Optical components

CoreOptics Nabs Another $28M

An obsession with 40-Gbit/s networking is paying off for subsystems vendor CoreOptics Inc. , as the company announced a $28 million funding round yesterday.

The money brings CoreOptics' total funding to $68 million, including rounds of $21.5 million and $15 million from 2002 and 2003, respectively. (See CoreOptics Raises $28M, CoreOptics Closes $21.5M, and CoreOptics Closes $15M Series C.)

The latest round included Belgian venture capital firm GIMV, which joins as a new investor, and prior investors Atila Ventures , Crescendo Ventures , High Tech Private Equity GmbH , Quest for Growth NV, and TVM Capital .

CoreOptics builds transponder subsystems targeting 40-Gbit/s links, although it's taken the practical step of also offering a 10-Gbit/s DWDM module. Though it's been a long climb, prospects for 40-Gbit/s networking are starting to look up, aided by statements like the recent AT&T Inc. (NYSE: T) pledge for an upcoming OC768 backbone. (See AT&T Readies 40-Gig Backbone.)

The big systems vendors have been making more noise about OC768, too. Cisco Systems Inc. (Nasdaq: CSCO) recently sold its first honest-to-goodness 40-Gbit/s link, in China, while Alcatel (NYSE: ALA; Paris: CGEP:PA), Huawei Technologies Co. Ltd. and Juniper Networks Inc. (NYSE: JNPR) closed out 2005 with 40-Gbit/s announcements. (See Cisco Sells an OC768, Alcatel Touts 40-Gig Deployment, Huawei Intros 40-Gbit/s DWDM, and Juniper Intros Interface.)

Not all investors are ready to buy into the OC768 story, though. The market for 40-Gbit/s links -- meaning real OC768, as opposed to four lanes of 10-Gbit/s -- is too limited, says Drew Lanza, a partner with venture firm Morgenthaler . It's true OC768 links will end up in undersea links and backbone networks, but "they're such small markets," he says.

Lanza sees 10 Gbit/s as a better bet, citing Infinera Corp. (Nasdaq: INFN), a company Morgenthaler didn't invested in. "I'm a big fan of Infinera. That's 100 Gbit/s, but it's 10 times 10 Gbit/s," Lanza says, referring to Infinera's chip that handles 10 links of 10-Gbit/s apiece. "Infinera will probably do better than all the 40-Gbit/s companies together times two."

CoreOptics' expertise lies in electronics rather than optics. Its technology involves using equalization to prevent signal distortion, freeing carriers from having to use dispersion compensation. The cleaner signal also makes it possible to use cheaper optics in the network, CoreOptics officials claim.

"It's the way we are seeing next-generation optical networking evolve: as applications of electronics into the optical layer," says Saied Aramideh, CoreOptics vice president of marketing.

The approach has landed CoreOptics some big-name customers and partners. The company helped Marconi -- since acquired by Ericsson AB (Nasdaq: ERIC) -- to develop 40-Gbit/s transponders destined for Deutsche Telekom AG (NYSE: DT) unit T-Com . CoreOptics separately worked with Fujitsu Microelectronics Europe (FME) to develop a handful of 40-Gbit/s chips that CoreOptics will sell. (See 40-Gig Begins Its Ramp and CoreOptics, Fujitsu Join on 40-Gig.)

More recently, Siemens AG (NYSE: SI; Frankfurt: SIE) signed up to use CoreOptics' equalization technology. (See Siemens Uses CoreOptics Tech.)

CoreOptics even has a play in more modest networks. The avoidance of dispersion compensation makes it possible to run 10-Gbit/s signals on a 2.5-Gbit/s backbone, something second-tier customers might be interested in, Aramideh says.

CoreOptics employs 70, including an R&D team researching modulation schemes for OC768. Some of the new funding will go toward continuing that basic research, Aramideh says.

— Craig Matsumoto, Senior Editor, Light Reading

SAramideh 12/5/2012 | 3:49:08 AM
re: CoreOptics Nabs Another $28M While 40G applications constitute an emerging target market for CoreOptics, I would like to point out that our success to date has come from our advanced 10G EDC technology and transponder subsystems.

The growing demand for our 10G distortion tolerant transponder solutions will continue to provide healthy revenue for the foreseeable future. We are heavily involved with a large number of OEMGÇÖs and our products are in 15 carrier deployments globally.

We look forward to a GÇ£distortion tolerantGÇ¥ future in 10G and are ready to address 40G applications as that market matures.

Saeid Aramideh
Drew Lanza 12/5/2012 | 3:49:07 AM
re: CoreOptics Nabs Another $28M Saied:

I'm a big fan of 10G EDC (and FEC, for that matter).

I didn't mean to imply anything negative about CoreOptics. I surfed over to your website after Craig called and you have an impressive list of product offerings. Congratulations!

But I believe that something very fundamental happened in the transition from 1 Gig to 10 Gig (or 2.4 Gig to 10 Gig) TDM bitrate and our world will never be the same. That has left me very pessimistic about investing in higher TDM bitrate technologies.

1/2.4 Gig was (and, I believe, will always be) the last TDM bitrate that 'spanned the whole food chain'.

It started in undersea and core and then migrated to metro and then to enterprise and, ultimately, to the desktop. And this was not new - this was the way all new bitrate technologies went from high value applications to commodity.

But the same is just not true at 10 Gig. Sure it went from core to metro to enterprise, but it's not going to the desktop anytime soon (if ever).

We just don't need 10G to our desktops (heck, I'm not sure we need 1G to our desktops yet).

And there are lots of other interesting things going on at 10G like the delta between fiber and copper thermals or the match up to CMOS process generations over the roughly equivalent bitrates of OC-192 and 10GbE.

And so my point about 40G. 1G will be the last end-user commodity TDM bitrate and 10G will be the last enterprise commodity TDM bitrate.

Any higher speed we'll just cobble together with more ports in parallel using either extra cables or more colors of light.

Of course, undersea and core will always take whatever they can get. If they can get 40G, they'll take it.

But the dynamic of companies investing in new TDM bitrates for undersea and core because they knew sooner or later it would get pulled into metro, then enterprise, and then desktop is gone forever.

ms777 12/5/2012 | 3:44:58 AM
re: CoreOptics Nabs Another $28M some reasons why 10G may even migrate to the desktop/end user:

VoD - load a DVD size video in 10 seconds rather than 100

PON to the desktop - 10 GB PON with splitting ratio ~10..100 may reduce cabling/infrastructure costs in office building

Storage - when LAN speed approaches/eccesses standard HDD access speed, new storage / backup concepts may arrive

I aggree that none of them is a killer application comparable to MP3 download, ebay, 3ple play, but considering the low fraction of component cost on total installation cost they may drive 10G to the desktop, or even home.
Of course, prior to that the temp problems and cost issues must be solved. Maybe in 2-3 years ?

Sign In