Breaking AdvancedTCA Bottlenecks

Tech Score

Breaking AdvancedTCA Bottlenecks

By Jeff Hudgins, VP of Engineering, NEI, Inc.  |  June 04, 2012

This article originally appeared in the June 2012 issue of INTERNET TELEPHONY.

Networking engineers I talk to all agree: the newly released Intel Xeon processor E5 Family, based on the company’s Sandy Bridge microarchitecture, will change how software applications run on AdvancedTCA (News - Alert) platforms.

It supports innovative networking through 10 gigabit Ethernet, and its features allow for advanced virtualization and cloud computing techniques.

Let’s take a look under the hood.

The E5-2600 chips are comprised of up to eight cores, each running up to 55 percent faster than its Xeon 5600 predecessor, to deliver stepped-up server performance to the enterprise market. New enterprise servers can offer support for up to 32 GB dual in-line memory modules and increased memory capacity from 288 GB to 768 GB using 24 slots.

E5-based ATCA compute blades with more limited board real estate are expected to support up to 256 GB in 16 VLP RDIMM slots at launch. Still, a 40 percent increase over prior ATCA compute blades is expected.

Power efficiency is another key benefit. The Intel (News - Alert) Xeon processor E5 family provides up to 70 percent performance gain per Watt over previous generation CPUs. This allows communications OEMs to create power-efficient dual processor blades made for service providers that fully meet or beat ATCA power specifications.

But the real ATCA game-changer lies in the E5-2600’s integrated I/O, from which comes the ability to reduce significantly latency and increase bandwidth. ATCA’s 40G fabric has been backplane-ready since 2010 in anticipation of an updated PICMG (News - Alert) specification release. Since then, solution providers have sought ways to eliminate bottlenecks and utilize as much of this fabric as possible.

This being the first time that Intel has integrated the new PCI (News - Alert)-Express 3.0 with 40 lanes aboard each PCI-Express processor, Intel's I/O bandwidth Quickpath Interconnect eliminates those bottlenecks, improves throughput and cuts I/O latency by up to 30 percent. And a dual CPU configuration offers 80 lanes, which equates to 200 percent more throughput over the previous generation.

The result: ATCA is all about I/O throughput, and the blades will now be able to deliver more than 10gbps per node. This is a critical milestone for newer wireless video applications that service providers are so hungry to launch in 2012.

Jeff Hudgins (News - Alert) is vice president of product management at NEI Inc. (

Edited by Braden Becker