Competitors of Cisco in the optical arena are doing quite well right now; listen to that.
Page 1 / 2   >   >>
Sign In