Comms chips

Cisco Goes Greenfield

Founders of Ethernet chip vendor Greenfield Networks are getting a second go-around with Cisco Systems Inc. (Nasdaq: CSCO), which announced today it's acquired the startup for an undisclosed sum.

Greenfield's four founders hailed from Grand Junction Networks, a 1995 Cisco acquisition. They added a fifth Grand Junctioneer, Bill Rossi, to take over as CEO when initial chief exec Gary Smeardon left to join a solar-energy startup. (See Grand Junction Vets Go Greenfield and Greenfield Hires Another Cisco Vet.)

Terms of the deal were not disclosed. Cisco plans to pick up most of Greenfield's 60 employees, but Rossi won't be among them. Cisco isn't elaborating. "It was just part of the deal," says Rob Salvagno, a director in Cisco's business development team.

Cisco doesn't typically spell out the origins of its acquisitions, but Salvagno did note that the Grand Junction link was not the key to getting this deal started. Grand Junction got acquired into Cisco a long time ago, and the resulting products, including the Catalyst 2900 and 3550, live on.

Cisco liked Greenfield's achievements in "developing configurable silicon that focuses on metro Ethernet features," Salvagno says. Cisco isn't giving many details about how the chips might be used, but they'll be going into desktop and modular switching systems, he notes.

Greenfield was founded in 2001 to produce Ethernet switching chips, taking on Ethernet chip giants Broadcom Corp. (Nasdaq: BRCM) and Marvell Technology Group Ltd. (Nasdaq: MRVL) as well as a host of competitors including Agere Systems Inc. (NYSE: AGR.A), SwitchCore AB (Stockholm: SCOR), and Vitesse Semiconductor Corp. (Nasdaq: VTSS).

Greenfield's pitch was that its parts were better suited for metro Ethernet, whereas the best-known Ethernet chips were aimed primarily for the enterprise market, not for carriers. Moreover, Greenfield's parts added some Layer 3 know-how, giving them some characteristics of a network processor.

Other companies were moving in that direction, too, however. Before Greenfield had launched, Switchcore and Vitesse had made a point of putting Layer 3 functions into their Ethernet switch chips, and Broadcom followed suit soon after. (See Broadcom Targets Metro Ethernet.)

Being more complex than many Ethernet chips, Greenfield's parts required lots of software savvy and backup, meaning it was a serious commitment for a systems vendor to adopt the chip. "The time to market takes so long for products in this range, that it made sense" for Greenfield join up with Cisco, says Kishore Jotwani, Greenfield's vice president of sales and marketing.

Greenfield had announced a couple of big customers in Asia, namely the H3C Technologies Co. Ltd. joint venture and Beijing Harbour Networks Co. Ltd. (See Huawei-3Com Picks Greenfield and Harbour Picks Greenfield.)

Naturally, Cisco won't be continuing to serve those companies. Jotwani says Greenfield has worked with them for finding alternatives. He wouldn't comment on either company's product status, but considering Greenfield's second-generation chip began shipping just this year, it's unlikely either company got too far into the product design cycle with it.

— Craig Matsumoto, Senior Editor, Light Reading

goundan 12/5/2012 | 3:34:58 AM
re: Cisco Goes Greenfield $10M + IPR + Engineers. Primarily pulled in by the chicken coops from Petaluma for their long in the tooth ONS product range.
qrious 12/5/2012 | 3:34:57 AM
re: Cisco Goes Greenfield If $10M is true, talk about a fire sale! Between letting the VCs get something back and keeping the engineers motivated, the 'second go-around at Cisco' is probably nothing like the first for the founders. But then again, they got something, which is a lot more than what many others can say!
TRPE 12/5/2012 | 3:34:53 AM
re: Cisco Goes Greenfield How reliable is $10M? Seems really low even in this market! Were they in trouble? They have operating on $21M raised 2 years ago. I wonder why Cisco bought them rather than a chip company?

Anyone know any more about this?
Mark Sebastyn 12/5/2012 | 3:34:52 AM
re: Cisco Goes Greenfield My thoughts here.

Pete Baldwin 12/5/2012 | 3:34:50 AM
re: Cisco Goes Greenfield I like Andrew's model of "How to get acquired by Cisco," but I'd like to propose another startup model -- riskier, and yet cheaper!

1. Decide what tech is going to be important to Cisco in two or three years

2. Patent some key aspects of it. (No actual invention required.)

3. Wait for Cisco to buy you out once they discover the stuff's been patented.

OK, yes, I stole that from the bubble days, when (i'm told) people actually tried this. It probably had a low success rate even back then, but heck, it's cheap...
schlettie 12/5/2012 | 3:34:50 AM
re: Cisco Goes Greenfield If the rumors that Cisco picked up Greenfield for $10M are true, then Huawei-3Com were really asleep at the wheel. More likely that the Huawei-3Com deal never materialized, which is why Greenfield's VCs pulled the plug.
Pete Baldwin 12/5/2012 | 3:34:50 AM
re: Cisco Goes Greenfield I like schlettie's point -- I think Andrew's got the right argument as to how the deal emerged, but you have to wonder why 3Com couldn't step in, especially if the price really was that low.
Lightbulp06 12/5/2012 | 3:34:49 AM
re: Cisco Goes Greenfield Hm, according to my knowledge the H3C Ethernet Switches are using the Marvell chip set.

Cisco is currently facing very heavy competition in the Metro-E Market. They have a tremandous need for something new, may the Greenfild move will help them to close the gap.
Sisyphus 12/5/2012 | 3:34:41 AM
re: Cisco Goes Greenfield metro e? the greenfield packetry ii specs read rather enterprise-ish to me: buffer size and no of queues don't seem geared towards sp infrastructure needs. then again, there's enough ethernet market insiders within cisco, so i am sure the technology has its useful niche there, especially if the price tag mentioned here is accurate. what's tough for a start-up to sell is sometimes surprisingly easy to sell if it comes with the continuity warranty of a large company... and thus it's a safe assumption that cisco indeed didn't want someone else to get this cheap, and use it as an enabler...

indeed one wonders why none of the chip guys (brcm, amcc etc) picked it up.
Sign In