×

TMCnet
ITEXPO begins in:   New Coverage :  Asterisk  |  Fax Software  |  SIP Phones  |  Small Cells
 

Feature Article
May 2004


Enabling The Next Generation Of Telecom Applications

BY MIKE WERNING

The telecommunications recovery will be driven by service providers rolling out new services and generating new income streams. However, the high cost of the infrastructure to support those services can limit their profitability and act as a strong disincentive to invest. The technology on which telecom equipment is based continues to evolve, with more and more emphasis being placed on open standards. Could this be the catalyst that drives down the cost of telecom infrastructure and helps to makes new services profitable? Or is something more needed?

Challenges To Growth
It�s no secret that the telecom industry has struggled significantly in the past few years. Reduction in capital and operational expenditures by the service providers has had a devastating effect on system-level solution providers at all levels. Solution providers have focused intensely on cost cutting and finding ways to operate more efficiently. Their short term solutions have revolved around optimizing existing equipment through incremental upgrades and optimizing existing networks. Today, convergence in the marketplace is starting to create a demand for new value-added services like push-to-talk, interactive media delivery, services on demand, voice and packet cable services, and video over DSL. New investments will be needed to offer these services to subscribers while at the same time controlling the rate of growth in capital expenditure.

The intense competition between telecom equipment manufacturers to provide these new services is squeezing development time windows. This increased time to market pressure coupled with the increased complexity of these new designs has forced telecom equipment manufacturers to rethink their value proposition. They must choose carefully where they want to focus their scarce engineering and research resources. Investment in non-value-added components will increase overall system costs while sacrificing any competitive advantage gained due to a lack of focus on core technologies. The high cost of proprietary system infrastructure can offset a service provider�s incentive to invest in the new application or enhanced service offering.

Evolution Toward Standardization
Telecom equipment, to a great extent, can be separated into two areas:

  • Common, non-differentiating features that are required but don�t necessarily add value in the end application; and
  • Truly differentiating items that add value or give a competitive advantage.
    Common features are likely to be outsourced, and therefore, cost-minimized based on economies of scale. Using standards-based commercially available products is the best way to leverage the investment and momentum the industry has created.

Standards have been a part of the telecom industry for the last 20 years. In the computer industry, some could argue that the PC was the first attempt to create a standard system architecture. Although not an official standard, this effort resulted in the creation of an ecosystem of compatible hardware and software. Economies of scale and intense competition has driven down prices and proliferated this architecture into many different facets of our life.

Technology offshoots from this market found homes in many telecom system-level products. The PCI Bus, originally created to replace the many custom microcomputer peripheral bus architectures in PCs, became the main communications bus in CompactPCI bus systems. Today, packet-based communications, driven by the explosion of the Internet, has become the defacto transport standard. The latest incarnation of standards-based systems created by the PCI Industrial Computer Manufacturers Group (PICMG) utilizes an embedded Gigabit communications network as its main transport mechanism. These systems, unlike the PC, are based upon open standards.

Open standards have been reducing the cost of telecom system design by allowing telecom equipment manufacturers to concentrate on the value-added components of their product or application. Open systems provide common infrastructure such as the chassis, power supplies, system management, control plane computing elements and data plane transport elements. The products are currently available as part of the value chain in the ecosystem that formed around the development of the standard. As the price performance of the individual elements improve, telecom equipment manufacturers can perform upgrades to their systems while providing new revenue generating services.

Preparing For The Future
As we look at the road ahead and anticipate the demands for next-generation architectures, there are limitations in power and network transport that lower the capacity of today�s systems. The desire to utilize state-of-the-art general purpose processing technology is driving the need for both increased power per slot and increased board space for next-generation switching technology. In PICMG systems today, these limitations can be worked around by utilizing new low-power processor technology, which is being driven largely by the commercial laptop market. When massive computing power is required by an application, the most compute intensive functions can be moved to state-of-the-art rack mountable, multi-processor servers.

However, if the system must go into a core network or central office environment, the price performance advantage of generally available servers is lost due to the additional cost of telecom environmental certification (NEBS) and quality that is a requirement in such locations. In addition, commercial server products are not designed from the ground up to pass NEBS testing, which may prohibit their use in such applications all together. Luckily, a new standard has been approved that addresses these concerns, and its availability is right around the corner.

The Next Big Thing
The most recent and potentially the most significant entry into the open standards telecom market is called the Advanced Telecom Computing Architecture, or AdvancedTCA. Like CompactPCI and its derivatives, AdvancedTCA has been developed with participation by more than 100 companies under PICMG. AdvancedTCA is based on a modular compute platform built on multi-vendor interoperable building blocks. These building blocks are commonly server blades, storage blades, chassis management, and control transport switching and data transport switching blades.

AdvancedTCA borrows from the idea used in proprietary systems, where functional elements are abstracted to the blade level and integrated into a chassis. The main exception, and main point of attraction, is that the interface to the system, physical form factor, and chassis interconnects have all been based on open standards. This allows for a multi-vendor ecosystem to enable blade-level solutions to be integrated into the system, eliminating dependency on sole source providers. The physical form factor, power, and cooling have all been increased dramatically, compared to CompactPCI, allowing the use of state-of-the-art technology, and accommodating a long roadmap of follow-on technology inside the system.
With AdvancedTCA, it�s now possible to collapse into a single box what has typically required a complete rack full of equipment. Applications areas that would benefit include wireline and wireless infrastructure; media, access, trunking, and telephony gateways; service nodes, echo cancellers; optical core networking; data center server and storage elements.

The adoption of AdvancedTCA is not limited to the top tier telecom equipment manufacturers. These vendors, who have large product lines of disjointed proprietary system-level products today, will benefit by leveraging a common platform that spans their product lines as well as reaping greater price productivity due to higher volume purchasing. The second and third tier of telecom equipment manufacturers will benefit by a leveling of the playing field. Standards-based equipment eliminates potential market barriers to entry by allowing products to be mixed in a competitor�s installed base. Standardized hardware, however, is only part of the solution.

Application Enabling
The next level of abstraction to be moved into the open standards world is in the management of the platform. In the past, platform management was done in proprietary software. This created a legacy software code base that needed to be ported and brought up to date as new underlying transport technologies were developed. The development of a standard set of APIs to manage the underlying implementation is clearly required to maintain platform portability and minimize application software impact.

This requirement has been recognized and addressed by the Service Availability Forum (SAF). The SAF specification defines a set of layers between the customer application and the low-level hardware. Communication between these layers occurs utilizing a standard set of APIs. This set uses the concept of a middleware layer that supports availability management, cluster management, message, event and check pointing services. This middleware removes the requirement for applications to have direct knowledge of the underlying hardware. The lower layer provides the interface to the operating system and hardware. The lower layer facilitates portability of the middleware across hardware platforms and manufacturers.

The combination of both AdvancedTCA hardware and SAF compliant middleware is a powerful enabler for telecom equipment manufacturers to concentrate on their value-added hardware and application software.

Conclusion
Telecom equipment manufacturers that include both standards-based hardware, as well as software in their products, will be able to provide the next generation of scalable, cost-effective telecom products. These new products, leveraging on the economies of scale, and building on the time-to-market advantage, will allow service providers to roll out timely and competitive new services and new revenue streams. After all, when traveling between points A and B, it is better to buy a car and drive it there than to start by building one from the ground up.

Mike Werning is principal staff engineer at Motorola Computer Group, a leader in the innovation of intelligent building blocks for standards-based embedded computing. These building blocks include open-architecture hardware, rich software, and application-ready platforms that enable equipment manufacturers to quickly and cost-effectively embed leading-edge functionality into their next-generation systems. More information can be found at www.motorola.com/computer.

If you are interested in purchasing reprints of this article (in either print or HTML format), please visit Reprint Management Services online at www.reprintbuyer.com or contact a representative via e-mail at [email protected] or by phone at 800-290-5460.

[ Return To The May 2004 Table Of Contents ]

 


Today @ TMC
Upcoming Events
ITEXPO West 2012
October 2- 5, 2012
The Austin Convention Center
Austin, Texas
MSPWorld
The World's Premier Managed Services and Cloud Computing Event
Click for Dates and Locations
Mobility Tech Conference & Expo
October 3- 5, 2012
The Austin Convention Center
Austin, Texas
Cloud Communications Summit
October 3- 5, 2012
The Austin Convention Center
Austin, Texas