A quiet evolution is underway in the mobile core network, as equipment vendors migrate vital network applications to commercial hardware and software platforms based on the Advanced Telecom Computing Architecture (ATCA) 3.x specifications from the PCI Industrial Computer Manufacturers Group (PICMG).

What this will not do is have an immediate, discernable impact on end-user services, like making it easier to connect to corporate networks, or quicker to download MPEG files. Exciting, huh?

Instead, ATCA will help equipment suppliers transition products from closed, legacy hardware, often based on ATM and voice switching platforms, to low-cost, off-the-shelf hardware designed and built by third parties. So, it's not even especially sexy in a propeller-head kind of way.

ATCA proposes a modular hardware platform based on standardized chassis form factors, management modules, and backplane interconnects, capable of supporting computer boards, based on commercial processors running standard system software (e.g. carrier-grade Linux) and middleware from a range of suppliers. (See Light Reading's AdvancedTCA Report for more on the design basics).

Backed by computer industry heavyweights including Hewlett-Packard Co. (NYSE: HPQ), Intel Corp. (Nasdaq: INTC), Motorola Inc. (NYSE: MOT), and Sun Microsystems Inc. (Nasdaq: SUNW), the idea is to create an ecosystem of chassis and board suppliers that network equipment manufacturers can leverage to produce high-availability telecom equipment at lower cost and in less time. In other words, ATCA is exciting in a CFO kind of way.

The argument holds that equipment vendors no longer differentiate their products according to the capability of their computing platforms, and should instead move to pre-integrated commercial hardware. This is expected to free up R&D resources and enable equipment developers to focus investment on the application-layer software and system-wide management and operational features that appeal to network operators.

ATCA-backers have targeted 3G wireless equipment makers specifically because they view 3G radio network controllers (RNCs) and core products, such as serving GPRS support nodes (SGSN) and gateway GPRS support nodes (GGSN), as being relatively immature and low-volume (i.e. less price-sensitive) products, with manageable throughput requirements.

This month's Unstrung Insider report – Wireless ATCA for 3G Networks – sets out to test this premise by asking vendors how, when, and why they expect to adopt ATCA. The report also looks at whether vendors will port control-layer applications, such as softswitch MSCs and IMS servers, and service-layer applications, such as push-to-talk, to ATCA.

The conclusion is essentially that ATCA is being adopted at varying rates, and for various reasons and applications, across the industry.

Among the early adopters, major names in the frame include Alcatel SA (NYSE: ALA; Paris: CGEP:PA), NEC Corp. (Nasdaq: NIPNY; Tokyo: 6701), and Siemens AG (NYSE: SI; Frankfurt: SIE), all of which have embarked on a strategic, long-term transition to ATCA-based open platforms across their product lines.

Aggressive second-tier vendors, such as Huawei Technologies Co. Ltd., are also early adopters of ATCA as a cross-platform play. For smaller vendors and startups focused on specific applications, the decision to use commercial hardware is similarly easy – examples here include Radvision Ltd. (Nasdaq: RVSN) (real-time video application) and Sonim Technologies Inc. (push-to-talk).

Some other vendors, however, remain a little more circumspect. Although these holdouts see long-term value in ATCA, they are unlikely to make a wholesale migration in the short term. Names in the frame here include Ericsson AB (Nasdaq: ERICY) and Nokia Corp. (NYSE: NOK).

Nokia, for example, is an enthusiastic supporter of the ATCA concept and sees the move to blade servers as a natural migration for today's core network products. The question is, at what rate will the transition occur?

Petri Pöyhönen, Nokia's vice president for core networks, IP communications, verifies that it will be a straightforward R&D effort to port operating systems, high-availability middleware, and server applications to ATCA hardware. But he cautions that the cost savings often attributed to ATCA are yet to materialize.

"We're in the middle of a transition that will take a fairly long time," says Pöyhönen. "So far we're capable of developing [non-ATCA] products with highly competitive cost structures."

This pragmatic view prevails across the wireless sector, with even early adopters stressing that their product development cycles are not hostage to the maturity and cost-effectiveness of ATCA. The result is that ATCA won't have a huge impact on the wireless equipment market in the short term, but will have a long-term impact on vendor competitiveness, as commercial off-the-shelf hardware platforms become pervasive across the sector.

— Gabriel Brown, Chief Analyst,Unstrung Insider

The report, Wireless ATCA for 3G Networks, is available as part of an annual subscription (12 monthly issues) to Unstrung Insider, priced at $1,350. Individual reports are available for $900. To subscribe, please visit: www.unstrung.com/insider.