AdvancedTCA

AdvancedTCA Featured Article

 

In a recent whitepaper, RadiSys explains in great detail how the AdvancedTCA (ATCA) platform standard can be used to build a flexible, Internet traffic-monitoring solution for modern networks.

 

The whitepaper is 12 pages long, and very detailed. If you’re looking for an overview that explains how AdvancedTCA (News - Alert) can be used to build systems, this article presents an executive summary.

 

 

In the whitepaper, RadiSys begins by examining various architectures that can be used to build a traffic monitoring solution, including the following:

The problem is that none of these provides the ideal mix of functionality and cost-effectiveness—either they are too expensive, lack critical performance requirements, or require complex programming.

 

“Today’s mobile, always-connected road warriors demand fast and protected connections to their mission-critical data,” RadiSys explains in the whitepaper. “Service providers vying for such highly profitable business users must deploy networks that offer security and high-bandwidth connectivity.”

 

Providing that type of service requires that the technology involved provides several key elements, including: 

 “The most promising way to meet the challenge is to use a network processor unit (NPU)-based AdvancedTCA platform with off-the-shelf components,” RadiSys concludes.

 

 

RadiSys’ AdvancedTCA-based solution uses the following components: 

 The ATCA-7010

 

In the whitepaper, RadiSys explains that the Promentum ATCA-7010, along with other components listed above, provides the type of open, modular architecture telecommunications equipment manufacturers (TEMs) and service providers need to stay competitive.

 

“The true competitive advantage of the ATCA-7010 is scalability,” RadiSys notes. “That scalability can be harnessed to easily upgrade the product to efficiently protect Internet applications from the security vulnerabilities of the future.”

 

 

One of Intel’s contributions to the solution, the IXP2850 network processor, makes it possible to meet “requirements for protected processing at high data rates.”

 

The processor includes on-chip support for industry data encryption standards, and interoperability with software developed for Intel’s IXP2800 network processor.

 

Intel also contributes its IXA Software Developer’s Kit 4.1, which includes “compilers, assemblers, debuggers, libraries, and simulation tools used to create software for the Intel IXP2XXX product line.”

 

RadiSys notes that, “A principal component of the SDK is the Transactor,

a cycle- and data-accurate software model of the NPU. It is used with the SDK’s graphical interface to test, debug, and tune application code.”

 

 

Any solution that uses sophisticated packet processing subsystems is, by its very nature, complex. In order to make things a bit easier to handle, RadiSys includes in its system multiple NPUs, pluggable I/O and fabric options, flexible data path switch, as well as optional accelerators such as TeleCommunications Access Method.

 

“A key challenge is to find a way to harness the power of the hardware system,” RadiSys notes. “The answer to this challenge is to abstract the underlying NPU hardware by creating an application-specific programming model implemented as a VM.”

 

In the whitepaper, RadiSys goes on to say that the VM approach gives programmers more flexibility by providing a machine-independent environment. Developers are free to focus on packet processing logic rather than being forced to consider the underlying NPU platform.

 

In a nutshell, the key advantages of the VM approach are portability, scalability, and robustness.

 

 

RadiSys’ solution also utilizes IP Fabrics’ Packet Processing Language, or PPL, to program the VM architecture. PPL is defined as “a high-level, functional programming language for describing the types of packet processing found in many of today’s networking applications.”

 

PPL capabilities include tracking connections, removing outer headers, translating IP addresses, encrypting packets, and scanning payloads for regular expressions, RadiSys says in the whitepaper These high-level abstractions result in enhanced productivity, enforced consistency, and reuse of code modules.

 

 

“Perhaps the most important byproduct of utilizing hardware and software building blocks is the reductions in TTM, software development expense, and lifecycle development expense,” RadiSys notes.

 

Using PPL and off-the-shelf hardware, along with the ATCA-7010, results in a solution that is quickly developed and therefore economical.

 

The PPL approach also provides allows for faster proof-of-concepts and/or prototypes, and greatly reduces financial risks inherent in speculative projects.

 

 

For TEMs, the advantages of the AdvancedTCA-based system described in this article include lower development costs and decreased TTM, supply chain flexibility resulting from a broad vendor ecosystem, and solid platform strategy.

 

Service providers benefit from fast time-to-revenue with new, innovative services; increased flexibility and scalability of networks; and the ability to manage network evolution with confidence.

 

 

To learn more about the type of solution described in this article, download RadiSys’ whitepaper here. Also, be sure to pay a visit to the company’s channels on TMCnet, AdvancedTCA and ATCA, and RadiSys’ website.