ATCA/Standard Servers

Whatever Happened to ATCA?

Wasn't AdvancedTCA supposed to be a $1 billion market by now?

More like multiple billions, actually. (See ATCA to Be Worth $3.7B in 2007.) But ATCA -- a set of standards from PCI Industrial Computer Manufacturers Group (PICMG) for building telecom hardware -- has come down to Earth, at least in terms of forecasts.

On top of that, big names Intel Corp. (Nasdaq: INTC) and Motorola Inc. (NYSE: MOT) have sold off their ATCA-related divisions, although the ATCA camp notes that neither company is abandoning the technology (more on that later). And you've got IBM Corp. (NYSE: IBM) still hoping to push its BladeCenter franchise into ATCA territory.

Even so, ATCA proponents say the technology and the market prospects remain sound. It's just that the numbers got smaller.

ATCA sales will close 2007 at $600 million to $700 million, estimates Crystal Cube Consulting analyst Ernie Bergstrom. He's "pretty confident we'll hit close to $1 billion" in 2008 and is predicting a $3 billion to $4 billion ATCA market in 2010.

So, what happened? For one thing, ATCA didn't turn out to have the scope some equipment makers were expecting circa 2002 or 2003.

"There was a sense they were going to be able to do a lot of augmentation of the existing systems," Bergstrom says. "That really did not happen, so it had an effect on the revenues."

Then there's the possibility that ATCA, like a lot of technologies, just turned out to be harder in real life than on paper.

"I've been watching one customer for the past year," says Mike Coward, chief technology officer for Continuous Computing Corp. . "They should have been in the market by now, but they had to increase the platform team from one person to five because of how badly they underestimated the integration. They slipped the entire product by a year."

Keep in mind, Continuous is hoping to make money on ATCA integration services; that's the crux of the FlexTCA offering it announced in October. (See Continuous Unveils FlexTCA.) Still, Coward notes that customers' expectations for ATCA have slipped out by two years, and he says it's because of integration.

Specifically, some equipment makers hoped to slap together an ATCA system within a year, using components from various suppliers. "I can't think of a single customer that's put together building blocks and gotten something out in a year," Coward says.

Buying time
ATCA just went through a minor acquisition frenzy. Intel sold its board business to Radisys Corp. (Nasdaq: RSYS). More notably, Motorola sold its Embedded Communications Computing (ECC) group, which has 1,100 employees and had 2006 revenues of $520 million, to Emerson Electric Co. . (See Sun Beams on ATCA and Emerson to Acquire Mot ECC.)

Intel and Motorola were two big proponents of ATCA early on, but most in the industry aren't too concerned about either company abandoning the technology.

Intel has been backing out of some telecom markets anyway. (See Intel Hands Off to Cortina and Marvell Takes a Bit of Intel.) The company still sees ATCA as a viable market for its chips; it's just that it doesn't want to sell the blades any more.

"They made it very clear in their keynote they are not at all backing away from AdvancedTCA," Bergstrom says of Intel's appearance at the AdvancedTCA Summit 2007 in October.

To Page 2

1 of 2
Next Page
AutoDog 12/5/2012 | 2:57:50 PM
re: Whatever Happened to ATCA? If ATCA has fallen short, ATCA300 (the transport-focused variant) seems to have completely fallen on its face.

I think the fundamental flaw of ATCA300 is not technical, but business. If you're an incumbent telecom equipment manufacturer (TEM), what possible incentive do you have to make your chassis 'open'?

If you make universal pluggable cards for a generic box, you could get designed out at the drop of a hat. Whereas if your proprietary, home-grown chassis & cards gets deployed into a Tier-1 carrier's network, you're golden.

I think that is one big reason why the number of big TEMs pushing ATCA300 = doughnut

Possession (of a socket) is 9/10ths of the law.
teleliar 12/5/2012 | 2:57:49 PM
re: Whatever Happened to ATCA? Autodog,

As far as the stated "fundamental flaw", there is no difference between ATCA or ATCA300!

In a big TEM, it is very difficult (politically) to unify architecture among products as it means reducing the number of redundancy.

As far as who should push for open telecom architecture, you are looking to the wrong place; Only Service Provider's can push and that will happen when they can not differentiate themselves using of the shelf products.
bollocks187 12/5/2012 | 2:57:49 PM
re: Whatever Happened to ATCA? A significant path for ATCA survivial relies upon new startups leveraging the platforms capabilities.

With very few HW startups......
Hanover_Fist 12/5/2012 | 2:57:47 PM
re: Whatever Happened to ATCA? One point commonly overlooked (or conveniently not addressed) by most ATCA manufacturers/vendors is that complexity in software kills their ability to provide COMPLETE, ready-to-implement, time-to-market solutions.

They would fail to address the biggest issue (or better yet, dance around with slick marketing slides) that using Common off-the-shelf HARDWARE components does not result in a faster time-to-market solutions because they ignored (e.g., avoided) the fact that the software which made the whole system work was in-fact a customized per system work endeavor far more complex than the ATCA hardware specification.

Bending sheet metal is a simple task compared to the millions of programming man hours required just to finish off ONE ATCA solution.

Ask each of the vendors mentioned in this article about the size of their ATCA software programming staff - I'm going to guess that you will be surprised at their answer.

Ask them specifics about how they plan to support the customized software required for each and every ATCA solution they propose to sell to the market? Remember that no two solutions are alike.

That's why the big guns got out of the ATCA market - each ATCA solution required a team of software programmers, which is an untenable business model.
Sign In