Also in this issue by RTC Media

Also in this issue Biometric Subsystems Ethernet Switch Boards The magazine of record for the embedded computing industry

September 2006 www.rtcmagazine.com

ROUGH READ Y AND

VME CELEBRATES 25 YEARS

An RTCâ&#x20AC;&#x2C6;Group Publication

Middleware for Servers

EPIC™ XE–900 1.0 GHz CPU

Our kits are the shortest path to a successful OS on an Octagon embedded computer. • Pick your Octagon SBC • Pick the OS you prefer: Linux, Windows, QNX Octagon delivers a high performance, total solution.

Features

XE–900

XE–800

XE–700

CPU

Via Eden

AMD Geode GXI

STPC

Clock speed

400 MHz; 733 MHz; 1.0 GHz

300 MHz

133 MHz

BIOS

General Software

Phoenix

Phoneix

DRAM support

to 256 MB

32/64 MB

Compact/Flash

Type I or II

COM 1

RS–232

RS–232/422/485

RS–232

COM 2

RS–232

RS–232/422/485

COM 3

RS–232

RS–422/485

COM 4

RS–232

COM 5

RS–232/422/485

COM 6

RS–422/485/TTL

LPT1

EIDE

USB

CRT

1600 x 1200

1280 x 1024

Flat panel

LVDS

yes

Digital I/O

24–bit prog.

48–bit prog.

24–bit prog.

Ethernet

10/100 Base–T

Dual 10/100 Base–T

10/100 Base–T

Expansion

PC/104 & Plus

PC/104

Power

3.6A operating

1.6A max.

1.6A max

Temp. range

–40° to 70/85° C

–40° to 80° C

–40° to 80/85° C

Shock/vibration

40/5g

Typical Linux kit includes: • Target CPU card • Preloaded OS image on 256 MB industrial CompactFlash • 256 MB SO–DIMM module • Interface cables • Hard copy of manual • Mouse • CPU OS bootable CD • Optimized OS version • Full driver support for on–board hardware • X–Windows support • Example applications and source code • Extra documentation

X–SRAM–2 MB • 2 MB high speed, SRAM • Read and write at full bus speed • Pointers to memory saved if CPU resets or loses power X–DIO–48 bit programmable digital I/O • 48 digital I/O, 5V compatible • Source and sink 16 mA per output • Direct connection to opto-module racks X–COM–2 dual UART • Up to 230.4 kBaud data rate • Supports RS–232/422/485 • RS–485 fault protected to ±60V

XBLOKs offer the best compromise in cost and function for both PC/104 and PC/104-Plus. Only 44% the size of a standard PC/104 card, you can add two functions to your system but increase the stack height by only one level. –40° to 85° C. Heat diagram shows enhanced cooling.

Designed for the XE-900, our conduction cooling system eliminates a fan even at 1.0 GHz.

X–LAN–1 Ethernet LAN • 10/100 Base–T, Intel 82551ER • Fully plug–n–play • High performance, PCI bus interface X–USB–4 quad USB 2.0 • Speeds up to 480 mbps • Mix and match USB 1.1 and 2.0 • Current–limited ports can supply 500 mA to external devices

For a full listing of Octagon Systems products, visit us at www.octagonsystems.com

GE Fanuc Embedded Systems

Imagine what we can do for you now. Our vision is to offer you the kind of products and services you’ve only dreamed of. If you could build the perfect embedded products company, it might be something like the one we’ve created by combining the imagination, energy, and expertise of the people at GE Fanuc Embedded Systems, Condor Engineering and SBS Technologies. It would be a company with the experience, resources and courage to develop truly innovative products — which is a cornerstone of the General Electric Company. It would be a company with a wide range of available products — everything from avionics and industrial controls to networking,

communications, and fully integrated systems — which is exactly what Condor Engineering, SBS Technologies, and GE Fanuc bring to the table. By creating this new company, we’ve taken a giant step toward our vision of a different and better kind of embedded company. So we invite you to open your mind and let your imagination run wild. Because our goal is to be right there with you helping make your inspirations into realities. www.gefanucembedded.com

Departments 9

Editorial: VME Turns 25...and Time Marches On www.rtcmagazine.com

11 Industry Insider 68 Products&Technology 74 Publisher’s Letter

Features Technology in Context

Biometric Subsystems

14 Biometric Access Control Gets Networked ahul Shah, Lantronix and Bashar Masad, Ingersoll Rand Security Technologies, R Schlage Recognition Systems

Solutions Engineering

VME 25th Anniversary

20 VPX and VPX-REDI Standards Coming on Strong Stewart Dewar, VITA 46 Working Group

Network-enabled biometric systems provide a central location where alarm conditions and door activities can be stored, viewed and controlled remotely via a PC browser. Such systems can use a wireless embedded device server, such as the Lantronix WiPort, for direct monitoring and control of multiple biometric readers. • Pg. 14

25 VMEbus at 25! Clarence Peckham, GE Fanuc Embedded Systems

30 VXS: Jump Start to Higher Performance for VME Users Dave Barker, VMetro

34 VME: Moving from Evolution to Revolution Dave Evans-Hughes, Concurrent Technologies

38 VME Market Bifurcation: The VXS and VPX Face-Off Eran Strod, Mercury Computer Systems

Industry Insight Ethernet Switch Boards

Representative VITA 46 Module. • Pg. 20

44 Blade Server Interconnect Converges on Switched Ethernet Mike Zeile, Fulcrum Microsystems

49 The 1/10 Gigabit Ethernet Ripple Effect Nauman Arshad, Curtiss-Wright Controls Embedded Computing

Executive Interview 55 “Parallel Buses Are Dead!” RTC Interviews Ben Sharfi, CEO General Micro Systems

Software & Development Tools Middleware for Servers 62 Off-the-Shelf Carrier-Grade Middleware: The Next Logical Step Niranjan Vaidya, OpenClovis

Photo Credit: VME has 100 percent market share . . . on Mars. Both the Spirit and Opportunity rovers, which have continued to run far longer than their projected service life, are based on a 6U radiation-hardened, R6000-based VME board from Lockheed Martin Federal Systems, installed in a backplane card cage with other custom cards that communicate with the rover’s peripherals and sensors. Photo courtesy of NASA/JPL.

Host Interface Board Extends PCIe Bus at x8 Speeds• Pg. 68 September 2006

September 2006 Publisher PRESIDENT John Reardon, johnr@r tcgroup.com VICE PRESIDENT, European Operations Zoltan Hunor, zoltanh@r tcgroup.com EDITORIAL DIRECTOR/ASSOCIATE PUBLISHER Warren Andrews, warrena@r tcgroup.com

Editorial

EDITOR-IN - CHIEF Tom Williams, tomw@r tcgroup.com SENIOR EDITOR Ann Thr y f t, annt@r tcgroup.com MANAGING EDITOR Marina Tringali, marinat@r tcgroup.com COPY EDITOR Rochelle Cohn

Art/Production

CREATIVE DIRECTOR Jenna Hazlett, jennah@r tcgroup.com GRAPHIC DESIGNER Melissa Gaeta, melissag@r tcgroup.com PRODUCTION DESIGNER Kirsten Wyatt, kirstenw@r tcgroup.com DIRECTOR OF WEB DEVELOPMENT Marke Hallowell, markeh@r tcgroup.com WEB DEVELOPER Brian Hubbell, brianh@r tcgroup.com

Advertising/Web Advertising

CALIFORNIA COASTAL ADVERTISING MANAGER Diana Duke, dianad@r tcgroup.com (949) 226 -2011 WESTERN REGIONAL ADVERTISING MANAGER Lea Ramirez, lear@r tcgroup.com (949) 226 -2026 EASTERN REGIONAL ADVERTISING MANAGER Nancy Vanderslice, nancy v@r tcgroup.com (978) 443 -2402 EMEA SALES MANAGER Marina Tringali, marinat@r tcgroup.com (949) 226 -2020 ASIAN ADVERTISING MANAGER John Koon, johnk@r tcgroup.com (858) 755 - 4999

Billing

Mag gie McAuley, mag giem@r tcgroup.com (949) 226 -2024

To Contact RTC magazine: HOME OFFICE The RTC Group, 905 Calle Amanecer, Suite 250, San Clemente, CA 92673 Phone: (949) 226-2000 Fax: (949) 226-2050, www.rtcgroup.com EASTERN SALES OFFICE The RTC Group, 96 Dudley Road, Sudbury, MA 01776 Phone: (978) 443-2402 Fax: (978) 443-4844 Editorial Office Warren Andrews, Editorial Director/Associate Publisher 39 Southport Cove, Bonita, FL 34134 Phone: (239) 992-4537 Fax: (239) 992-2396 Tom Williams, Editor-in-Chief 245-M Mt. Hermon Rd., PMB#F, Scotts Valley, CA 95066 Phone: (831) 335-1509 Fax: (408) 904-7214 Ann Thryft, Senior Editor 15520 Big Basin Way, Boulder Creek, CA 95006 Phone: (831) 338-8228

September 2006

Published by The RTC Group Copyright 2006, The RTC Group. Printed in the United States. All rights reserved. All related graphics are trademarks of The RTC Group. All other brand and product names are the property of their holders.

A Compute Blade This Powerful Deserves A Cape.

The CPC5564 64-Bit AMD Opteron™ Single Board Computer The CPC5564 is the world's most powerful PICMG® 2.16 compute blade, and the first to be based on single- and dual-core AMD Opteron™ processors. The AMD Opteron™ processor provides a highly scalable x86 architecture that delivers next-generation performance as well as a flexible upgrade path from 32- to 64-bit computing. Its multi-core architecture offers advanced processing speed while reducing heat and power consumption. With up to 8GB of ECC memory, multiple storage options and Linux, Solaris™ and Windows® operating system support, the CPC5564 is an ideal computer for high-end packet processing or multi-threaded software environments found in wireless, softswitch and defense applications. Is it the superhero of the compute world? We like to think so.

Learn more at:

www.pt.com/sbc

WE GO t o g r e a t l e n g t h s f o r o u r c u s t o m e r s.

From the toughest terrains in the universe to the antiseptic environment of the operating room, Hypertronics delivers the highest reliability electronic connectors for our customersâ&#x20AC;&#x2122; most demanding applications. But our commitment doesnâ&#x20AC;&#x2122;t end there. We provide complete, custom design solutions for customers, including cabling, mechanical design, integrated electronics and electrical modeling capabilities. Hypertronics custom solutions can save you valuable engineering and manufacturing time and ensure the overall reliability of your final product. Our complete solutions are as dependable as our connectors, and our service and support is unmatched. For more information on Hypertronics reliable connectors and dependable solutions, call 1-800-225-9228, email info@hypertronics.com or visit our web site at www.hypertronics.com/wego.html.

Hypertronics: reliable connectors. dependable solutions. Hypertronics Corporation

Hudson, MA

1-800-225-9228 www.hypertronics.com ISO 9001, AS9100, 13485 & 14001 certified

Editorial September 2006

t’s been an amazing run. I got a phone call from Munich back in 1981 announcing the birth of the VME specification from someone whose name now escapes me over the years. However, I still remember the line, “We’ve agreed on the design of a coliseum in which to compete.” At that time, the competition centered around Motorola’s 68000 processor family. Since then, the parade of silicon—in the form of CPUs, DSPs, NPUs, FPGAs, bridge chips, networking chips, etc. that has appeared on the VME form-factor—has been a mirror of the advancements in the semiconductor industry. Through it all, VITA and VSO, the VITA Standards Organization, have managed to maintain backward compatibility with legacy boards so that developers could upgrade systems with selected newer modules while retaining their legacy equipment and never having to face the dreaded “forklift upgrade” that requires a total replacement of a system. Advancement of the technology has been voluntary, selective, strategic—and steady. Advancements on the backplane, from 32-bit to 64-bit to 2eSST and now to serial fabrics over the P0 connector with VITA 41, have always included backward compatibility to the first level. Of course, any advanced card that needed to communicate with an earlier one had to do so at that card’s level, but the connection was there. Interestingly, VME appeared about the same time that IBM introduced its first PCs to the consumer market. At that time, Motorola processors were the first-line choice for industrial and military users. Hardly anyone in that market used Intel chips and the 386 had yet to appear. To borrow a phrase, “No one foresaw the breech of the levees.” The PC phenomenon exploded across the world and would ultimately have a huge impact on the embedded computer industry. When you start seeing IBM PCs hooked up to $2 million MRI machines, it is a sign that things are changing. With advancements in the x86 architecture, the proliferation of systems with the inevitable drop in costs and, finally, with the advent of the PCI bus, x86-based embedded solutions became ever more attractive. The first PC/104 modules appeared in 1987 and the specification was published in 1992. Two years later, PICMG was formed, offering bus-based modules with their roots in PC technology. Today, there is an overlap of approximately 75 percent between VITA and PICMG members, which speaks volumes about the attractiveness of modularity and the desire for a full range of options. Today, also, we are witnessing two developments. One is that serial switched fabrics are replacing bus-oriented interconnects. The other is that the concept of what constitutes a “platform” is being raised. Generally speaking, a platform represents the level of integration at which an OEM customer can start adding unique value. The move away from bus-oriented interconnects means that

the value of a given form-factor rests more in terms of the size of application it can accommodate, how it can be cooled and the requirements of the environment for ruggedization. What was once sold as modules that an OEM plugged together and integrated with operating system software is moving to a pre-integrated system with pre-integrated operating system, system management software and middleware. This is especially true in the networking

VME Turns 25... and Time Marches On by Tom Williams, Editor-in-Chief and communications arena. Industrial control and instrumentation (including medical), on the other hand, are increasingly oriented toward maintaining legacy and custom I/O design while maintaining the option to upgrade CPU performance. This is being addressed by modular concepts such as EPIC, COM Express, ETX and custom form-factors. Today, both VITA and PICMG continue to advance innovation and standards, although both are also departing from their original base technologies. VITA 41 is probably the last true VME specification and appears to have a good future. After that there is, of course, VITA 46, which is a departure. In fact VITA 46 and VITA46 REDI appear to be aimed at the most rugged, highest performance military and aerospace applications. Given the interest of large prime defense contractors in these specs, VITA 46 may turn out to be a back door mil spec. The proposed VITA 58 can concept is, of course, also a departure from VME. VITA 46 will most probably be a niche technology while VITA 58 could conceivably reach commodity status if it catches on. It must, however, prove itself against a number of similar concepts that are further along such as the effort coming out of the Liquid Cooled Embedded Computer consortium and several proprietary efforts. On the PICMG side, ATCA and AMC have left the organization’s PCI roots and are addressing the needs of what is hoped will be the rebirth of the telecom market. But there are also calls for increased ruggedization of AMC/microTCA to make it more suitable for industrial applications. AMC in particular appears to have a bright future due to its front-removal, hot-swappable nature. All in all, the sweep of innovation continues and a salute is due to the steady history of VME and what it has contributed realizing that VME as VME has probably reached the end of its evolution. Its market will continue strong for some time to come even as its history points the way to developments taking us well beyond it. September 2006

Design 10X Faster with Graphical System Design

DESIGN

PROTOTYPE PROTOTYPE

DEPLOY DEPLOY DEPLOY

“With graphical system design through NI LabVIEW and CompactRIO, we designed a motorcycle ECU prototyping system in three months versus two-anda-half years with traditional tools.”

> Control design > Intellectual property libraries > Digital filter design > Dynamic system simulation

> I/O modules and drivers > COTS FPGA hardware > VHDL and C code integration > Design validation tools

> Rugged deployment platforms > Distributed networking > Human-machine interfaces > Firmware management

– Carroll Dase, design engineer Drivven, Inc.

Graphical System Design Accelerate your embedded design using National Instruments LabVIEW graphical programming, third-party tools, and commercial off-the-shelf hardware. Graphical system design empowers you to rapidly design, prototype, and deploy embedded systems.

Learn how to design faster through Webcasts by Analog Devices, Celoxica, and Maplesoft at ni.com/embedded.

© 2006 National Instruments Corporation. All rights reserved. CompactRIO, LabVIEW, National Instruments, NI, and ni.com are trademarks of National Instruments. Other product and company names listed are trademarks or trade names of their respective companies. 7225-821-101

(800) 450 8999

Industry Insider

September 2006

Flex Circuit-Based Chip-to-Chip Interconnection Concept Validated by Intel Team The performance potential of high-speed, copper-based, direct interconnections between IC packages by means of a flexible circuit has been proven again, this time by Intel Corporation. The results were discussed in a paper at the IEEE CPMT Society’s Electronic Component and Technology Conference (ECTC), July 2006 in San Diego, CA in a paper titled: “Flex Circuit Chip-to-Chip Interconnects.” The paper was authored by a team of scientists and engineers from facilities in Hillsborough, OR, Santa Clara, CA and Chandler, AZ. The authors concluded that “20 Gbit/s signaling on channels of relevant lengths is possible,” however, the authors also noted that “we are not ready to introduce flex I/O into any mainstream products.” Even so, they also concluded that “the possibility of a chip-to-chip interconnect with minimal discontinuities and propagation loss is a strong attractor and corresponding high-speed solutions will be in demand.” Get Connected with technology and SiliconPipe OTT technology (anproviding acronymsolutions for Off-the-Top or Over-the-Top) is viewed by many companies now as an important enabling technology not only for meeting data requirements as Get Connected is a new resource for further high-speed exploration reported, but also forinto low power potential. In a demonstration vehicle, code named “Sideproducts, technologies and companies. Whether your goal is to research latest datasheet fromsignals a company, directly over distances up to 30 winder,” the team showed it isthe possible to send at speak 10 Gbits/s withtwo an Application Engineer, jump a company's technical the anticipated transmission inches through connectors andordo sotousing less than 2%page, of the goal of Get Connected is to put you in touch with the right resource. power while maintaining a 60% timing margin. “However,” one team Whichever level of service you require for whatever type of technology, member noted, “there is still more benefit to mine and this area SiliconPipe is continuing to develop and Get Connected willextract help you in connect with theand companies and products are searching for. evaluate new you copper-based interconnection solutions to meet the cost performance needs of www.rtcmagazine.com/getconnected future electronics. It appears Stair Step Package (SSP) will be an important element of that future.”

Ad Index

has deployed their products with more than 200 customers around the globe, including 25 of the top 50 global telecommunication service providers, and many Fortune 500 and enterprise customers. Convedia’s full family of media servers will continue to be developed and enhanced under the RadiSys corporate brand. In addition, the two companies will leverage Convedia’s eXMP (eXtended Media Processing) software technology to operate on RadiSys’ market-leading ATCA platforms to further expand the market opportunities of the combined entity.

RTI and Concurrent Team for Navy’s Aegis Open Architecture Program

Real-Time Innovations and Concurrent Computer have announced they have partnered to provide RTI’s standards-based realtime data distribution middleware with Concurrent’s RedHawk Linux media processing technologies technology. Several companies StarGen Lays Off 80% of in support of the U.S. Navy’s and IP media server products. that are currently successful in Workforce Aegis Open Architecture program. Convedia’s platform products the PCI Express arena originally StarGen Inc., an Get interconnect Connected with technology and companies providing solutions nowof Voice-overAegis Open Architecture is the are at the heart Advanced Switching on chip supplier that raised more ishad the the modernization Get Connected a newroadmaps, resource forbut further exploration into products, and companies. Whetherfoundation your goal is tofor research latest Internettechnologies Protocol (VoIP) networks their are currently than $41 million datasheet in venture from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connected is to put you of the Navy’s Aegis-equipped and emerging IP Multimedia downplaying expectations that it funding since 1999, inhas halted touch with the the right resource. Whichever level of service you require for whatever(IMS) type of technology, cruisers and destroyers. The open Subsystem deployments, willyou emerge soon. and products you are searching for. Get Connected connectany withtime the companies sale of its PCI-based circuits andwill help architecture approach will allow enabling telecommunication StarGen’s initial technology laid off more than 80 percent of www.rtcmagazine.com/getconnected the Navy to readily incorporate service providers to deliver a broad was a physical-layer switching its employees. software and hardware updates range of value-added multimedia chip that could be used in a StarGen, which had at one into its Aegis systems. This enables services to their residential and variety of backplane topologies, point raised over $41 million, is those systems to be maintained at business customers. along with network processors keeping a small staff to attempt the highest level of capability while The acquisition of Convedia and control-plane processors, but to sell intellectual property employing low-cost, off-the-shelf will make available to both the company later put emphasis related to its original StarFabric solutions. companies’ customers a broader on the ASI standard originally switching technology. However, Concurrent’s RedHawk Linux set of technologies and solutions proposed by Intel. Today, even there are a limited number of provides a fast response to external delivered by an expanded team Intel doesn’t talk about ASI very customers for that technology and events, optimized interprocess with greater scale and breadth. much. Other original members of the demand for it does not appear communication, and the high I/O The combination will facilitate the ASI-SIG have also lost interest to be growing. In addition, it was throughput needed for time-critical RadiSys’ penetration of the in the technology. less clear what wouldwith become applications. RedHawk Linux, Get Connected companies and high-growth VoIP equipment Get Connected of theproducts newerfeatured technology related with the popular Red in this section. and IMS infrastructure marketsin thiscompatible with companies mentioned article. to thewww.rtcmagazine.com/getconnected PCI Express Advanced RadiSys to Acquire Hat distribution, guarantees that www.rtcmagazine.com/getconnected as well as accelerate RadiSys’ Switching Initiative (ASI). While Convedia a user application can respond to strategy to provide turnkey StarGen is paying lip service to RadiSys has announced that an external event in less than 30 networking platforms. Convedia ASI, there has not to date been it has entered into a definitive much resonance in the market agreement to acquire privately Get Connected with companies mentioned in this article. signaling a demand for the held Convedia, a maker of IP www.rtcmagazine.com/getconnected Get Connected with companies and products featured in this section.

Products

End of Article

www.rtcmagazine.com/getconnected

September 2006

Industry Insider

Event Calendar 10/3-5/06 ARM Developers Conference Santa Clara, CA www.arm.com/developersconference

10/9-11/06 2006 AUSA Annual Meeting & Exposition Washington, DC www.ausa.org

10/9-11/06 WiMAX World USA Conference and Expo Boston, MA www.wimaxworld.com

10/17/06 Real-Time & Embedded Computing Conference Patuxent River, MD www.rtecc.com/paxriver

10/19/06 Real-Time & Embedded Computing Conference Washington, DC www.rtecc.com/tysonscorner

10/23-25/06 MILCOM Military Communications Conference Washington, DC www.milcom.org

10/23-26/06 9th Annual Systems Engineering Conference San Diego, CA www.ndia.org

11/7/06 Real-Time & Embedded Computing Conference Montréal, PQ www.rtecc.com/montreal

11/9/06 Real-Time & Embedded Computing Conference Ottawa, ON www.rtecc.com/otttawa

If your company produces any type of industry event, you can get your event listed by contacting sallyb@rtcgroup.com. This is a FREE industry-wide listing.

September 2006

microseconds on a dedicated processor. RTI Data Distribution Service provides a rich publish-subscribe middleware environment that complies with the Object Management Group’s (OMG) Data Distribution Service for Real-Time Systems (DDS) standard. RTI Data Distribution Service significantly reduces the time required to implement communication within a distributed real-time system while providing a more scalable, maintainable and manageable infrastructure.

SanDisk Will Buy Israeli Data Storage Supplier

SanDisk and msystems have entered into a definitive agreement for SanDisk to acquire msystems in an all-stock transaction. This combination joins together two flash memory companies with complementary products, customers and channels. Together the combined company will have the people, technology, manufacturing and IP to play a leading role in creating new markets and accelerating the penetration of flash memory into existing storage applications. “The NAND flash data storage business is in its early stages and we believe the market opportunity is largely untapped,” said Eli Harari, Chairman and CEO of SanDisk. “msystems is a leader in flash memory systems addressing mobile, portable and embedded markets and they have a strong team, significant IP and important OEM customers. SanDisk has a record of creating new market categories, worldclass manufacturing capabilities and leading market share in the retail channel. Both companies are noted for their innovation, and this acquisition is intended to further accelerate our pace of innovation. In the near term, this transaction better positions SanDisk to serve the expanding storage needs of handset manufacturers and mobile network operators. In the long term, the combination with msystems will be a catalyst in the development of next-

generation flash-enabled consumer applications. We are extremely excited about joining forces with the msystems team to achieve our shared vision. We are committed to serving msystems’ OEM customers after the transaction closes.”

Altera, AMD, Sun and XtremeData Launch Computing University Program

Altera has announced the development of a new university program to support academic research into high-performance computing. AMD, Sun Microsystems and XtremeData are participating in the program that will donate $1 million in workstations and development software to universities. Using the workstations, participating universities will be able to research and drive the adoption of FPGA co-processing for high-performance computing applications such as medical imaging, data analytics, text searches, network security, bioinformatics and energy. Twenty Sun Ultra 40 workstations, each powered by single or dual-core AMD Opteron processors with Direct Connect Architecture and an XtremeData XD1000 FPGA coprocessor module, are being made available under the program. The XD1000 coprocessor module includes Altera’s largest Stratix II FPGA, the EP2S180. The FPGA module is pin-compatible with an AMD Opteron processor and allows researchers to speed up algorithms running on the Sun platform by up to 100 times and applications by up to 10 times. The University of Illinois at Urbana-Champaign, home of many of the earliest and largest computer systems since 1952, is the first university to receive workstations through the program. The workstations will complement the Trusted ILLIAC, a 500-processor programmable hardware/software cluster that utilizes FPGA coprocessors to make large-scale computing more reliable and secure.

RadiSys and GoAhead Partner for Pre-Integrated High-Availability ATCA Solution

RadiSys and GoAhead Software have announced the first standards-based, pre-integrated high-availability Advanced Telecom Computing Architecture (ATCA) solution. The integration of RadiSys Promentum, a common managed platform for network element and dataplane applications, with GoAhead SelfReliant, standards-based high-availability middleware, is intended to help equipment manufacturers accelerate time-tomarket and reduce project costs. The integrated solution includes validated, out-of-box capabilities such as platform management and comprehensive high availability and system management. Areas of focus include resource discovery and system model instantiation, shelf manager integration, alarm management, hot swap management and integration with RadiSys-specific systems management capabilities. All of this is accomplished using the Service Availability Forum (SA Forum) Hardware Platform Interface (HPI). With this functionality, equipment manufacturers are able to realize the benefits of integration without having the burden of this significant undertaking themselves, enabling them to shift focus to the value-added application layer. One such customer is Operax, a provider of carriergrade solutions for Quality of Service (QoS) control in multiservice networks. Operax utilized both RadiSys and GoAhead for its next-generation triple play application, Bandwidth Manager 5500. This application is in the critical path for IP Multimedia Subsystem (IMS) services as well as non-IMS services such as IPTV. Service providers commonly regard it as a network element in terms of requirements for performance, scalability, availability and management.

SHARE YOUR VISION. WE’LL SHARE OUR INNOVATION.

solutions for the most demanding embedded applications. Tell us what Share your vision with us. We’ll customize our products to your requirements or partner with you to develop custom products all the way through deployment. Either way, you’ll get leading-edge board level

you need at info@vadatech.com. • Comprehensive Hardware/Software solution for Intelligent Peripheral Management Interface (IPMI) version 2.0 • Symmetric Multi-Processing CPU for AMC/VME Modules • A complete line of ATCA and AMC carriers • High dynamic range A/D Converters

w w w. v a d a t e c h . c o m

702.896.3337

exploration er your goal eak directly l page, the resource. hnology, and products

TechnologyInContext Biometric Subsystems

Biometric Access Control Gets Networked Once relegated to the realm of science fiction, the use of biometrics for access control is becoming a viable security solution, finding its way into a wide range of everyday applications. by R ahul Shah, Lantronix and Bashar Masad, Ingersoll Rand Security Technologies, Schlage Recognition Systems

he use of unique physical or behavioral characteristics, such as hand shape or a signature, to verify someone’s identity is called “biometrics.” Biometric access control systems consist of a reader or scanning device, software that converts the gathered information into digital form and a database that stores the information for comparison with previous records. These readers, or scanning devices, panies providing solutions now can scan for a fingerprint, hand geometry, ration into products, technologies and companies. Whether your goal is to research the latest iris/retina, facial ication Engineer,signature, or jump to a company's technical page,recognition, the goal of Get Connected is to put you voice print, ce you require for whatever type ofvascular technology, pattern and even es and productsDNA. you are searching for. This technology can be used for Figure 1 Biometric access control a number of applications including time is widely used in many and attendance reporting, building access security applications, such control, verification of signatures, pointas to check employees of-sale identity verification, process conentering nearly every trol security and cellular phone security. airport. Verification is a simple process for users (Figure 1). A PIN number is entered sult, the reader pulls up a template taken into a keypad, a magnetic stripe/barcode of the person’s biometric data at the time card is swiped or a proximity card is used of enrollment. Depending on the type of to touch the biometric reader. As a re- reader, the user places a hand on the unit or a finger on the window or provides another type of biometric input. If the resultGet Connected with companies mentioned in this article. ing template matches the stored template, www.rtcmagazine.com/getconnected the person is verified.

End of Article

September 2006 Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

Although this can be considerably more convenient than current access methods such as passwords and cards, many think of the technology as confined to heightened security applications. It is true that biometrics are used to check employees coming into almost every airport and to guard almost every nuclear plant. These access systems are also the mode of entry at embassies around the world. However, the majority of implementations are used in common, everyday locales including hair salons and restaurants.

Embedded Biometric Applications

Biometric sensors are now being embedded into systems such as cell phones or PDAs to customize features or settings and provide increased comfort and convenience. The stand-alone biometric reader, although seldom thought of as an embedded system, has been on the market for some time. A newer alternative features template management performed by a smart card. Breaches of security can be minimized in many ways. One method limits the opportunity for infringement. For in-

TechnologyInContext

Success Factors and Design Challenges

The main goal of any access control system is to keep some people out and allow others to get in. Although this sounds simple, some key factors must be considered early on when designing a biometric application. These include user acceptance, throughput, accuracy, encryption and identity theft aversion. User acceptance of the access control device is one of the most critical factors in the success of a biometric-based implementation. In order to prevent improper use, which can cause access errors, the device should not cause discomfort or concern and must be easy to use. Throughput, which is application-dependent, is the total time required to use the device. The elapsed time from presentation to identity verification is known as verification time. Most readers can verify identity within one second. However, when considering the use of biometrics for access control, the total time it takes a person to use the reader must be considered. This includes the time it takes to enter the ID number and the time required to get into the right position for scanning. The total time required for each person varies.

Comparison of Biometric Technologies Requirement

Fingerprints

Hand Geometry

Retina

Iris

Face

Signature

Voice

Ease of Use

High

Low

Medium

High

Changing signatures

Noise, Colds

Factors Increasing Error Incidence

Dryness, Dirt, Age

Hand Injury, Age

Glasses

Lighting

Lighting, Age, Glasses, Hair

Accuracy

High

Very High

High

User Acceptance

Medium

High

Long-Term Stability

High

Medium

High

Medium

Figure 2 Access control technologies have varying levels of effectiveness.

Accuracy is vital to the acceptance of the biometric type chosen. If it does not accurately read the person’s biometric input, the system will no longer be used for access control because of its inaccuracies. Letting the wrong people in or denying access to the correct people poses serious problems. The two errors a unit can make are false acceptance and false rejection. When the live biometric template is compared to the stored template, a matching score confirms or denies the identity of the user. System designers set this numeric score to accommodate the system’s desired level of accuracy, which is mea-

sured by the False Accept Rate (FAR) and False Rejection Rate (FRR). The FAR is the probability that an unauthorized user will be allowed to pass for someone else. This error rate must be low enough to present a real deterrent for a given application. In today’s biometric access control systems, FAR ranges from .0001% to 0.1%. In comparison, the biometric hand geometry reader used on the front entry area of 60% of U.S. nuclear power plants has a FAR of 0.1%. False Rejection Rate is just as crucial as FAR. The FRRs quoted for currently available systems vary from .00066% to 1.0%. A low FRR is important because

5% 4%

% Error Rate

stance, if the database of those authorized for access to a facility does not rely on networked hardware systems, the chance of someone infiltrating the system is reduced, as are the possibilities of downtime. Although new methods and attributes, such as vascular recognition and DNA matching, are promising, they are not yet mature enough for mainstream usage. Alternatives further along in development already exist and a variety of common methods are commercially available today. Selection of the appropriate biometric access control technology depends on a number of application-specific factors, including the environment in which the identity verification process is carried out, the user profile, accuracy requirements and overall system cost (Figure 2).

False Reject

False Accept

3% 2% 1% 0% 40

60 Increasing

Figure 3

100

120

Sensitivity

140

160

180

200

Decreasing

For a given biometric device, the point where false accept and false reject error curves intersect and equal one another is called the Equal Error Rate (EER). The EER’s corresponding sensitivity setting is shown on the lower axis: the smaller it is, the better. September 2006

TechnologyInContext

Figure 4

01_WirelessRTC_Ad.qxd

9/5/2006

4:08 PM

Page 1

wLAN Ho! E M P- 8 6 0 2

• Up to 400mW (26dBm) transmission power support

802.11a/b/g for maximum range, signal strength, and receive sensitivity

• Utilizes the Atheros AR5006 Solution (AR5413 Single Chip CMOS MAC/Baseband/Radio, 6th Generation)

• 802.11e standards supported for Wireless Multimedia Enhancements QoS

EUB-362 EXT

• Up to 200mW high output power radio and 2 dBi removable antenna (SMA RF Connector)

• High Speed Data Rate up to 108 Mbps in Super G mode and support 802.11b/g

• USB 2.0 interface E M P- 3 6 0 2

• Up to 250mW transmission power support 802.11b/g for superior range, signal strength, receiving sensitivity

• Utilizes the Atheros AR5005G Solution (AR2413 Single Chip CMOS MAC/Baseband/Radio, 5th Generation)

• Support for draft IEEE 802.11e, h, i and j standards Discover value and performance through EnGenius’ PowerGear leading technology of Long-Range, high powered ISM band embedded radios that provide greater coverage and increased deployment flexibility. Call today for a FREE catalog (RTC1006): 888-735-7888 EnGenius Technologies Inc. • wlan@engeniustech.com • www.engeniustech.com

September 2006

this type of error can occur with almost every use of the access control device. Error curves give a graphical representation of the device’s performance (Figure 3). The point where the false accept and false reject curves intersect and the error rates equal one another is called the Equal Error Rate (EER). The smaller the EER’s sensitivity setting, the better. Setting the system to produce a very low rate of false accepts usually causes a higher rate of false rejects. Conversely, if the system is set to produce a low rate of false rejects, the rate of false accepts is higher. The application determines the acceptable levels of false accepts and false rejects. Encryption keys and digital signatures enhance secure authentication in networked applications. For example, biometric information can be transferred using public key infrastructure systems that incorporate encryption to help make a networked system more tamper-resistant. The use of smart cards can help avoid identity theft. Since the card stores the user’s ID and biometric template, there is no database that can be hacked. When the smart card is presented to the smart card reader embedded in a biometric reader, the user’s hand is placed on the unit. The reader compares length, width, thickness and surface area measurements against the template stored in the smart card to verify identity. This process takes approximately one second and is virtually foolproof.

Networked vs. Stand-Alone Biometric Systems

Biometric readers can either operate as stand-alone systems or be integrated into a network environment. Stand-alone biometric systems have been employed

TechnologyInContext since the late 1980s. These are not only biometric readers, but also provide complete door control, including locking and unlocking the door based on who is authorized to enter and when. For example, the Schlage Recognition Systems HandKey II reader provides access for a very small group of people to a unique private library and museum at the New York Weill Cornell Medical Center. However, the majority of access control applications must manage more than one door. A network-enabled system of biometric readers can provide a central location where alarm conditions and door activities can be stored, viewed and controlled via browser on a PC located virtually anywhere. Networked systems also provide convenient template management and allow user profiles to be deleted or changed on the PC. This process also allows users to enroll at a single location while their templates can be made available at multiple locations. All major biometric readers can be integrated into conventional access control systems in a variety of ways. The most common way is card reader emulation, where the biometric device works with the access control panel in the same way a card reader does. The biometric readerâ&#x20AC;&#x2122;s card reader output port is connected to the panelâ&#x20AC;&#x2122;s card reader port. This method can be very effective for integration into existing card-based systems. Integrating biometrics with cards or smart cards is becoming common and provides dual authentication. By adding a biometric reader to the access control system, a badge alone cannot be used to gain access. Both the badge and the personâ&#x20AC;&#x2122;s biometric template are required, providing a higher level of security. Since the template only resides on the card, the solution also eases individual privacy concerns.

latest LANs (local area networks), WANs (wide area networks) and other networks, a TCP/IP stack, support for security and support for biometric hardware/software. Building Ethernet connectivity into a product requires a significant investment in hardware and software integration. Device servers provide an easy and economical way of connecting the serial device to the network. They allow independence from proprietary protocols and the ability to fulfill a number

of different functions. When the access control system is networked, the biometric template can be checked against a centralized database and records are made of all accesses made within the organization. For example, a biometric system can utilize the Lantronix WiPort embedded device server to develop a complete networked biometric system (Figure 4). This allows the direct monitoring and control of an almost unlimited number of

Success Factors for a Network-Enabled Biometric System

A number of requirements must be met to connect the biometric reader to a networked environment. These include an ultra-high level of reliability, an operating system capable of working with the September 2006

GE Fanuc Embedded Systems

We can save you time. How you spend it is up to you. With 50+ single board computers to choose from, we probably have what you’re looking for. Finding exactly the right single board computer can be a time-consuming task, but it can be lot simpler when you start your search at GE Fanuc Embedded Systems. We have such a vast selection that you’re likely to find what you’re looking for, no matter what the form factor, processor, ruggedization level, I/O or configuration. This PICMG® 2.16 compliant computer, for example, offers 3 Gigabit Ethernet ports, 16 programmable GPIO ports, a 167 MHz MPX system bus, and gives you the choice between a 1 GHz or a 1.4 GHz Freescale™ PowerPC® G4 processor. It can be

ordered with convection or conduction cooling, and it’s just one of a dozen of our off-the-shelf cPCI single board computers, most of which are available in multiple configurations. We not only offer you cPCI in 3U and 6U, Intel® or PowerPC with convection or conduction cooling, we also have a range of VME, PMC and AdvancedTCA® and AdvancedMC™ processor boards. Take a look at our selection. We’ve spent a lot of time pulling it all together so you don’t have to.

C2K High Performance PICMG 2.16 6U CompactPCI® PowerPC Embedded Computer

www.gefanucembedded.com

TechnologyInContext readers across the enterprise. This device server’s wireless technology connects directly to an IP network, eliminating the need for traditional security control panels and expensive wiring. A traditionally complicated configuration and installation becomes as simple as installing wireless access points on a network.

do not cost much and the cost of access cards has recently fallen, the true benefit of eliminating these methods is reduced administrative efforts. For instance, a lost card or key must be replaced and reissued, whereas hands and fingers are not lost, stolen or forgotten, and they do not wear out or need to be replaced.

Network-Enabled Hand Recognition Readers

The use of biometric hand geometry technology can eliminate the need to track employee time and attendance. Hand geometry readers can easily handle any size of user population volume while providing a high level of reliability, with dramatically lower FRRs and failure-to-enroll rates. Because each hand is unique, the system can eliminate time fraud and reduce payroll costs. In addition, employees tend to consider hand geometry less intrusive than biometric devices that read fingerprints or scan the eye. The HandPunch HP-4000 from Ingersoll Rand Recognition Systems automatically takes a three-dimensional reading of a hand’s size and shape and verifies the user’s identity in less than one second. In most applications involving smart cards and embedded biometrics hand geometry is the preferred technology, since a hand template uses up only 9 bytes versus an average of 300 bytes per finger required by fingerprint readers.

Future Applications

New technologies and developments aimed at high-security applications include using vein pattern, ear shape, body odor and body salinity for identification. Interest in vascular recognition, or vein patterns, and DNA for faster identity verification is growing. Since an individual’s blood vessels never change, vein pattern verification is more secure because vein patterns are unique to each individual. DNA, on the other hand, requires a more lengthy process of identification than other forms of biometrics. For users, the biggest appeal of biometric access control is the elimination of keys and the various types of access cards used. Although keys

Lantronix Irvine, CA. (800) 526-8764. [www.lantronix.com]. Schlage Recognition Systems Campbell, CA. (408) 341-4100 [www.schlage.com].

Extreme environments require extreme solutions...

The Challenge - Develop the next evolution of ATR cooling.. Mission Complete!

Liquid Cooled 100W/Slot 1 ATR 9 9 9 9 9 9 9 9 9

Top load 1ATR short Brazed construction Liquid cooled card cage side walls Accepts conduction cooled cards Cools up to 100W/slot power dissipation 8 Slot VITA 41 VXS or VME64x Backplane with custom I/O MIL-STD-704 800 Watt power supply MIL-STD-461 EMI

Hybricon Corporation is a true solutions company providing integrated system solutions, electronic enclosures, and backplanes for high performance military and ruggedized COTS applications. Hybricon has a successful track record of proven and deployed COTS solutions that have been designed to meet stringent military requirements. Hybricon Corporation 12 Willow Road Ayer, MA 01432 ISO 9001-2000 Certified

www.hybricon.com Call Today 1-877-HYBRICON September 2006

SolutionsEngineering VME 25th Anniversary

VPX and VPX-REDI Standards Coming on Strong The latest set of standards expected from the VITA Standards Organization (VSO) will address the support of robust serial interconnects over the backplane along with ruggedized specs that provide for a selection of cooling methods. by S tewart Dewar, Chairman VITA 46 Working Group

Ad Index

Get Connected with technology he VPX and VPX-REDI standards companies mally known as VITA 48, offersand extended then a range of subsidiary specifications, providing solutions now are currently a major focus area mechanical configurations designed to commonly called “dot specifications,” afGet Connected is a new resource for further exploration for the VITA Standards Organi- into support higher functional density. products, technologies and companies. Whether Being your goal ter the convention in which they are numzation (VSO). VPX, formally referred targeted primarily is to research the latest datasheetfor fromharsh-environment a company, speak directly bered. The base specification for VPX is an Application Engineer, orapplications, jump to a company's technical the to as VITA 46, offers 6U and 3Uwithboard embedded VPX and page, VPXVITA 46.0, “46” being derived from the goal of Get Connected is to put you in touch with the right resource. formats with a modern high-performance REDI go hand in hand to enable defense number of the VSO working group develWhichever level of service you require for whatever type of technology, connector set that is capable of supportsystem integrators to fully exploitand today’s Get Connected will help you connect with the companies products oping the VPX standard. Figure 2 shows ing today’s plethora of high-speedyoufabric advanced component technology. a representative VPX module in a conducare searching for. www.rtcmagazine.com/getconnected interfaces such as Serial RapidIO, PCI Figure 1 illustrates the suite of docu- tion-cooled format. Express, 10 Gigabit Ethernet and others. ments that together make up the VPX VITA 46.0 defines the basic mechanWhile VPX maintains the very familiar standard. As with other backplane stan- ical and electrical underpinnings of VPX: mechanical and dimensional aspects of dards, such as AdvancedTCA (ATCA), • 6U and 3U board sizes and mechanical 6U VME and 3U CompactPCI such as VPX uses the concept of a base specificadetails based heavily on IEEE 1101.1 the standard 0.8” pitch, VPX-REDI, for- tion to define the common elements and and IEEE 1101.10 for air-cooled cards Get Connected with technology and companies providing solutions now and IEEE 1101.2 for conduction-cooled Get Connected is a new resource for further exploration into products, technologies and companies. Whether your goal is to research t cards—all on a 0.8” pitch datasheet from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connect the usetypeofof technology, the MultiGig 7-row RT2 in touch with the right resource. Whichever level of service you require for• whatever 46.10 Get Connected will help you connect with the companies and products youconnector, are searchingrated for. for signaling rates of VITA 46 Module Rear Transition Test Data www.rtcmagazine.com/getconnectedModule up to 6.5 Gbaud • the use of a mechanically robust alignment and keying system—the alignment aspect brings the payload card into alignment with the backplane slot before the connectors mate, while the 46.xx System 46.20 46.1 46.3 46.7 46.4 46.9 keying aspect prevents cards from beManagement Switch Slot VME SRIO 10GbE PCIe XMC/PMC IO ing inserted into an incorrect slot accidentally

Products

End of Article

VITA 46.0 (Base Standard)

Figure 1

Get Connected with companies and VITA 46 Specification Hierarchy products featured in thisDiagram. section. www.rtcmagazine.com/getconnected

Get Connected

with companies mentioned in this article. www.rtcmagazine.com/getconnected

September 2006 Get Connected with companies mentioned in this www.rtcmagazine.com/getconnected

High-Performance 3U CompactPCI Intel® Pentium® M / Intel® Celeron® M Processor SBC with Dual GbE on PCI Express

ETXexpress Module with Intel® Pentium® M and Mobile Intel® 915GM Express Chipset As the first member of the ADLINK ETXexpress family, the ETXexpress-IA533 uses low power Intelfi Pentiumfi M processor 760 at 2.0GHz and the Mobile Intel fi 915GM Express Chipset. Both the processor and chipset are part of the embedded Intelfi Architecture that ensures a long production life for applications that need extended availability. The ETXexpress-IA533 supports dual channel DDR2 533MHz memory and comes with a single on-board Gigabit Ethernet port. In addition to the on-board integrated graphics, a Graphic PCI Express x16 slot is also available. The board connects up to four additional PCI Express x1 devices. The module has legacy support for 32-bit PCI and ISA through LPC.

The cPCI-3915 CPU board is a 3U 4HP or 8HP system slot single-board computer with up to 2.0GHz for the Intelfi Pentiumfi M CPU. This SBC is based on the Intelfi 915GM chipset with 400 or 533 MHz FSB. One SO-DIMM socket carries up to 1GB 400/533 MHz non-ECC DDR2 RAM. The cPCI-3915 offers two Gigabit Ethernet ports, two RS-232 ports, one IDE ATA 100/66/33 port, one CompactFlash socket, two SATA-150 interfaces, quad USB 2.0, and a VGA/DVI interface. The Ethernet ports support PXE remote boot. For more info, go to: www.adlinktech.com/products

Come Visit Us at ESC Boston at Booth 1716

For more info, go to: www.adlinktech.com/products

Two AdvancedTCA Processor Blades with Dual Intelﬁ Xeon 800FSB

Full-size Intel® Pentium® 4 (LGA775) Processor SBC The new NuPRO-851DV has a high-computing capability. It supports an 800MHz FSB and Intelﬁ Pentiumﬁ 4 (LGA775) processor with Hyper-Threading technology. It has a large communication bandwidth with two PCI Express x1 Gigabit Ethernet ports. It supports Serial ATA for high-speed storage, USB 2.0 and other generic features including VGA, COM, keyboard, mouse and hardware monitoring. For more info, go to: www.adlinktech.com/products

� PICMG 3.4 ATCA Advanced Switching Option3 � Quad ASI x4 Fabric Interface Channels up to 5-Slot Full-Mesh Configuration, 32Gbps � Dual Intelﬁ 64-bit LV-Xeon 2.8GHz, 1MB L2 Cache, Hyper-Threading Processors � Dual DDR2-400 REG/ECC Channels with 16GB max.

AdvancedTCA Base/Fabric Switch Blade with 10GbE Uplink Port � 24-Port 10/100/1000BASE-T Base Interface: 12 Node-Slot, 8 Egress, 2 ShMC Ports, 1 Inter-Switch and 1 Control Plane Port � 24-Port 1000BASE-BX Fabric Interface: 14 Node-Slot, PICMG 3.0 Option1 Ports, 2 Egress and 8 RTM Ports � Layer 2/3 Packet Switching and Routing � Vitesseﬁ VSC7310 GbE Switch-on-a-chip 5U Five-Slot AdvancedTCA Rackmount System

A Compact Multi-Application Computer with New CPU Options, and the Latest Chipset The GEME-4000 offers multiple CPU choices and USB 2.0 connectors, LVDS support for display and DDR RAM in addition to the standard GEME features. The GEME-4000 comes with an Intelﬁ Ultra Low Voltage Celeronﬁ M 600MHz processor. They can be controlled from remote locations and run continuously in many critical applications such as machine tools and digital video capture. It also comes equipped with anti-vibration features. For more info, go to: www.adlinktech.com/products

� 19" Rackmount, 15.3" Depth � Five-Slot Full Mesh Topology Fabric Interface � Five-Slot Standard Rear Transition Module � Dual Bussed IPMB

Call us toll-free at (866) 4-ADLINK or email info@adlinktech.com

www.adlinktech.com

SolutionsEngineering • definition of voltages and current capacities of the power rails • utility signals such as geographical addressing, SYSRESET*, Reference Clock and JTAG • signals allocated for a system management bus implementation The VITA 46.0 specification provides a highly capable baseline for module implementations. Some notable aspects of the VPX connector system as implemented on a 6U module include: • A total of 707 non-power electrical contacts • A total of 464 signal contacts (nonpower, non-ground), providing for: - 64 signals implemented as 32 highspeed differential pairs for core fabric - 104 signals used for a full VME64 implementation - 268 for general-purpose user I/O including 128 high-speed differential pairs - 28 for system utilities (reset, geographical addressing, etc.) and spares • Each high-speed differential pair is rated up to 6.5 Gbaud The large number of high-speed differential pairs supported by VPX modules allows for a rich complement of backplane I/O including: multiple ports of high-speed fabrics such as PCI Express, Serial RapidIO, 10GbE XAUI and InfiniBand; mass storage interfaces such as Serial ATA and high-speed digital video; and special-purpose interconnects such as RocketIO from Xilinx FPGAs. The VPX specification hierarchy presented in Figure 1 illustrates a range of

Figure 2 22

Representative VITA 46 Module. September 2006

optional elements for a VPX module, one or more of which need to be implemented to provide a functional module. For 6U VPX modules an implementation of a full A32:D32 VME64 bus is defined within the VITA 46.1 specification. Such an implementation enables a VPX module to provide the performance benefits of a modern fabric-connected module while still maintaining the ability to interface to standard VME modules, assuming that an appropriate hybrid backplane is provided. This eases the transition to VPX by allowing previous investments in VME modules to be leveraged for new systems. Integral to the working concept of VPX is what is referred to as the core fabric interface. A 6U VPX module is specified to have 4 core fabric ports, each port having 4 lanes (a Tx and Rx pair) in each direction. The allocation of 4 lanes per port fits naturally with Serial RapidIO, 10GbE XAUI and 4-lane implementations of PCI Express and InfiniBand. By provisioning for four (4) ports of fabric, a VPX module provides enough fabric connectivity to implement a large range of systems without the need for a dedicated switch slot—an important consideration for space, weight and power (SWAP)-sensitive systems often associated with military/aerospace applications. To work together in a system all of the VPX modules must use the same core fabric, and while from one perspective it may have been ideal if one, and only one, standard fabric could have been chosen for VPX, due to the range of application scenarios envisioned for VPX no such “one size fits all” approach was deemed possible. So, like ATCA, a range of fabric options is provided for VPX. Currently, dot specifications detailing the implementation of Serial RapidIO (VITA 46.3) and PCI Express (VITA 46.4) are available for VPX. A placeholder, VITA 46.7, for an eventual 10GbE implementation is defined. When other fabrics become of interest to the working group additional dot specifica-

tions can be created. The core fabric dot specifications contain two main types of information. The most fundamental information they contain are pin-out tables for the specific fabrics. This establishes a basic level of interoperability among backplanes and modules—ensuring that the Ins and Outs of the signals have been connected properly. However, for true multi-vendor interoperability in a set of cards connected together by a serial fabric such as Serial RapidIO, it is necessary to ensure that the serial links are meeting the required signal integrity budgets. This guarantees that when signals arrive at the receiver inputs they are of sufficient amplitude and low enough noise to be reliably received. Thus the fabric dot specifications define signal integrity budgets for VPX plug-in modules and backplanes so that system integrators can mix and match products from multiple vendors with confidence that they will work together reliably. Support for PMC and XMC modules is an important element of VPX—in fact a key selection criterion for the 7row RT2 connector is that its dimensions are compatible with the use of a standard-length PMC/XMC module. To facilitate multi-vendor interoperability the VPX series of standards includes VITA 46.9, a specification that standardizes the mapping between the Pn4 and Pn6 I/O connectors of PMC/XMC modules and the VPX backplane connectors. VITA 46.9 includes a definition of how pins should be paired to support the routing of differential signals from Pn4 and Pn6 to the backplane. This will allow makers of PMC/XMC modules that incorporate high-frequency signals such as Fibre Channel or digital video to be able to design their module with confidence that—from a signal integrity perspective—it will work on any VITA 46.9-compliant basecard. The scope of VPX also includes the definition of a System Management (SM) bus and backplane signals are reserved for that purpose. The VPX SM bus will be fully defined in an eventual addition to the VPX suite of standards. The VPX SM bus specification, referred to as 46.xx in Figure 1, will be tailored to the requirements of embedded systems and will carefully consider the

SolutionsEngineering tradeoffs between functionality and the amount of board real estate required for the SM interface circuitry. Some of the capabilities expected in an eventual SM implementation are the collection of BIT results and the real-time monitoring of operating conditions such as temperature and power supply voltages. While the rich fabric connectivity supported by VPX is expected to enable a high percentage of applications to be implemented by directly connecting modules together in mesh-based architectures, there will be VPX applications that require a centralized switch. The definition of a VPX switch slot will be encapsulated in the VITA 46.20 specification and encompass definitions for both single-fabric and hybrid (fabric plus GbE) switches. Given that VPX is targeted to harsh-environment military/aerospace applications, it was very important to substantiate that the module would hold up to the intended application environments. To this end, the VITA Standards Organization (VSO) VITA 46 Working Group successfully conducted an extensive third-party qualification program including the following environmental parameters: • Vibration (power spectral density of 0.1 g2/Hz from 50 Hz to 2000 Hz, 1.5 hours/axis) • Shock (50g, 11 msec) • Temperature (-40°C to 100°C) • Humidity • Durability (500 mate/unmate cycles) • Dust and sand • Electrostatic discharge (ESD) protection (EN 61000-4-2)

usable PWB area and improved thermal management (air-cooled, conductioncooled, liquid flow-through cooling), which enables increased functional density. VPX-REDI also supports use as a line replaceable module (LRM). The VPX-REDI formats are described in a suite of specifications consisting of VITA 48.0 (common elements), VITA 48.1 (air-cooling), VITA 48.2 (conduction-cooling) and VITA 48.3 (liquid flow-through cooling).

Concurrent with the finalization of the VPX and VPX-REDI standards within the VITA Standards Organization, product development and announcements from the vendor community are continuing. In addition to a growing list of VPX module vendor product announcements there have also been chassis from Elma Bustronic, Parker-Hannifin and Hybricon. More information on VPX and VPXREDI can be found at www.vita.com.

The test results are publicly available and may be downloaded on the VITA Web site. As mentioned above, the VPX mechanical formats closely follow those of IEEE 1101.x and preserve the 0.8” pitch that is the standard for VME and CompactPCI. VPX-REDI (Ruggedized Enhanced Design Implementation) is a complementary mechanical standard to VPX, and defines alternate mechanical formats that go beyond the traditional IEEE 1101.1 and .2 formats to provide new enhanced capabilities. These new capabilities include an increase in September 2006

Extreme performance across the board. NEW! TP32™ (Intel Pentium-M based SBC) 2.0 GHz Intel® Pentium-M® processor 4GB DDRII-400 SDRAM VME interface Two Gigabit Ethernet ports Two Ultra320 SCSI ports Up to four PMC slots XMC support on optional PMC carrier card Four USB and four serial ports Solaris™ 10, Linux and Windows support PS/2 keyboard and mouse port Up to 30G shock

NEW! TA64™ (AMD Turion 64 Mobil-based SBC) Low Power, AMD® Turion™ 64 Mobile CPU Single/2-slot VME configurations Up to 4GB ECC DDR memory VME interface Four PMC slots, two USB ports PS/2 keyboard and mouse port Compatible with USPIIe-USB™ SBCs 64-bit Solaris 10, Windows® and Linux support Up to 30G shock

NEW! TPPC64™ (IBM PowerPC 970FX-based SBC) 6U VME64 engine with up to two IBM® 1.8 GHz PowerPC® 970FX CPUs Up to 4GB of DDR400 SDRAM Universe II VME64x interface Two Gigabit Ethernet ports Six USB ports Two serial ports Support for Linux™ OS Up to 30G shock

When the highest performance is critical to your mission, turn to Themis SBCs. In mission critical applications, there’s no substitute for high performance. Themis single board computers have proven themselves time and again in the most demanding applications and operating environments. The Themis family of single board computers includes 64-bit AMD, Intel and IBM PowerPC processors in addition to our leading UltraSPARC products on VME and Compact PCI. So we can support applications in Solaris, Windows, Linux and UNIX®.

All Themis products offer maximum configuration flexibility and life cycle support for your technology refresh cycle process, reducing your Total Cost of Ownership. So when mission success depends on higher performance, you can rely on Themis. Across the board. www.themis.com (510) 252-0870

Transformational. © 2006. Themis Computer, Themis, Themis logo, TP32, TA64 and TPPC64 are trademarks or registered trademarks of Themis Computer. All other trademarks are property of their respective owners.

SolutionsEngineering VME 25th Anniversary

VMEbus at 25! A continuity of specifications across a quarter of a century has enabled VME to grow and adapt new technologies and increase performance while keeping compatibility with established designs.

by C larence Peckham GE Fanuc Embedded Systems

ME at 25. I don’t think anyone would have expected that to happen! Good planning? Good luck? Actually I believe the answer is pretty simple. The VMEbus specification was written to meet the needs of the market and has been modified since the early 80s to continue to meet the requirements of the users. The VME specification was developed by the VITA Standards Organization to support early 16-bit microprocessors—most notably the Motorola 68000. These processors required a lot of MSI/ LSI technology support logic. The earliest nies providing solutions now VME systems utilized separate boards for on into products, technologies and companies. Whether your goal is to research the latest andtechnical I/O with tion Engineer,CPU, or jumpProcessor to a company's page,all the commugoal of Get Connected is to put you nications between the boards occurring you require for whatever type of technology, and productsover you are for. thesearching VMEbus. In the VME specification, support was provided for both 3U (160 x 100 mm) as well as 6U (233 x 160 mm) board forFigure 1 Initial VME implementations consisted of one or more SBCs plugged into a backplane with I/O cards. For commercial applications, these mats. The 3U board form-factor never were mostly air-cooled, though many military applications were also gained much popularity for two main reahoused in conduction-cooled ATR chassis. sons. First, the single DIN connector only allowed for a 24-bit address and a 16-bit data bus—adequate, but only for the 8/16- single DIN connector on the 3U board did well as providing 64 I/O pins on the secbit processors of the time. The second, not have any available pins for I/O signals ond DIN connecter. and more important reason, is that the on the backplane. This left only the front As technology developed and propanel available for I/O connectors. The grammable logic and 32-bit processors 6U format, on the other hand, was much became readily available, the VMEbus Get Connected with companies mentioned in this article. more user friendly, with its support for and 6U board form-factor became the www.rtcmagazine.com/getconnected 32-bit address and data implementation as platform of choice for building 16- and

loration our goal k directly age, the source. ology, products

End of Article

September 2006 Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

SolutionsEngineering

Figure 2

design.

develop. d

The addition of mezzanine cards in the form of PMC and, later, processor PMC modules allowed customization and the addition of enhanced capabilities to VME SBC cards.

deploy.

Are You Board With Me?

design. The next generation of connected devices.

Start with an AMCC PowerPC 405 Processor. Specify your I/O requirements.

develop. Your products based on our platform.

deploy. Your solution faster. 26

Choose your operating system. Relax while Embedded Planet delivers your rugged and reliable system. 4760 Richmond Rd / Cleveland, OH 44128 Tel: 216.245.4180 / Fax: 216.292.0561 www.embeddedplanet.com

September 2006

EP885 Stack Advert V4.indd 1

1/11/2006 1:24:11 PM

32-bit systems. Eventually the technology reached a point where the processor memory, disk and Ethernet I/O were on one board for performance reasons, and the VMEbus was typically used to communicate with I/O cards or between processor cards since the bus architecture supported multiple processor cards via an arbitration system. One of the highlights in the history of the VMEbus was the availability of the VME interface chip. The first globally available interface chip was the VIC chip designed by a consortium of board vendors and Cypress Semiconductor. Other chip solutions followed, but it was the VIC chip that released the board designer from designing a VME interface based on programmable logic designs. The VIC chip, even with its “warts,” provided a common interface, one that provided the biggest advance in interoperability on the VMEbus since the specification was released. Since the VIC chip there have been other designs all leading to the latest VME interface chips from Tundra Semiconductor—The Universe II and Tsi148 Chips. Another feature of the VMEbus is that it is the only standard that encompasses environmental requirements as well as mechanical and electrical specifications. The VITA 47 specification provides support for all of the environments required for VME-based systems—both conduction- and convection-cooled. So where are we with the VMEbus specification? Well if you want to build a system based on PowerPC or Intel Pentium processors with lots of I/O, graphics or communications devices, the released specifications provide a variety of solutions. In Figure 1, a traditional use of the VMEbus, one or more processor cards and a card cage full of I/O and other function cards is shown. In the Figure a commercial type configuration is shown for air-cooled applications and a military conduction-cooled system is shown as well. For the first fifteen years these configurations would have been the majority of the systems. In the past ten years the expansion of the VMEbus to include support for faster transfers (VME64/VME 2eSST) and PCI peripherals with the inclusion of PMC modules on all cards has opened

SolutionsEngineering

Single Board Computers

InfiniBand Switch

Legacy VME Boards

Figure 3

Newer VITA 41 cards can take advantage of serial switched fabrics, such as InfiniBand, while maintaining backward compatibility with legacy VME boards.

VME Bus

up new configurations of VME systems. The addition of the P0 connector provided more I/O pin options, as well as allowing other connection methods to be implemented. The end result is that each card can now be considered a system. A single board computer can have two PMC cards (Figure 2) that can provide extended capabilities such as high-performance graphics, FPGA-based computing, I/O or another processor using a processor PMC. The end result is that the VMEbus chassis provides interconnect between the “system” cards and supports the environmental requirements of the application. So what is available today to make VME still an attractive solution for building systems? The basic VME specification still provides a viable solution for both commercial and military applications. The list of features that make VMEbus a viable solution are: • Basic 32-bit address/32-bit data bus supporting data transfers from 8 bit programmable I/O to 320 Mbit/s block transfers • Ability to support multiprocessor solutions sharing common I/O assets • Proven environmental specifications for convection as well as conduction-cooling. VMEbus systems have been qualified for many ruggedized military applications and deployed in many military platforms • Continued support of technology advances such as switched fabric backplanes via VITA 41.x family of specifications • Large supplier base providing board and system level solutions

The most powerful DSP to ever hit the screen

BittWare’s new 6U VME / VXS board. Ruggedized. Conduction-Cooled. COTS Compliant. The military has specific needs. BittWare has the solution. The new T2-6U-VME board (T2V6), which supports the VITA 41 VXS (VME switched serial) standard, is the very first COTS VME-based board featuring eight ADSP-TS201 TigerSHARC DSPs from Analog Devices, providing up to 28.8 GFLOPs in a single slot. Designed and built for demanding real-time signal processing systems, the T2V6 is particularly effective in applications such as sonar, radar, imaging, and communications. Available in both air-cooled and conduction-cooled versions, it’s ready for Airborne, Ground, or Naval environments. The T2V6 is the newest member of the T2 Family of boards from BittWare which includes PCI, cPCI, and PMC form-factors. Building on BittWare’s decade-long expertise in SHARC-based products, the T2 Family is supported by a complete line of development software and run-time tools.

What’s on your screen?

www.bittware.com TigerSHARC DSP is a registered trademark of Analog Devices, Inc.

September 2006

SolutionsEngineering

The key factor in the VMEbus future is technology insertion.

• Backward compatibility. Older VMEbus boards can be used with the latest approved VMEbus specifications. Allows reuse of older I/O boards in new systems The key factor in the VMEbus future is technology insertion. Moving from simple bus model and adding block transfers, VME64 extensions and support for 320 Mbit block transfers, were all major enhancements. However, the addition of support for switched fabric backplanes via the VITA 41 family of specifications was a major shift in technology. Still supporting legacy VMEbus, but at the same time providing support for high-performance serial links such as InfiniBand or serial

Rapid I/O, enables VITA 41 to support much higher performance systems. Figure 3 is an example of an InfiniBandenabled system supporting multiple processors and legacy VME boards with the ability to connect multiple systems via InfiniBand. What is being developed for the future? VME-based products will continue to support serial fabrics as a method of transferring high bandwidth and large packets of data. Future developments rest on the latest specification, and the one that is the most dramatic for the future of VMEbus—VITA 46. The VITA 46 specification is the first enhancement that will not support legacy boards or systems because a new

high-performance connector structure is defined as well as support for fabric switches. The 3U and 6U form-factors are maintained in VITA 46 but there are a couple of options on card spacing. Since VITA 46 will not support legacy VME cards, any requirement for I/O that was available will require a relay out of the card. On the surface this sounds like a less than ideal solution, but with the new RoHS requirements as well as the need to provide support for another 10-25 years for VITA 46 applications, it makes sense to refresh some of the older technology VME cards. Another VITA specification that will affect future VMEbus efforts is

PCI Express SHBs COVER PAST, PRESENT and FUTURE NEEDS

Choices between old and new bus architectures and option card technologies have you frustrated? Relax,we got you covered! Trenton's system host boards • Single Processor SHBs with Intel Pentium 4, Intel Pentium M and Intel Core ®

(SHBs)combined with our PICMG® 1.3 backplanes simplify system designs by merging legacy solutions with today's latest computing innovations. In addition to the latest PCI-X and PCI Express™ system architectures, our SHBs and backplanes support legacy ISA and PCI buses and option cards. Visit us today at www.TrentonTechnology.com to learn more about our growing line of PCI Express products.

FOR THE LATEST INFORMATION: TRENTON Technology Inc.

• • • •

Phone: (800) 875-6031 or (770) 287-3100 • Fax (770) 287-3150 • www.TrentonTechnology.com

September 2006

Duo Processor Dual Processor SHBs featuring multiple Intel® Xeon® processors Server-class and Graphics-class PCI Express configurations PCI Express™ x16, x8, x4 and x1 link support PICMG® 1.3 Compatible System Host Boards (SHBs)

™

SolutionsEngineering VITA 47. Efforts by the embedded industry to set uniform standards for ruggedization levels (temperature, shock, vibration, etc.) were, at best, hit-andmiss until VITA published ANSI/VITA 47-2005. This standard sets a clear, precise and comprehensive set of guidelines to which every manufacturer in the industry can adhere. Perhaps more importantly, it allows customers to compare products from various manufacturers by applying the same set of benchmarks to everyone. Now that a uniform, open standard has been approved, ANSI/VITA 472005 should make it possible for both the makers and consumers of rugged electronics to compare products from multiple vendors with far more accuracy and simplicity. However, due to the historical lack of consistency, it will continue to be critical for some time (possibly years) for customers to clarify with their suppliers the precise levels of ruggedization for individual products that were released

prior to the ratification of VITA 47. For products produced going forward, though, the process of compliance verification will become much more straightforward. In some ways, VITA 47 was a matter of organizing and sorting out an existing body of standards and test or manufacturing methodologies. Those explicitly defined by VITA 47 are often based on earlier MIL specifications. When the commercial off-the-shelf (COTS) movement took effect, many military specifications were no longer updated, and so VITA 47 was partially an effort to reaffirm or redefine many testing methods, manufacturing methods and certification levels that had lapsed or fallen into a certain degree of disarray. In some cases VITA 47 was also a kind of negotiation among the members/vendors as to the levels of ruggedness that should be formalized— usually levels that were already in use. What is the outlook for the next 25 years for VME-based applications? Will

traditional VME bus systems continue to be used on new design wins? Will VITA 46-based systems provide a new line of design wins? These are all tough questions to answer. Clearly VME-based systems will continue to ship for the foreseeable future as there are many current design wins that use the system. Overall I do not see a lot of competition for VME, but users will have to consider competing specifications when they review VITA 46 since it is all new. That may open the door for CompactPCI or MicroTCA system designs. Overall, I predict that VME-based applications will be shipping for at least another ten years. After all, VMEbus is a proven specification, there are hundreds of diverse products that can be used to build a system and users know how to construct systems using VMEbus specifications. GE Fanuc Embedded Systems Huntsville, AL. (256) 880-0444. [www.gefanuc.com].

See

GE FANUC EMBEDDED SYSTEMS at the RTECC JT Daugherty Conf Center October 17, 2006 Pax River You’re invited to be a special guest of GE Fanuc at the Pax River

Real-Time & Embedded Computing Conference (RTECC). Talk face-to-face with experts about why the basic VME specification still provides a viable solution for military and commercial applications. With VME celebrating a 25th birthday, now is the time to find out it’s future. Patuxent River To view the full exhibitors’ listing and all open-door technical seminars you can attend for free, go to www.rtecc.com/paxriver. Pre-register online to have your badge waiting when you arrive. Walk-in attendees welcome; bring your colleagues to take full advantage of this one-day conference. Event hours are 8:30 am–3:00 pm. Seminars, Exhibits, Parking and Lunch are Complimentary!

www.rtecc.com September 2006

SolutionsEngineering VME 25th Anniversary

VXS: Jump Start to Higher Performance for VME Users VXS provides a path to get substantially more performance into systems, but at the same time provides an opportunity for reuse of existing hardware, software and experience by leveraging familiar VMEbus technology. by D ave Barker VMetro

Ad Index

ne of the reasons VMEbus has Get adding features intended to stretch Connected with technology and perforsolutions thrived for over 25 years is that the companies mance toproviding new levels helpsnow designers meet technology continues to evolve to Get challenges asresource adding new exploration system Connectedsuch is a new for further products, technologies and companies. Whether your goal meet increasing performance demands into functions and reducing processing time. is to research the latest datasheet from a company, speak directly over time, yet preserves a base of compat- This is an issue in many programs where with an Application Engineer, or jump to a company's technical page, the ibility. This compatibility lowers goal risk,of leVME is isused. Because life of Get Connected to put you in touch of withthe the long right resource. verages the learning and experience of de- levelthese programs, refreshes are Whichever of service you require technology for whatever type of technology, Get Connected will help you with more the companies and products velopers, and keeps system development required to connect provide functionality you are searching for. costs down by allowing existing com- and performance without the necessity of www.rtcmagazine.com/getconnected ponents to be reused. At the same time, undergoing a “forklift upgrade.” Traditional VME system with a bottleneck between DSP boards

A recent evolutionary development in VME Technology is VXS, which can help existing VME users to “jump start” their applications in exactly these ways. First, let’s explore the basics of the VXS technology. Then, we’ll discuss two scenarios where users can use the technology to create new systems: incrementally adding processing to existing VMEbus I/O in a hybrid approach, and a complete remake of a system to take full advantage of the VXS high-speed serial links system-wide.

VXS Highlights Get Connected with technology and companies providing solutions now

VXS combines the event-driven parGetSBC ConnectedVME is a DSP new resourceVME for further and companies. Whether your goal is to research t VME DSP exploration into products, technologies datasheet from a company, speak directly with an Application Engineer, or jump to a company's page, the goal of Get Connect allel VMEbus with technical enhancements to supin touch with the right resource. Whichever level of service you require port for whatever type of technology, high-speed serial links with optional Get Connected will help you connect with the companies and products you are searching for. switch fabric protocols over a new highVME bus www.rtcmagazine.com/getconnected

VME I/O

System utilizing VXS to eliminate bottleneck between DSP boards

VME I/O

VME I/O VME SBC Products

VXS DSP

VME bus

Get Connected with companies and products featured in this section. www.rtcmagazine.com/getconnected

Figure 1 30

speed P0 connector. Several protocols are mapped out for VXS including Gigabit Ethernet, Serial RapidIO, PCI Express, InfiniBand and Aurora—Xilinx’s highspeed serial lightweight protocol. VXS maintains backward compatibility with existing backplanes without a conflicting legacy P0 scheme. VME’s parallel bus architecture provides bus control and maintenance data, handling everything

End of Article Get Connected

VXS 4x links

with companies mentioned in this article. www.rtcmagazine.com/getconnected

Moving from a VME system to hybrid VME/VXS.

September 2006

Get Connected with companies mentioned in this Get Connected with companies and products featured in this section. www.rtcmagazine.com/getconnected

www.rtcmagazine.com/getconnected

SolutionsEngineering

from single byte transactions to over 300 Mbyte/s block data transfers. Combining the venerable VMEbus with high-speed VXS links creates a powerful and flexible architecture suited to many high-end, embedded, real-time applications. VXS is fully ratified as ANSI/VITA 41, and products are shipping today. Key system benefits of VXS products are: • Improved bandwidth – VXS payload modules with two 4x VXS links can provide a total of 2.5 Gbytes/s of bandwidth using today’s high-speed serial and fabric technologies with 3.125 Gbit/s links, and bandwidth will increase to 8 Gbytes/s as the highspeed serial and fabric technologies evolve to 10 Gbits/s. Because VXS links are point-to-point interconnects, as opposed to the shared parallel architecture of the VMEbus, the total system bandwidth increases as the number of VXS payload modules is increased. • Backward compatibility to VME64x – By preserving the e lectrical and mechanical properties and locations of the VME64x P1 and P2 DIN connectors, and ensuring the new P0 connector does not interfere with non-P0 cards, standard VME64x backplane slots are created in VXS systems and insertion and use of legacy VMEbus cards is possible. • Increased power delivery – VXS

references ANSI/VITA 1.7, which studied the current carrying capability of the VME64x 5-row DIN connectors in depth and found them capable of supplying 2A per pin for nearly twice the +5V power as earlier thought for VME32 systems. With these benefits to designers, VXS has two common usage scenarios: hybrid systems designed to use existing VME I/O with an improved processing subsystem, and completely new systems built around the VXS high-speed serial interconnect architecture for both processing and I/O.

Hybrid Systems Step up Processing

A typical scenario is an existing VME-based system requiring a technology refresh to increase the system performance. Often, analysis indicates there VXS Switchless System

VXS backplane implementing a full mesh topology between three boards using 4x VXS links

VXS

VME bus

is a data throughput bottleneck in one particular part of the system that keeps it from achieving the required performance. How can the developers increase the performance in this one area of the system without doing a wholesale change of technology—the dreaded “forklift upgrade?” Keeping risks and costs low while incrementally increasing the performance capability of the processing subsystem would be the ideal solution. Incorporating VXS in a hybrid approach with legacy VME can help developers solve this problem. Since VXS uses the same P1 and P2 connector scheme, with a new P0 connector for high-speed serial VXS links, backplanes are easily designed with just a few VXS payload slots and several additional legacy VME slots. In these systems, users can employ the high-speed serial VXS links between the VXS slots, replacing slower interconnects such as VMEbus, VSB, RACE, RACE++, Ethernet, FPDP or sFPDP, and increasing the bandwidth available to high-performance applications such as digital signal processing. Interconnected directly by high-speed, low-latency serial links, VXS processor cards can exchange data quicker than was previously possible with an overlay bus. An example of a small hybrid system is shown in Figure 1. Often, the largest cost of forward migration is not moving the processor complex to new boards,

VXS Switched System

VXS backplane implementing a replicated star topology with seven VXS payload boards and one VXS switch board using 4x VXS links

VXS

VME bus

Figure 2

VXS example topologies.

September 2006

SolutionsEngineering but moving a large variety of I/O forward. Changing hardware, porting and revalidating driver and application software— whether it has been developed in-house or comes from multiple vendors—is a daunting task in many cases. The benefits of increasing processing power can be offset by the costs of moving the I/O. This is where the hybrid VXS approach pays off—its real advantage is that the I/O model remains unchanged. By incorporating legacy VME slots in the backplane, VXS users can simply drop in the I/O from their older system without hardware or software modifications. I/O cards can include existing VME boards or PMC modules used in PMC sites on VXS payload or VMEbus cards. This approach offers a way to step up to faster processing with increased bandwidth and reduced latency between processor cards, yet bring proven I/O technology forward quickly and cost-effectively.

Supercharged Systems Go Everywhere Faster

In another scenario, the opportunity to supercharge systems is worth the effort, and VXS can meet the challenge. In a system designed from scratch—a “clean sheet” design—users can benefit in a number of ways from designing a system architecture entirely around the VXS highspeed serial links. The obvious benefits are associated with the increase in available bandwidth for both processing and I/O as high-speed serial interconnects are made available through-

Figure 3

out the system. But VXS brings other important benefits. For instance, designers can eliminate the front-panel cables or backplane overlays they had to use to interconnect boards and move data throughout the system—the high-speed serial links that are inherent in VXS are designed to move large amounts of data between payload boards. Using VXS backplane connections also increases reliability, improves maintainability, and reduces system costs by eliminating extra interconnect hardware. Another important benefit is the new system architecture and design options that are now viable. Since VXS does not define a backplane topology, it allows for very creative backplane designs to optimize connections between VXS payload cards. In simple configurations, such as a small three-board mesh, VXS payload cards can be connected directly. In more complex configurations, processing and I/ O payload boards can connect to a switch board, increasing topology options as shown in Figure 2. As seen in the hybrid system we looked at earlier, using the high-speed VXS serial interconnects increases the available bandwidth between processor cards. As the system scales, another advantage becomes very clear—VXS transactions can occur simultaneously between multiple VXS payload boards, increasing aggregate bandwidth considerably over the parallel, arbitrated architecture of VMEbus. By redesigning the I/O subsystem to also leverage the highspeed VXS serial

One example of a VXS product available now is the VMetro Phoenix VPF1. It brings together dual Freescale 7447A processors with dual Xilinx Virtex-II Pro FPGAs. September 2006

interconnects, I/O speeds can be supercharged to levels not possible over the VMEbus. High-bandwidth I/O such as sFPDP, Fibre Channel and video can be moved across VXS links and distributed quickly to the processing complex. I/O cards can be feeding in new data while processed data is simultaneously being passed between boards on other VXS links.

Good Links Make Good Neighbors

There is yet another benefit of a clean-sheet design that may not be apparent at first. Prior to VXS, “on-board links” (such as PCI local bus) were much faster than “off-board links” (typically VMEbus or CompactPCI). As a result, designers would cram functionality and PMC sites onto a processing board in order have as many “near-neighbors” as possible. With VXS, there is no longer a need to cram functionality or to make use of PMC sites directly on processor boards solely to improve performance. With VXS, every board in the system is now a near-neighbor due to the eight high-speed serial links running at up to 3.125 Gbits/s for a total bandwidth of 2.5 Gbytes/s on each VXS payload board. The problems associated with distance and latency between boards no longer exist with VXS. This even extends to serial interconnects available to XMC modules—their functions can also appear as near-neighbors to every board in the system. In fact, any processor has low-latency, high-bandwidth access to any I/O function that is connected via high-speed serial links. I/O within a VXS system appears local no matter what its physical location in the system might be. With the concept of a near-neighbor redefined by VXS, the resulting benefits are more apparent. For instance, it is straightforward to design a system architecture with high-speed I/O and processing on separate VXS payload boards that utilize high-speed VXS links to pass data across the backplane. Other possibilities open up, such as utilizing a dedicated large memory board (perhaps with 32 Gbytes of SDRAM) with the ability to move up to 5 Gbytes/s of data on and off the board utilizing the full-duplex capability of the VXS links. These are but

SolutionsEngineering two examples of the numerous architectural possibilities available to designers as they look for new ways to apply VXS technology. We have focused on the benefits of the VXS high-speed serial interconnects, but lower speed control functions still exist in most systems. These control functions can be implemented over the parallel VMEbus. This allows the higher speed data paths on the VXS links to proceed with their tasks uninterrupted by supervisory or control messages. Conversely, with only control messages on the VMEbus, control functions have quick and ready access without waiting for data transactions to release the bus.

Starting with VXS Today

VXS is much more than a concept— the ANSI/VITA 41 standards behind it are mature, products are available today, and designers can start putting it to work right away. Many of today’s embedded, real-time Digital Signal Processing (DSP) applications utilize both PowerPC/AltiVec processors and

powerful FPGAs to effectively distribute the processing load. The Phoenix VPF1 offers a combination of PowerPC and FPGA processing in a single-slot design to provide a very powerful reconfigurable DSP processing engine. An example of how VXS is being utilized in embedded DSP applications is the Northrop Grumman Airborne Laser Mine Detection System (ALMDS). The MH-60 helicopter-borne system is housed in a pod attached to the helicopter. The system uses LIDAR (Light Detection and Ranging) blue-green laser technology to detect, classify and localize mines in shallow waters, helping surface ships to maneuver through clear sea lanes. The processing power for the system is provided by multiple VPF1 boards interconnected via a VXS switch board inside a liquid-cooled ATR chassis housed inside the pod. VXS and the VPF1 offer the high level of functionality and performance required in an application like ALMDS that has to move and process large volumes of image data in real time. Over the history of VME, it is clear

that the power of the technology is in its flexibility. Designers have taken the VME building blocks and interconnected them creatively to achieve their specific goals. The diversity of those implementations is quite impressive. VXS increases the options available to system designers. By leveraging VME and PMC I/O capability, incremental processing performance can be added while keeping the existing I/O functions intact. By allowing for new high-speed serial interconnects between VXS and XMC modules, processing and I/O can both be enhanced to much higher levels than previously possible. Some designers will choose VXS for the hybrid system approach. Others will choose VXS for the clean-sheet approach. In either case, designers will find VXS helps them extend their VME systems experience to meet the challenges of today and tomorrow. VMETRO Houston, TX. (281) 584-0728. [www.vmetro.com].

Accelerate your telecom application development. Ottawa October 3

Chicago October 10

Stockholm October 17

Munich October 24

Tokyo November 9

San Jose December 5

Boston December 12

Telecom Application Acceleration Seminars for Network Equipment Providers This free seminar focuses on all aspects of platform design and integration–from hardware components such as microprocessors, DSPs, open modular AdvancedTCA boards and shelves, to software components such as operating systems, protocols, middleware, and database technology.

Complimentary lunch-provided by your participating hosts

Seating is limited—Register on-line today! www.aaseminar.com Silicon & Software Systems

Some Facilities Provided by

Enea_adv13.indd 1

Kontron, Monta Vista and S3 are members of the Intel Communication Alliance. The alliance is a global community of communications and embedded developers and solutions providers committed to the development of modular, standards-based solutions based on Intel technologies. Learn more at www.intel.com/go/ica

9/11/06 4:06:08 PM 33 September 2006

SolutionsEngineering VME 25th Anniversary

VME: Moving from Evolution to Revolution VME has undergone a profound and compatible evolution over its 25-year life. Today that evolution is undergoing a deliberate and orderly development as new serial interconnects are adopted. The result is not a break but an extended ecosystem that promises to unfold for many years to come.

Ad Index

by D ave Evans-Hughes Concurrent Technologies

Get Connected with technology and companies providing solutions now

Get Connected is a new resource for further exploration into products, technologies and companies. Whether your goal ince its adoption in the early 1980s, ties have grown, so must communicais to research the latest datasheet from a the company, speak directly with an Application or jump to a company's technical the the VME architecture has evolved tionsEngineer, between them, particularly in page, backgoal of Get Connected is to put you in touch with the right resource. continuously, and remarkably suc- plane systems utilizing several intelligent Whichever level of service you require for whatever type of technology, cessfully, to keep pace with the improveboards. bus Get Connected will helpParallel you connect with technologies the companies andhave products ments in microprocessor and communicabeen you are searching for. stretched so far, but now the future

VME has moved from the 20 Mbytes/s backplane speed where it began, through 40 and 80 to 320 Mbytes/s offered by 2eSST technology. Each step has pushed the limits a little more, and www.rtcmagazine.com/getconnected tions technology. In the early 1980s the is plain—serial interconnects with ever- stretched the backplane signal timings so world was learning about PCs that used 5 increasing bandwidth are the choice for that skew or signal quality effects are inMHz CPUs, 640 Kbyte RAM and the ISA the future. VME has willingly embraced creasingly causing concern for further enbus. Now we are growing used to dual this change with the introduction of three hancements. Serial interconnects reduce core 2 GHz CPUs with 4 Gbyte RAM major standards: VITA 31, VITA 41 and the scale of these problems significantly, and PCI Express. As the CPU capabili- now VITA 46. and allow not just improved speeds now Get Connected with technology and companies providing solutions now (in excess of 1 Gbyte/s), but the promise Get Connected is a new resource for further exploration into products, technologies and companies. Whether your goal is to research of significant increases to come. This is datasheet from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connec to ofsome extent by increased in touch with the right resource. Whichever level of service you requirecountered for whatever type technology, Get Connected will help you connect with the companies and products you are searching for. latencies and overheads associated with www.rtcmagazine.com/getconnected the serial protocols (see sidebar “The Good and Bad Sides of Serial InterconPC Satellite Receiver nects,” p.36). LAN The boards in VME systems, as in other types of systems, have become more and more intelligent over time, and with CPU DSP DSP DSP DSP this intelligence the level of inter-board communications has changed. As an example, consider an advanced radar system using several VME CPU boards to

End of Article

Products

VME bus

Get Connected with companies and Figure 1

products featured in this section. www.rtcmagazine.com/getconnected

Radar Data Acquisition using VME64.

Get Connected

with companies mentioned in this article. www.rtcmagazine.com/getconnected

September 2006

Get Connected with companies mentioned in this www.rtcmagazine.com/getconnected

SolutionsEngineering

collect and process imaging satellite data via multiple DSP-based acquisition boards. Figure 1 shows the way these boards used to be connected for this real-world application. The DSP boards acquire data from the satellite receivers, then process the data to filter noise and to create digitized information that is passed via VME bus to one or more CPUs, normally located in the same VME chassis. The user interface is provided by an external PC connected via a LAN. In this situation, the VME bus provides both the control and data planes for the acquisition boards. As a result, in higher-end applications, the number of VME slots used by the DSP boards, together with the total VME bus bandwidth requirements, limits the total processing capacity of the system. Now consider Figure 2, where the DSP boards have been replaced by FPGA PMC boards on VITA 41 baseboards, and the main CPU board has been replaced by a much more powerful unit using a very fast dual-core processor. The data processed by the FPGA boards is now passed via backplane Gigabit Ethernet connections to a CPU board in the same chassis. The VME bus remains in use as the control plane while the VITA 41 serial interconnect is used as the data plane. The system connects via the chassis switch boards to the LAN, which hosts the PCs providing the user interface. In this system, flexibility is greatly improved by utilizing the in-system LAN to provide both the internal and external connectivity, allowing CPUs or additional FPGA engines to be added either inside the chassis or even in an entirely separate chassis. Splitting the control and data planes for the FPGA boards also improves responsiveness to control functions by minimizing interference from data transfers. In adopting a serial interconnect, the traditional backplane â&#x20AC;&#x153;busâ&#x20AC;? is no longer present. Several interconnection topologies are possible, but the most common are the star and dual star configurations shown in Figure 3. The obvious complication with these topologies is the need to include one or two additional switch boards, which

Satellite Receiver

FPGA PMC

CPU

FPGA PMC

CPU

FPGA PMC

CPU

FPGA PMC CPU

VME bus VITA 41

VITA 41

VITA 41.3 Switch PC LAN

Figure 2

Radar Data Acquisition using Serial Interconnect.

adds to the costs and of course lowers the system MTBF. However, a dual star configuration also provides multiple interconnection paths between boards, potentially

improving overall system reliability. An alternative topology, also shown in Figure 3, is the mesh, which allows boards to directly connect to each other without using

Star

Dual Star

CPU

A B

VME bus

Switch

Mesh CPU

CPU

A B

A BB

VME bus

Figure 3

Backplane Network Topologies.

September 2006

SolutionsEngineering

Figure 4

The Concurrent Technologies VX 405/04x is a 6U Intel Core Duo-based SBC.

switch boards. This improves the cost but makes the implementation of multiple interconnection paths much more complex.

VME – The Serial Interconnect Story

For VME, the modern serial fabric evolution started with VITA 31, which adopts for P0 the connector style and pinout used by the newer CompactPCI standard for its Packet Switched Backplane (PSB). By using Ethernet technology for the fabric, speeds up to 2 Gbits/s full-duplex are possible, with dual fabric support

for redundancy or bandwidth improvements. One big virtue of this standard is the ease with which it can be adopted. Although new backplanes are needed, existing CompactPCI switch boards can be used and the remaining VME64x features can be retained, including the ability to deliver more power to each backplane slot. VITA 31 supports only star and dual star fabric topologies. Surprisingly, the demand for VITA 31 systems does not seem to have been strong. Although several board and backplane vendors offer compatible products,

the takeup has been relatively slow. Perhaps this is because VITA 31 is the first stage in this serial evolution, and has been the herald of change rather than being seen as the change itself. VITA 41 (VXS) takes the serial fabric evolution further, supporting multiple alternative fabric technologies such as Gigabit Ethernet, RapidIO, InfiniBand or PCI Express. Like VITA 31, VXS retains the existing 160-way DIN connectors of VME64x, so standard VME bus connectivity and power delivery are unchanged. By using a newer style of P0 connector, more I/O pins are also available too. Even better, VXS boards fitted with P0 connectors will still fit in existing VME backplanes that do not have obstructions or connectors in the P0 region. Likewise, VXS boards without their P0 connectors and with no obstructions in that region will fit into full VME64x backplanes. Hybrid backplanes combining VXS and VME64x are available, and the fabrics can use different topologies, including a mesh as well as the star and dual star concepts offered by VITA 31. The backplanes themselves allow for many different interconnect technologies, although only one can be used in any one backplane. Serial RapidIO, InfiniBand, Ethernet and PCI Express forms of the standard have either already been ratified or are on the point of ratification, so the options are increasing. Some vendors

The Good and Bad Sides of Serial Interconnects The downside of the change to serial interconnects is some loss of determinism as hardware and software protocol overheads come into play. Although arbitration overheads occur in a parallel bus system such as VME, the timings are much more accurately known. Techniques exist to minimize the effects of these overheads with serial protocols, but are less than perfect solutions. In many cases, the sheer speed of the interconnect may overcome some of the limitations. Until now, the use of serial interconnects also posed a problem because the devices at the endpoints were still using inherently parallel connections. This is changing rapidly as the serial interconnect is integrated into a multitude of new devices. The transition from PCI bus to PCI Express is a particularly good example of this progression. Accepting the need for serial interconnects is only the start. With Ethernet, PCI Express and RapidIO well established, and others like InfiniBand and StarFabric waiting in the wings, the choice of protocol is a significant one for board vendors and their cus-

September 2006

tomers. Although Ethernet is relatively slow (up to 2 Gbits/s full-duplex) and has relatively high overheads, and thus is not ideally suited to low-volume data transfers, it remains probably the most ubiquitous, cost-effective, flexible and best-understood protocol for general use. PCI Express offers much improved bandwidth compared to Ethernet by supporting multiple transfer lanes (theoretically up to 64 Gbits/s for a x16 interface), but needs more complex switching protocols. Serial RapidIO has similar capabilities to PCI Express, though its overheads are slightly reduced and its peak bandwidth is lower at around 20 Mbits/s for a x4 configuration. InfiniBand, despite being the driving force for some of the technology improvements, is still somewhat specialized and limited to a small number of suppliers. For this eason vendors such as Concurrent Technologies adopted Gigabit Ethernet for their first “fabric of choice.”

SolutionsEngineering have developed boards that implement the interconnect protocols in an FPGA, allowing them (at least in theory) to be reconfigured for the fabric in use. Although the use of VXS products is still in its infancy, there is a much stronger ground swell of interest than with VITA 31. The prospect of much greater interconnect bandwidth and the provision for many more I/O signals may be some of the reasons. Hot-swappable switch boards and the availability of system management options such as the Intelligent Platform Management Interface (IPMI) may also spark more interest from users concerned about 24/7 availability and lower maintenance costs. The Concurrent Technologies VX 405/04x is an example of a new high-performance board that has been designed to take advantage of the connector level compatibility between VXS backplanes and traditional VME64x backplanes, while also supporting the new features offered by VXS (Figure 4). The next stage in this development is no longer an evolution. VITA 46 (VPX) and VITA 48 Ruggedized Enhanced Design Implementation (REDI) change the rules significantly, while offering both large improvements in outright speed and greatly improved options for power delivery, system management and cooling. Arguably, this is no longer VMEâ&#x20AC;&#x201D;certainly it is not VME as we have known it. Although hybrid backplanes (VPX and VME64x) are an option, the VPX backplane looks and is completely different to what came before. Even though 6U and 3U form-factors are supported, there is no connector compatibility with VME64x, but the VME bus electrical and â&#x20AC;&#x153;softwareâ&#x20AC;? interface is retained, though not in a simple way. VPX is a new standard and is still being finalized, but several vendors, particularly those supplying to aerospace and defense customers, are extensively involved with the creation of the specifications, and have already announced products. Product costs are likely to be high, especially in the early stages, and of course there is the need to completely reconsider the backplane technology. Many VME customers have been using the traditional backplanes for some time, and are comfortable with them. To get them into the zone where they are equally comfortable with what

RS232 Serial LAN

IPMI Shelf Monitor CPU

CPU

VME bus

Chassis Temp

IPMB0 IPMB1 PSU

Figure 5

Fans

IPMI connectivity.

inevitably must be seen as something other than VME may take some time.

System Management Improvements

Along with CPU speed, integration levels have improved, and the number of I/O signals available on a single VME board has also increased dramatically. Of course, as a result of all this improvement, the power demands have also increased, and despite the chip manufacturers efforts to reduce power dissipation while maintaining CPU speeds, the demands of the applications continue to push the envelope for backplanes and power supplies. Regardless, customers expect and demand higher levels of maintainability, and the concepts of hot-swapping and system management are moving into the limelight. System Management features arrived on VME with the introduction of VITA 38, adding the option for an Intelligent Platform Management Interface (IPMI) on the 160-way P1 connector used for VME64 and VME64x, with dual Intelligent Platform Management Buses (IPMB) supported on VME64x. IPMI offers a way to connect with in-chassis sensors to detect and record significant operating events, as well as provide chassis inventory information. For example, power supply

voltages, temperature and system fans may be monitored, allowing improved system fault diagnosis or just enhancing routine maintenance. Some systems may support external connections to the IPMB interfaces, which allows for remote monitoring of an installed unit or subsystem via LAN, WAN or other connections. Figure 4 outlines some ways in which IPMI can be connected. VME is once again changing. Despite its age, it continues to move forward to accommodate the changes in technology that are the key to its survival. Although the original designers could not have foreseen many of the developments to come in the 25 years since its birth, VME has proved remarkably durable. The introduction of switched fabric interconnects, the inclusion of new system management capabilities, and finally the jump to VPX and REDI, offer major improvements to the performance and flexibility for the VME systems of the future. VME will continue to be the bus of choice for many industrial, aerospace and defense applications for many years to come. Concurrent Technologies Ann Arbor, MI. (734) 971-6309. [www.gocct.com]. September 2006

SolutionsEngineering VME 25th Anniversary

VME Market Bifurcation: The VXS and VPX Face-Off With the advent of two new specifications, the VME market might appear to be bifurcating. While true in a sense, the reality is that two complementary technologies are becoming available to offer developers even more choices in performance, flexibility and ruggedness.

Ad Index

by E ran Strod Mercury Computer Systems

Get Connected with technology and companies providing solutions now

Get Connected is a new resource for further exploration

ME has enjoyed robust growth into the foreseeable future. There Whether is a percepproducts, technologies and companies. your goal since it was adopted in 1981. Today tion that VME is in from retreat in commercial is to research the latest datasheet a company, speak directly an Application Engineer,but or jump technical the the merchant VME board with market markets, thistoisa company's not entirely thepage, case. goal of Get Connected is to put you in touch with the right resource. is north of $1B and growing at a respect- Electronic Trends Publications forecasts Whichever level of service you require for whatever type of technology, able clip. In aerospace and defense,Get nearly growth industrial Connected will helpinyouVME connectin withthe the companies andcomproducts 80% of the merchant boards procured puting you areare searching for. market at more than 7% CAGR www.rtcmagazine.com/getconnected VME. Furthermore, VME is expanding through 2010. in this market at nearly 8% CAGR for

Interestingly, despite much merger and acquisition activity, the top players in the VME merchant market have remained relatively stable. Table 1 shows the top VME board vendors. Mercury was named the top VME board vendor for 2005. VME has enjoyed continual renewal as newer and better technologies are added to it in an evolutionary fashion. Over the past 25 years, the data transfer rate over the tried-and-true 3rd Party Applications DIN connector has increased by an order of magnitude; the latest innovation being Get Connected with technology and companies providing solutions now VME320. VME64x added a fabric conNumeric Libraries and Protocol Stacks Get Connected is a new resource for further exploration into products, technologies and companies. Whether your goal is to research t nector that further increased inter-slot data datasheet from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connect rates. VMEâ&#x20AC;&#x2122;s success has partly depended in touch with the right resource. Whichever level of service you require for whatever type of technology, Data Movement Middleware Get Connected will help you connect with the companies and products are searching for. onyouhaving maintained backward compatwww.rtcmagazine.com/getconnected ibility at the pin level.

Foundation Software and Systems Services

VXS (VITA 41) vs. VPX (VITA 46) Fabric Interface

Figure 1

A/D FPGA D/A

Fabric Interface

Products

Fabric Interface

CPU

CPU Switch Mez

Mez

A representative system architecture example supporting a high-end Get Connected with companies and application could conceivably be realized on a variety of form-factors. products featured in this section. The differenceswww.rtcmagazine.com/getconnected appear when mechanical issues such as ruggedization are considered.

The most recent tension in the VME market relates to VXS (VITA 41) and VPX (VITA 46). Some customers are concerned the market is bifurcating. They are right! The market is bifurcating, but not in the way that is commonly thought. VXS and VPX actually have a great deal in common.

End of Article Get Connected

with companies mentioned in this article. www.rtcmagazine.com/getconnected

September 2006 Get Connected with companies mentioned in this www.rtcmagazine.com/getconnected

DPM • VME64, VITA 31.1 • Pentium M to 1.8 GHz • 855 GME chipset with embedded graphics • Low power consumption • -40°/+85°C operation

CPM1 • CompactPCI, PICMG 2.16 • Pentium M to 1.8 GHz • 855 GME chipset with embedded graphics • Two PMC sites • -40°/+85°C operation

KPM-S1 • Custom embedded • Pentium M 2 GHz • 7520 chipset with up to 8 GB DDR 400 • ATI M26 Mobility Radeon graphics on PCI Express

Conduction cooled versions of the VME and CompactPCI/PICMG 2.16 single board computers • Extended temperature to -40°/+85°C • Tested to MIL STD 810 and 901D • Tested to high shock/vibration • Conformal coating available • Customized interfaces supprted

Let Dynatem provide your next embedded processor solution

www.dynatem.com 800.543.3830 949.855.3235 sales@dynatem.com Dynatem, Inc., 23263 Madero, Suite C, Mission Viejo, CA 92691

SolutionsEngineering

Figure 2

Application platforms and their underlying form-factors span a range from benign to extreme and from small to quite large. Developers tend to choose the mix that best fits their requirements to optimize cost.

Figure 1 shows the typical architecture of a sensor computer used in a radar or signals intelligence application. The system has a radio frequency module to transmit and receive signals, an A/D board, a quad-processor card, and a single-board computer with dual-mezzanine I/O slots. The boards are interconnected with a high-speed fabric that is depicted using a fabric switch card. AITech Concurrent Computer Corporation, Integrated Solutions Div. Curtiss-Wright Controls Embedded Computing (CWCEC) Dynatem GE Fanuc Embedded Systems (including SBS) General Micro Systems, Inc. Kontron AG Mercury Computer Systems, Inc. Motorola Embedded Communications Computing (MECC) North Atlantic Industries Pentek, Inc Radstone Technology Spectrum Signal Processing Thales Computers VMETRO Table 1 Top VME vendors in

alphabetical order. 40

September 2006

In the software domain, there is system software, data movement middleware, numeric libraries such as VSIPL and of course, the user application. Note that this canonical system is form-factor-agnostic. It could be VXS. It could be VPX. It could be something else completely. All of these form-factors use the same basic technological ingredients: PowerPC, RapidIO, Linux, etc. In fact, Mercury has demonstrated families of products based on VXS, VPX-REDI, AdvancedTCA, CompactPCI and other form-factors—all based around a common processor/fabric/software architecture. With a single underlying software and hardware architecture, the ultimate form-factor deployed can reflect an application designer’s needs and preferences in mechanical robustness and ecosystem.

VXS – VPX Differences and Similarities

The mechanical differences in VXS and VPX are fairly straightforward if you read the standards. The differences are summarized in Table 2. VXS is exactly like VME64, maintaining the P0 and P2 connectors, but adds an improved P0 connector that supports modern multi-GHz serial fabrics like RapidIO. With a VXS backplane, system engineers can carry forward VME64 cards without the neces-

sity of a hybrid backplane. Note, the VXS P0 obsoletes the VME64x P0. This is a difficult development for VME64x customers who are now coping with the fact that the leading VME64x fabric silicon company appears to no longer be a growing concern. Unlike VXS, VPX is a ground up redesign of the backplane chassis. VPX can support VME, but does not mandate that VME be carried forward. It eliminates the VME P1 and P2 DIN connector in favor of the new multi-GHz connectors. As shown in Table 2, it has much greater I/O pin capacity. When teamed up with the REDI standard (VITA 48), the result is truly revolutionary, supporting higher slot power budgets, enhanced ruggedization and allowing for an array of open-standard cooling methodologies: liquid flow through and liquid spray in addition to conduction and air. It’s natural to think that VXS and VPX will compete with each other if you focus only on the mechanical differences between them. In reality, they complement each other. This can be understood only if we examine the ecosystems of the respective standards. Each standard addresses a different set of application requirements—one “demanding” and the other “extreme.” Over time, system designers will always gravitate toward the simplest solutions possible. For example, if a single Pentium motherboard running Linux will solve a problem at hand, why look further? If the application is fairly straightforward, designers will typically select a benign solution (Figure 2). Benign applications are generally stationary, run in temperaturecontrolled environments, require one or only a few processors and modest connectivity. These solutions play an important function in the overall ecosystem by setting a baseline for price and performance. Further, they improve relentlessly, forcing competing solutions to follow a similarly aggressive continuous improvement curve.

Rugged VME Systems

As embedded designers know, commodity technologies are great, but not if they can’t get the job done. Some applications require NEBS compliance (telecommunications) or must be capable of surviving shock, vibration or

SolutionsEngineering temperature extremes (defense). Some applications have to scale to a large number of I/O ports or require numerous inter-connected processors. When an application can’t get by with “benign” product, system engineers usually adopt more rugged technologies (Figure 2). VME’s bread-andbutter niche, where VXS is assuming the leadership mantle, has been in 6U rugged applications. Air-cooled CompactPCI and AdvancedTCA are more robust than the benign alternatives, and that’s fine for the telecom central office, but for mobile applications like defense and transportation, VME still offers a dominating value proposition. In rugged computing, customers tend to adopt technology at the board level. With hundreds of VME module vendors, customers have an enormous ecosystem from which to choose. The VITA Web site lists 24 categories of VME modules. These offer a wide array of processors, field programmable gate arrays (FPGAs), interfaces and functionalities. With an open-standard backplane, customers can choose best-inclass modules from numerous vendors and then integrate them together. This provides customers with enormous flexibility and choice. The vendor competition in the VME space means that new technologies are made available very quickly. For example, processor MHz increases are constantly being fielded on new boards as vendors jockey for performance leadership. The evolution of technology around an openstandard backplane is rapid.

Example VXS System

A very high-performance VXS system available today, Mercury’s PowerStream 6100 system, can house 68 PowerPC processors for 761 GFLOPS and supports more than 42 Gbytes/s of fabric bandwidth. That represents an order of magnitude performance leap over what we could do just two short years ago. Is it enough? For many applications it is, but there are some customers who can never have enough processor performance, fabric bandwidth, I/O connectivity or ruggedization. These customers adopt technologies in the extreme computing category (Figure 2). Extreme computing is a category in which the application requirements exceed what can be supported by simply integrating off-the-shelf boards.

Standard Characteristic

VME64x VME2eSST

VXS

VPX VPX -REDI

Specification

ANSI/VITA 1.1 ANSI/VITA 1.5

VITA 41

VITA 46 VITA 48

Standard Bandwidth

VME: 320 MB/s

VME: 320 MB/s 16 differential pairs: 5 GB/s @ 3.125 Gb/s

VME: 320 MB/s Up to 192 differential pairs: 10 GB/s @ 3.125 Gb/s 30 GB/s @ 10 Gb/s

Switch Fabric

Via the P2 connector

Any; central switch slot defined

Any topology

Faceplate user I/O

Yes

Backplane user I/O

205 single ended pins

110 single ended pins + 32 differential pairs + 31 pins RFU (future use)

48 single ended pins + 192 differential pairs

User I/O for 3U systems

80 pins on J2

Backplane I/O bandwidth

205 pins @ 1 Gb/s

110 pins @ 1 Gb/s 16 pairs @ 10 Gb/s

192 pairs @ 10 Gb/s

Defined differential X/PMC I/O mapping

Yes

Existing VME64x forward compatibility

Yes

Yes for VME cards without Yes, with hybrid backplane VME64x P0, use hybrid Also supports VME signal backplane otherwise mapping

Slot Pitch

0.8”

0.8” 0.85” or 1.0” w/REDI

Available Power

5 V: 90 W 3.3 V: 66 W

5 V: 120 W 12 V: 384 W or 48 V: 768 W

Cooling

Air, conduction

Air, conduction, liquid w/REDI enhancements

Table 2

Key computing standards for signal processing.

A large prime contractor came to Mercury with a challenge: “We need 125 PowerPCs in a ruggedized air-cooled chassis the size of a medium-sized refrigerator.” Later, a customer asked us to fit 69 Xilinx FPGAs in this same box. These challenges could not at the time, and still cannot today, be accomplished with VME. In order to solve the problem, system architects had to start with a clean sheet of paper. The backplane fabric was based on parallel RapidIO. The system was so densely packed with processors that Mercury invented a technique to cool it called “managed air.” Mercury developed its own backplane architecture and form-factor called MultiPort. The result is a system-level product called PowerStream 7000, which is being used on the Multi-Platform Radar Technology Insertion Program (MP-RTIP), among other major programs.

The PowerStream 7000 is an example of a system solution where processing, fabric, I/O and software are all designed holistically. You can’t take out one module and use it somewhere else—everything in the system is designed interdependently. Detractors of this approach will characterize it as proprietary, but that is not a realistic view. The system is completely based upon open standards: RapidIO, PowerPC, PCI Mezzanine Cards, POSIX, etc. Customers who require extreme levels of performance, whether that involves processing, I/O connectivity or ruggedization, simply have had to look beyond the open backplane PICMG and VITA systems. Today that situation is changing with the advent of VPX (VITA 46) and REDI (VITA 48).

VPX-REDI

The extreme computing application segment will soon be able to adopt VPX, September 2006

SolutionsEngineering an open-standard backplane architecture, which provides processing and connectivity for the industry’s most demanding applications. VPX supports 768W per slot. In first-generation systems, power is actually more limited by cooling capability than by inlet capacity. Thus, VPX is particularly revolutionary when paired with the Ruggedized Enhanced Design Implementation—REDI (VITA 48). With REDI, designers have open-standard mechanical specifications for four cooling methodologies: air, conduction, liquid spray and liquid flow through. REDI supports many other ruggedization enhancements that enable systems to address the harshest environments on earth or beyond. In order to address the diverse needs of extreme applications, the team working on VITA 46 has defined a range of options. Table 3 describes the major options available to designers of VPX-compliant solutions. A large number of options is good, of course, because it preserves flexibility. However in practice, there are so many options for VPX that it makes one wonder about the likelihood of third-party THA068_187x121_RTC US_GB 21/03/06

Form Factor

3U, 6U

Mechanical

IEEE 1101 or REDI

Slot Pitch

0.8”, 0.85” or 1.0” (REDI)

Cooling Methodologies

air, conduction, liquid flow through (REDI), liquid spray (REDI)

Fabrics

RapidIO, PCI Express, Ethernet

Optional VMEbus and Rear-Transition Module definitions Mezzanines

XMC, PMC

Table 3 VPX options.

vendor interoperability. There are a few vendors who have announced VPX product to date. It is too early to know, but it is possible the small community of VPX vendors will select different form-factors, fabrics or mechanical specifications such that modules cannot be mated together in a system. Still, large program adoption is driving VPX solutions toward Q4 2006 availability. As is natural in the early phase of technology 17:48 Page 1adoption, the pioneering

VPX-REDI vendors are making an investment to establish the market. In practice, this means most VPX-REDI vendors are building system-level product. Why? Because there is not yet an ecosystem from which to draw. For example, Mercury is integrating A/D, processing, FPGA and I/O functionality together in a conduction-cooled VPX-REDI architecture. This VPX-REDI system is called the PowerStream 6600. Since there is no VPX-REDI ecosystem to tap, Mercury is building the entire 6600 system with enabling software from the ground up. Based on experience with the extreme computing market, Mercury is confident the VPX-REDI ecosystem will emerge over time. Electronic Trends Publications forecasts that VPX products will start to ship in 2006 and grow to exceed $30M by 2010. It sees VXS growing to exceed $80M in that same time frame. Both standards will thrive, but VXS is widely expected to have a running start because it more easily leverages the VME-board ecosystem. Someday, VPX-REDI will certainly

Nothing empowers performance like PowerMP! The PowerMP concept is designed to provide offthe-shelf and off-the-chart performance for your critical computing needs in demanding environments. Each MP system is a high-performance, low-cost COTS-based multiprocessor computing solution based on industry standards and Pentium and/or PowerPC architecture. The new PowerMP6 – a multi-Pentium, readyto-use solution When you place a premium on software productivity and performance turn to the turnkey computer system that sizzles—PowerMP6. The newest in the PowerMP line, the PowerMP6 consists of multiple Pentium-M boards in a 19-inch rack. Running Red Hat Linux on the Intel processors supports software productivity and portability through an extensive set of open source and commercial tools and libraries. Performance is dictated by the number of Pentium M processors the system runs and the PowerMP6 available in various customized configurations of up to eight processor boards in a rack. The PowerMP6 features optimized message passing interface (MPI) for multiprocessor communications and contains software tools geared for such tasks as real-time performance analysis, remote control operations and monitoring system management.

www.thalescomputers.com 42

September 2006

The PowerMP4-60 – RapidIOTM system entry. The PowerMP4 fills the embedded industry’s need for reliability, increased bandwidth and faster bus speeds. It combines PowerPC and Pentium-M technology and takes advantage of the outstanding compute power to power dissipation ratio of the PowerPC technology as well as the wide spectrum of software tools available on PC platforms. PowerMP4-60’s RapidIO TM high-performance and packet-switched interconnect technology meet your demanding embedded system needs. The specialists in embedded performance Thales’ Computers include development of commercial and ruggedized VMEbus & CompactPCI systems solutions based on PowerPC and Pentium microprocessors. Thales’ products are optimized for a wide variety of applications in the military, aerospace, transportation, communications, and industrial markets and are used by blue chip customers worldwide.

For more information please contact: Luc Torres Tel: 33(0)4 98 16 33 95 e-mail: luc.torres@thalescomputers.fr

SolutionsEngineering have matured enough so customers will be able to adopt at the board-level and plug modules from different vendors into the same system. After all, that’s what an open-standard backplane is all about. The industry will get there, but a brand new technology like VPX-REDI requires a bit of time for the ecosystem to crystallize. What if an application requires more than VPX-REDI can offer? That’s the market that the MultiPort form-factor addresses. Now that VPX-REDI is coming to market, some customers will demand an order of magnitude better performance than even VPX-REDI can deliver. Future systems will deliver an enormous performance leap using the Cell Broadband Engine—a binary compatible G5 PowerPC processor with vector engines from IBM. There are today and always will be applications for which there is never enough performance and connectivity no matter how much COTS vendors deliver. The goal of the MultiPort family is to innovate a generation ahead of the PICMG and VITA standards, even as the standards organizations continually improve their offerings.

The real bifurcation in the VME market is between a la carte VXS board customers who want to perform their own module integration and VPX-REDI system customers who adopt technology at the holistic system level. This bifurcation will affect many different aspects of the VXS and VPX-REDI markets. In VXS, since adoption is at the modular level, pricing models will be unbundled. In the VPX-REDI market, software, hardware and services will be sold as a bundle. As a board market, the VXS ecosystem will enable customers to choose best-in-class modules from different vendors and integrate them together. Anyone who has spent time in the lab knows that integration of complex systems involving multiple vendors takes significant resources. Adopting VPX-REDI technology at the system level trades vendor flexibility for quicker time-to-market. VXS customers may be comforted in knowing that large vendors are now offering module-integration services to assist customers in defraying the significant effort of intervendor module integration.

There is a temptation to regard VXS and VPX-REDI as competing with each other, but in reality, they have different capabilities; and in the first generation of products, will draw upon their respective ecosystems differently. VXS is emerging as a modular-board standard that addresses rugged applications that cannot be satisfied with benign computing solutions. VPX-REDI will emerge as a system-level standard, with an ecosystem of a select few, highly committed vendors. VPX-REDI will go beyond rugged computing and address the needs of applications with extreme performance and ruggedization requirements. Both form-factor architectures will stand as distinct and viable alternatives, even as they share fabrics, processors, I/O interconnects and software. Mercury Computer Systems Chelmsford, MA. (978) 256-1300. [www.mc.com].

See

MERCURY COMPUTER SYSTEMS Herndon, VA

October 19, 2006

at the RTECC

Hilton Washington Dulles Mercury Computer Systems invites you to be their VIP guest at the Airport Hotel Real-Time & Embedded Computing Conference. Attend their open-door

seminar at 9:15am to learn how VPX-REDI Enables Multi-Function Radar in the Harshest Environments. Plus you’re invited to attend a special Luncheon Keynote on Cell Broadband Engine Architecture in Embedded Computing Applications ~ presented by Joey Sevin, Mercury’s Partner Development Director, Defense Business Unit. Mercury brings their experts to talk in depth about your needs, Thursday, October 19th! To view the full exhibitors’ listing and all technical seminars you can attend, visit www.rtecc.com/tysonscorner. Walk-in registrations welcome; bring your colleagues to take full advantage of this one-day conference. Event hours are 8:30 am – 3:00 pm. Seminars, Exhibits, Parking and Lunch are Complimentary

www.rtecc.com September 2006

IndustryInsight Ethernet Switch Boards

Blade Server Interconnect Converges on Switched Ethernet Improvements in switching technology and other advances will lower Ethernet’s latency and processing burden as it evolves to become the single interconnect for next-generation blade servers.

by M ike Zeile Fulcrum Microsystems

he rapid rise of the blade server system industry has introduced a new model for how to build out an enterprisecomputing data center. By using a chassis with computing blades added as needs grow, a blade server consumes much less space and is much easier to manage than multiple stand-alone servers. Switched Ethernet has become a staple on the backplane

of blade servers to provide networking services and access to users. Today, designers of next-generation blade servers are eyeing the technology as a converged fabric, delivering computing and storage interconnect as well as networking. The need for more storage and faster access to it, as well as faster networking, has become a priority in blade server de-

Blade Server Backplane Interconnect Protocols Ethernet (networking and occasionally computing)

Fibre Channel (storage)

InfiniBand (computing)

10 Mbits/s 100 Mbits/s 1 Gbit/s 10 Gbits/s

1 Gbit/s 2 Gbits/s 4 Gbits/s 8 Gbits/s (future)

8 Gbits/s 16 Gbits/s (future)

Ubiquitous, low cost

Exclusively used in storage networks

Focused on high-performance computing interconnect applications

Transport protocol (TCP/IP) is bulky, but well known

Figure 1

Three protocols currently dominate today’s blade system backplane: Ethernet for networking (and occasionally computing), Fibre Channel for storage and InfiniBand for computing. September 2006

sign. However, a chassis may contain up to three different backplanes: Fibre Channel for storage, InfiniBand for computing and Ethernet for networking and sometimes computing (Figure 1). This architecture means that storage resources are interconnected and computing resources are interconnected, but these subsystems are either not connected to each other or are bridged together, which robs performance. A “converged” interconnect is needed in next-generation blade servers to integrate computing, storage and networking. Switched Ethernet is the most likely candidate to play the role of converged interconnect. But Ethernet’s legacy as a networking protocol means that it lacks several key requirements that have allowed Fibre Channel and InfiniBand to thrive as specialized storage and computing interconnects.

Integrating Fast Storage Access

Integrating fast access to storage is becoming a critical element of next-gen-

IndustryInsight

eration blade servers. Storage blades are Application Application BufferBufferavailable that utilize compact hard drives app app User User designed for laptop computers and are I/O library I/O library well suited for networking tasks, such as domain name server (DNS), Web and OS OS firewall applications. Most other storage BufferTCP/IP OS needs are satisfied by farms of storage OS kernel OS kernel servers connected to blade servers via a Device driver Device driver Bufferstorage area network (SAN). Regardless driver of the way storage is interconnected, trafBufferfic must pass across the backplane to comBufferNIC NIC TCP/IP w/RDMA adapter adapter municate with the computing blades. When considering the storage conStd. IP Packet Std. IP Packet nectivity needs of a blade server, 10 w/RDMA extensions Gigabit Ethernet must meet the same performance levels as Fibre Channel. Fibre Figure 2 Without RDMA, packets are buffered at various levels of the OS, Channel’s advantages over Ethernet have taxing the CPU. With RDMA, the NIC can communicate directly with the historically been high bandwidth, with application buffer. 2 Gbit/s and 4 Gbit/s solutions widely available and 8 Gbit/s solutions nearing cally been lower than that of 10 Gigabit burden at full line rate is arguably as high market readiness. This situation changes Ethernet because it leverages a lighter as 10 GHz. Offloading this processing with the appearance of 10 Gigabit Ethweight transport protocol than the ubiquifrom the host is a critical need for interploration tous TCP that provides the same reliable connect applications. your goal ernet, which is now widely available at transport characteristics without the prok directly a price tag that is dropping below that of age, the Fibre Channel solutions. cessing overhead. However, a lightweight Attacking Ethernet Latency and source. But bandwidth is not the only criteria protocol standardized by Myricom runOverhead ology, ning on Ethernet has shown that 10 GigaSeeking a new opportunity for Ethd products for storage interconnect. Fibre Channel offers low latency as well, and this can bit Ethernet can be adapted to beat even ernet in the burgeoning blade server marboost throughput enough to overcome the InfiniBand in latency. ket, the industry has responded with new bandwidth differences between 8 Gbits/s While InfiniBand is starting to outtransport protocols and extensions to the and 10 Gbits/s. Technologies like iSCSI pace Ethernet again in bandwidth with the Ethernet standard and improvements in and Fibre Channel over 10 Gigabit Ether- emergence of DDR, 10 Gigabit Ethernet switching technology that make it comnet allow storage to be connected to com- holds the upper hand in terms of end-topetitive for blade server interconnect. puting using IP. But for blade server back- end connectivity. Although InfiniBand is The changes are coming from stannies providing solutions now planes, the real impact will come when efficient, it is not ubiquitous, so at some dards boards and leading Ethernet and ion into products, technologies and companies. Whether your goal is to research the latest storage can be direct-attached toGet a Connected point it will high-performance computing interconnect ation Engineer, or jump tocards a company's technical page, the goal of is to need put youto be converted into Eth10whatever Gigabittype Ethernet network. ernet for transport outside of the bladed players. For example, at the 2006 Internayou require for of technology, and products you are searching for. system. This adds latency and cost into tional Supercomputer Conference, MyriInfiniBand Utilized for the network design, as well as managecom announced that its low-latency mesComputing Interconnect ment complexity. sage-passing software will operate over On the computing interconnect front, Latency is critical for both storage 10 Gigabit Ethernet and that the company InfiniBand has staked out a place for high and computing interconnects, but another will sell network interface cards (NICs) performance and low latency. InfiniBand factor holding 10 Gigabit Ethernet back that will be able to pass data through stanaggregates 2 Gbit/s channels together to from this application area is the amount dard 10 Gigabit Ethernet switches. The offer bi-directional throughput of typi- of processing required by the host server new technology, called MX/Ethernet, uses cally 8 Gbits/s, moving to 16 Gbits/s with CPU to transmit data. The rule of thumb 10 Gigabit Ethernet as a Layer 2 network emerging double data rate (DDR) capa- for TCP/IP is that every bit of data transwith an MX EtherType1 to identify MX bilities. InfiniBand latency has histori- mitted on Ethernet requires 1 Hz of propackets, or frames. The same network and cessing power from the host CPU. Ever Myricom’s 10 Gigabit Ethernet NIC can faster processors have managed to keep carry TCP/IP traffic along with the MX Get Connected with companies mentioned in this article. pace with the network’s demands, but traffic, but MX/Ethernet uses MX’s effiwww.rtcmagazine.com/getconnected with 10 Gigabit Ethernet the processing cient reliability layer rather than TCP/IP.

End of Article

September 2006 Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

IndustryInsight

Other industry initiatives are targeting Ethernet endpoints with new standards to address about two-thirds of the latency in a network—the rest resides in the switch—as well as nearly 100% of CPU processing overhead. Among these initiatives is iWARP, a suite of enhancements and extensions of TCP/ IP, which includes remote direct memory access (RDMA), transport offload and operating system bypass. Together, these three elements can eliminate the vast majority of CPU overhead related to networking. The RDMA component of iWARP lowers latency by lessening the impact of the TCP/IP transmission protocol on the host computer. Today’s non-RDMA Ethernet adapter cards have a latency of about 30 microseconds. By comparison, InfiniBand adapter cards accomplish the same handoff in about 3 microseconds. With RDMA, computers can directly place information onto another computer’s memory with minimal demand on the memory bus and on the CPU, replacing TCP/IP copy requirements that add latency and consume CPU and memory resources (Figure 2). Transport offload is accomplished through a TCP offload engine (ToE), a specialized protocol processor on a NIC that offloads all TCP operations from the host CPU. With this dedicated processor, the operations are accelerated and CPU cycles are made available for running server or storage applications. Ethernet adapters with both ToE and RDMA are now on the market. In test labs, they have demonstrated latency performance equivalent to that of InfiniBand and Fibre Channel. The final element in the CPU overhead equation addressed by iWARP is bypassing the OS through the use of an external context switch. When an application wants to communicate to the Ethernet NIC, it must transmit its command to the OS and, in the process, save the application context so that the OS can return the result of the completed process to the right application. This requires a context switch that absorbs a lot of compute cycles. The OS bypass capabilities of iWARP transfer the processing of these context switches to

Switching can be the interconnect bottleneck...

The Ethernet switch is the bottleneck

Cumulative Latency

Ethernet today (>80us)

Latency (us)

Ethernet with RDMA Endpoints (~40us) 0

InfiniBand (<7us) Endpoint

Switch

Endpoint

... but low-latency switches even the playing field. 80

Cumulative Latency

Latency (us)

Low-Latency Ethernet (<8us) 0

Figure 3

Endpoint

Switch

Endpoint

InfiniBand (<7us)

Even with low-latency RDMA endpoints, Ethernet cannot meet latency levels that are competitive with InfiniBand. Only when the switch latency is reduced will end-to-end latency be competitive.

the Ethernet channel adapter, allowing the application to post its commands directly to the network adapter. The impact of this OS bypass is enhanced in a data center application because the very bursty nature of the traffic.

Improving Switch Latency

While these standards take care of the NICs, the interconnect within a blade center is built using switch chips. This is either done on a backplane that centralizes all of the switching, or distributed throughout the chassis on the blades themselves using a passive backplane for transport. Here, latency is the primary issue. The impact can be large and the time it takes data to pass back and forth among a number of cards can be a significant source of endto-end latency (Figure 3). The switch latency issue is being addressed by 10 Gigabit Ethernet switch chips that target latency in several ways. Like each successive generation of switches, they are growing in overall data capacity. At the same time, they incorporate the latest Ethernet congestion management standards from the IEEE that come closer to lossless switch fabrics on par with Fibre Channel and InfiniBand. Use of these techniques has

resulted in a dramatic drop in latency to around 200 nanoseconds in switch chips available today. With the new technologies encompassed in iWARP and through new switch congestion and efficiency capabilities, Ethernet has once again been adapted to a new market. If history is a guide, the benefits of increased adoption of Ethernet as a blade server interconnect will include lower cost and continued performance increases, as well as a reduction in the islands of connectivity, providing more access to computing resources. As Ethernet becomes a truly converged fabric, blade platforms are becoming dramatically more practical and scalable. Each slot in the chassis can arbitrarily support the addition of CPU resources, storage resources or networking resources, allowing system resources to scale arbitrarily to meet current and future needs. Fulcrum Microsystems Calabasas, CA. (818) 871-8101. [www.fulcrummicro.com].

September 2006

loration our goal k directly age, the source. ology, products

IndustryInsight Ethernet Switch Boards

The 1/10 Gigabit Ethernet Ripple Effect Ethernet continues to adapt to meet the ever increasing demands of high-performance networking. New standards nearing completion will make possible a unified 1/10 Gigabit Ethernet fabric solution.

by N auman Arshad Curtiss-Wright Controls Embedded Computing

thernet is hands down the most perToday, momentum is clearly buildvasive networking technology in the ing for 10 Gigabit Ethernet (GbE) in the world. Nearly all wireline traffic on commercial space for high-end servers, the Internet starts or ends with an Ether- network backbones and storage applicanet connection. As standards have evolved tions. The number of 10 GbE ports used from 10 Mbit/s, 100 Mbit/s and 1 Gbit/s to in commercial markets is predicted to nies providing solutions nowrates, Ethernet has continu10 Gbit/s data grow from approximately half a million in on into products, technologies and companies. your goal is to research the latest ally transformed markets Whether by creating new tion Engineer, or jump to a company's technical page, the goal of Get Connected is to put you Standard Ratification PHY Types ones, while defeating competitive network you require for whatever type of technology, or Draft Date technologies. and productsinterconnect you are searching for. In the process, a multi-billion dolStd. 802.3ae 2002 10GBASE-SR 10GBASE-LR lar ecosystem has arisen of companies 10GBASE-ER that have invested in Ethernetâ&#x20AC;&#x2122;s continued 10GBASE-LX4 ability to adapt to meet the ever increasing 10GBASE-SW demands of high-performance network10GBASE-LW ing. This continued commitment to Eth10GBASE-EW ernet has led to high-volume economies Std. 802.3ak 2004 10GBASE-CX4 of scale that drive ongoing investments in P802.3an June 2006 10GBASE-T R&D, technology innovations and stanP802.3aq Sept 2006* 10GBASE-LRM dardization activities.

End of Article

P802.3ap

Get Connected

with companies mentioned in this article. www.rtcmagazine.com/getconnected

10GBASE-KX4 10GBASE-KR

Medium

Reach (max)

MMF SMF SMF MMF/SMF MMF SMF SMF

300m 10km 40km 300m/10km 300m 10km 40km

Coax

15M

Cat 6 UTP or better

100M

MMF

220M

4-lane FR backplane 1-lane FR4 backplane

1M 1M

* Expected ratification date

Figure 1

Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

March 2007*

2006 to over 9 million in 2010, a substantial projected increase over the next four years. At these rates, 10 GbE is emerging as one of the fastest growing networking technologies. With a strong growth curve, 10 GbE is following in the successful footsteps

10 Gigabit Ethernet Standards and Draft Standards. September 2006

IndustryInsight

Figure 2

A VME64x High-Performance 1 Gigabit Ethernet Star Topology with 10 Gigabit Ethernet on board stacking.

of its 1 GbE predecessor. 10 GbE is a disruptive technology that promises to transcend multiple applications and markets over time. This momentum in the commercial space is creating a ripple effect that is finding its way into the emerging market of network-centric infrastructure for rugged applications such as defense. Implementation of network-centric warfare is driving battlefield communications to embrace bandwidth-hungry secure IP networks. With compulsory DoD requirements for network-ready resources that can seamlessly interface into the Global Information Grid (GIG), switched 1 GbE has emerged as the preferred interconnect for high-bandwidth IP platform networks. The natural progression is to move to multi-speed 1/10 GbE to enable a scalable interconnect solution that can keep up with the increasing needs of highbandwidth applications over several years. This progression from 1 GbE to 1/10 GbE has already started to appear at the component level. The market dynamics of hardware components, such as switch chips, PHYs and NICs, are shifting from handling auto-negotiation among 10/100/1000 Mbit/s Ethernet to auto-negotiation between 1 and 10 Gbit/s Ethernet. Lead50

September 2006

Figure 3

A VME64x High-Performance 1 Gigabit Ethernet Star Topology with 10 Gigabit Ethernet off-board stacking and boxto-box uplink.

ing switch chip vendors such as Broadcom, Fulcrum and Marvell provide multi-speed 1/10 GbE switch chips, so from a switching perspective, low-latency, line-rate 10 Gbit/s technology is available today. However, one of the major performance hurdles with both 1 and 10 GbE is the CPU performance bottleneck when a 1 or 10 GbE interface terminates at a processor card. As packets arrive through a high-capacity Ethernet interface, more processor compute cycles are diverted to processing the incoming packets that carry a TCP/IP protocol overhead than are focusing on application processing. This bottleneck became apparent with 1 GbE technology but it was not critical for communication over a 1 Gbit/s interface that operated at less than full capacity, since part of the processor’s cycles could be used for handling the network interface and part of it for the application. With 10 GbE this limitation is painfully obvious and a standardized working solution is a required prerequisite for usable IP-based 10 GbE termination. For example, handling a single 1 Gbit/s termination at full speed maxes out current 1 GHz 32-bit processors, let alone a full duplex 10 Gbit/s termination. How-

ever, with the evolution of TCP offload engine (TOE) technology, which offloads the CPU from TCP/IP protocol processing, and Remote Direct Memory Access (RDMA), which enables high-performance memory-to-memory data transfers with deterministic low latencies, both 1 and 10 GbE termination bottlenecks are being alleviated. Recently, vendors such as such as NetEffect have emerged from “stealth” mode to provide Internet Engineering Task Force (IETF) standardized Internet Wide Area RDMA Protocol (iWARP) NIC chip implementations that feature built-in TOE and RDMA functionality. iWARP is not just limited to WAN networks, but is finding its way into embedded designs as a way to enable high-performance blade-to-blade communications over 1 and 10 GbE.

10 GbE Standardization Efforts

Although various 10 GbE standards have been around for a number years (Figure 1), cost and performance hurdles have so far delayed it from becoming widely adopted. The primary factors that limited widespread adoption in the commercial world were due to initial standards that required costly 10 GbE optical components or due to the limited short reach of 15 meters using coax.

IndustryInsight

In June, the IEEE Standards Association standards board approved a new copper standard, IEEE P802.3an (10GBASET), which defines 10 GbE operation over unshielded or shielded twisted-pair, e.g., augmented CAT 6 or better. With the introduction of a lower-cost copper solution, 10GBASE-T shows promise to follow the success and economies of scale of 1000BAE-T. Additionally, improvement in distances obtainable with twisted-pair copper cabling now extends the reach of 10 GbE to 100 meters over low-cost copper, making it ideal for various enterprise backbones. 802.3aq (10GBaseLRM) provides an efficient Small Form-Factor Pluggable (SFP) optical solution for running 10 Gbits/s over multimode-fiber backbones at distances of up to 220m. Although 10GBaseLX4 was the first optical interface standard developed to run 10 Gbits/s over multimode fiber, 10GBaseLRM is expected to establish a leadership position due to its lower cost, smaller size and low power. Its smaller sized connectors are ideal for real-estate limited 6U and 3U VME/VPX boards. Additionally, the use of optical interconnection for box-to-box platform networks is preferred in rugged environments, due to its immunity to EMI and RFI. Although not used heavily by defense board manufacturers, the current dominant 10 GbE technology for board and backplane routing in the commercial embedded systems space is XAUI (802.3ae), which requires eight pairs of wires (16 pins) per channel. In embedded defense applications, XAUI has mostly been limited to an onboard interconnect while the backplane interconnect has been dominated by switched 1 GbE. However, an up-and-coming new draft standard under development, 802.3ap, aims to define new 1 and 10 GbE PHY transport mediums that can be used in high-speed backplanes. The 1 GbE PHY defines 1000BaseKX that provides 1.25 Gbits/s over a single lane with 8B10B encoding. The 10 GbE parallel version, 10GBase-KX4, splits the 10 Gbit/s signal into four lanes of 3.125 Gbits/s with 8B10B encoding per lane, while the 10 GbE serial version,

Figure 4

Conceptual 1/10 Gigabit Ethernet System Leveraging next-generation VITA46 and 802.3ap standards.

We Listen. Think. And Create. Digital I/O

Distributed I/O

Serial I/O

Expand with Sealevel PCI and PC/104 boards for robust communications backed by a Lifetime Warranty

Industrial Computing

HMI

PCI and PC/104 Serial Solutions Provide: • 1, 2, 4, 8, and 16-Port Models • RS-232, RS-422, and RS-485 Interfaces • Data Rates to 921.6K bps • 128-Byte FIFOs for Error Free Communication • Design Control and Long Term Availability • Extended Temperature Versions

F CUS On Success

September 2006

PCI-to -PMC

Technobox produces custom and off-the-shelf PMCs.

Serial

Ultra320

FPGA

async I/O, SATA, Ethernet, and IDE controllers. Plus a variety of adapters, extenders, and tools for PMC development or integration.

Adapters SCSI

solutions includes versatile FPGA-based digital I/O,

Quad Ethernet

Async

Our extensive range of

Make us your first stop for

SATA

PMC Solutions.

Reconfigurable

RoHS

CardBus

Digital I/O

Fast/Wide

SAS

PrPMC

PMC-to-PCI

IDE

www.technobox.com

CompactFlash

Your Source for PMCs, Adapters, and Tools. RS-232/422/485 LVD/SE

PMC-X

PIM

Test Access & Bus Analysis

Carriers Extenders

140 Mount Holly Bypass â&#x20AC;&#x201C; Unit 1 Lumberton, NJ 08048 Tel. 609-267-8988 Fax 609-261-1011

IndustryInsight

10GBase-KR, defines one lane at a full 10 Gbits/s with 64/66B encoding. Although 802.3ap is not expected to be finalized until March 2007, it shows great promise for physical standards that can be used in 1/10 GbE backplane routing for next-generation, high-performance VITA46 backplanes.

10 GbE and Legacy VME64x Systems

In embedded defense boards and subsystems, popular backplane standards such as VME64x are limited in their ability to support 10 GbE’s higher multi-GHz signaling. Due to this limitation—as well as to the processing overhead bottleneck—current use of 10 GbE for VME64x-based systems has been limited to onboard stacking of switch chips, front panel off-board stacking/cascading of switch blades, or as box-to-box highperformance platform backbones. An example of onboard 10 GbE stacking can be seen on Curtiss-Wright’s VME-680 SwitchBlade, which uses dual 12-port Broadcom BCM5690 multilayer switch chips that are stacked using a single 10 Gbit/s interface (Figure 2). Onboard stacking allows up to 24 x1 Gbit/s ports for larger applications while simplifying management and configuration by using a single IP instead of two separate IP addresses for each onboard multi-layer switch. Off-board 10 GbE stacking is also possible. In the company’s VME-682 FireBlade, the 10 GbE interfaces are routed to the front. With two VME682s in adjacent slots, the 10 GbE interfaces on each card can be stacked using a simple low-profile 10 GbE front panel overlay (Figure 3). This will enable a 48-port stacked solution in a VME64x chassis. This approach enables larger port counts in existing VME64x systems that can scale from benign to extremely harsh environments. The off-board 10 GbE ports can also be used to architect high-performance platform-wide backbones. With 10 GbE-capable switches in each box, the switches can be connected using their 10 GbE ports. Meanwhile, inside the box, 1 GbE interfaces can be

fanned out to connect each blade within a VME64x box, providing a balanced network topology. The system-wide 1/10 GbE backbone provides a unified infrastructure that can facilitate easier inter-blade or inter-box management, configuration, monitoring and data traffic transportation. The 10 GbE box uplink port can also be combined with security features on an advanced router card to control the IP traffic flowing into and out of the box through the use of firewalls, filtering and packet inspection.

10 GbE and Next-Generation VITA 46 (VPX)

The VITA 46 (VPX) next-generation backplane standard is specifically targeted at rugged environments. It offers seven rugged Tyco Electronics Multi-Gig RT2 connectors, P0 to P6, to form the backplane for a 6U card and three connectors, P0 to P2, for smaller form-factor 3U cards. P0 is a half-height utility connector used in both 6U and 3U versions, while the other connectors are all full height. VITA 46 provides unmatched pin-count density, with a total of 384 pins per slot. Designed for extremely harsh defense environments, VITA 46—along with the VITA 48 (VPX-REDI) draft standard that defines cooling specifications for rugged systems—offers a unique solution of increased backplane performance, support for high-speed signaling fabrics, increased pin counts, rugged connectors with built-in ESD protection and support for in-field two-level maintenance. The base specification, 46.0, defines a distributed switching architecture that enables five card mesh clusters, where each payload card can connect to every other card. To apply 10 GbE within the VITA 46 standard, the 46.7 substandard must be completed. The 802.3ap 10GBaseKX4 would be an ideal fit for 46.7, while 10GBaseKR would enable increased scalability in future implementations to provide 40 Gbit/s bandwidth using the same backplane routing. Furthermore, the completion of another key sub-standard, currently known as V46.20, would introduce centralized

switch slots that could support star, dualstar and hybrid (e.g. star + mesh) architectures within VITA 46. This would be valuable for systems that do not need meshed architectures or have dual topology requirements where meshes and star/ dual stars are required. Until the new physical interface 802.3ap 10 GbE specifications, and backplane VITA 46 along with all its dot specifications, achieve the status of official standards, their benefits remain conceptual. The benefits of a hypothetical VITA 46/802.3ap solution would yield a unified, multi-speed fabric that can support IP traffic, an ideal fit for NCW and GIG interoperability (Figure 4). Furthermore, the unified messaging over IP would simplify application design and portability without the need to learn new fabric protocols. Finally, the massive software ecosystem, tools and trained professionals already available for IP protocols could be leveraged to drastically accelerate the time-to-market and long life cycle support required for defense applications. With exciting new standards like 802.3ap and VITA 46 getting closer to completion, a unified 1/10 Gigabit Ethernet fabric solution for defense platforms is possible in the near future. Once again, Ethernet’s tide is poised to ripple across multiple markets including defense. The ubiquity of 1/10 GbE is a certainty, but the application possibilities are wide open. Curtiss-Wright Controls Embedded Computing Ottawa Ontario, Ottawa, Canada. (613) 599-9199. [www.curtisswright.com].

September 2006

VME PRODUCT SHOWCASE Featuring the latest in VME Product technology

T2-6U-VME -- Octal ADSPTS201 TigerSHARC 6U VME board

BittWare, Inc. Phone: (603) 226-0404 Fax: (603) 226-6667

Eight ADSP-TS201S TigerSHARC® DSPs providing 28.8 GFLOPs and 115 BOPS of processing Two Xilinx Virtex-II Pro FPGAs – interfacing, coprocessing ATLANTiS architecture – 8 GB/sec of external I/O throughput via Virtex-II Pro One PMC site w/PMC+ extension Tundra Tsi148™ PCI-X-VME bridge w/2eSST support 64-bit 66 MHz PCI interface via BittWare’s SharcFIN PCI-DSP bridge E-mail: info@bittware.com Web: www.bittware.com

Elma VME Chassis Family Chassis in VME, VME64x, VXI, VXS, and VPX Configurations in 19” rackmount, desktop, portable and Rugged COTS/ATR. Enclosures sizes in 1U-14U heights and various ATR styles. Wide selection of modular enclosures makes customization fast, easy, and cost-effective Largest offering of VXS platforms, with (two) 5, 8, (two) 12, 18, and 20 slot sizes. Leader in VITA 46 VPX platforms

Elma Electronic Phone: (510) 656-3400 Fax: (510) 656-3783

SystemTrace VME Monitor

The Leader in On-Time Delivery

Time-Correlated Monitoring of up to 32 VME chassis. Local or Remote Setup and Control via Ethernet. Monitor VME, 1553 & Ethernet in the same embedded system, ALL AT THE SAME TIME! SystemTrace makes it possible. ITCN is an ISO 9001:2000 Certified company.

ITCN, Inc. Phone: (800) 439-4039 Fax: (937) 439-9173

VME/VME64X/CPCI Vectorbord® Backplanes Vectorpak DIN or EIA RS-310 Subracks Custom Configurations Shorter Lead Time Factory Stock

Vector Electronics & Technology, Inc. E-mail: sales@itcninc.com Web: www.itcninc.com

Phone: (818) 985-8208 Fax: (818) 985-7708

Vortex Data Recorders

VMETRO, Inc.

E-mail: inquire@vectorelect.com Web: www.vectorelect.com

Phoenix VPF1

VME/CompactPCI Recording, playback, and analysis systems Up to 385MB/s sustained performance Architecture makes custom I/O and feature enhancements easy Web-browser GUI allows intuitive local or remote operation Fibre Channel SAN offers an array of storage options

Phone: (281) 584-0728 Fax: (281) 584-9034

E-mail: sales@elma.com Web: www.elma.com

VME/VXS Processor Board 2x PowerPC 7447 CPU nodes 2x Xilinx Virtex-II Pro FPGA nodes 8x 2.0-3.125Gbit/sec off-board serial communications channels Ethernet, RS232, RS422 64-bit, 66MHz PMC site VxWorks, Linux operating systems

VMETRO, Inc. E-mail: info@vmetro.com Web: www.vmetro.com

Phone: (281) 584-0728 Fax: (281) 584-9034

E-mail: info@vmetro.com Web: www.vmetro.com

Executive Interview

“Parallel Buses Are Dead!” RTC Interviews Ben Sharfi CEO, General Micro Systems

RTC: We continue to hear tales of stan- new applications will be dard bus architectures such as VME based on the VME/VXS. and cPCI becoming obsolete, yet GMS VXS gives users the added continues to develop new products performance they require based on these architectures. Can you over the backplane, yet it comment on whether or not you believe still supports the old hardthese architectures are facing extinc- ware. GMS definitely will tion and how that has impacted your continue its product development on popular those days), VME was pronounced corporate growth? VME/VXS: this istechnology a growthandarea and dead. When PICMG introduced the ATCA, Get Connected with companies providing solutions now not declining. VME was again pronounced dead. But the a new resource further exploration Sharfi: Let’s first talk about VME. There Get Connected Now asisto cPCI, thefor6U cPCI form- defense market refuses to let VME die. your degoal is some truth that VME is declining, butinto products, factor,technologies althoughand it’scompanies. used inWhether limited Defense and some automation companies is to research the latest datasheet from a company, speak directly not for defense applications. VME has fense applications, it is primarily used continue to design in VME mostly due to with an Application Engineer, or jump to a company's technical page, the indeed been phased out from most in Telecom and two reasons: A large installed base and goalindusof Get Connected is to put youautomation in touch with theapplications. right resource. Whichever you require for whateverwhere type of technology, trial automation applications. Large OEMlevel ofInservice the Telecom market, GMS has preservation of applications software. We will help exposure, you connect with companies and OEM products will continue to see VME 64 and VME users such as Applied Materials Get andConnected Ku- limited wetheare seeing you are searching for. licke & Sofa are no longer using it, and users moving toward AdvancedTCA. 64X in existing, upgraded systems and we www.rtcmagazine.com/getconnected I don’t believe many new automation ap- We do see a major cutback on use of will see the VMEbus in new designs using plications will be using the VME platform cPCI products and consequently there’s VXS and possibly VITA 46. As to cPCI, in the future. Automation companies that limited new development from GMS in since the big markets that have adapted are going through an upgrade are trying the 6U convection-cooled form-factor. cPCI (Telecom and Datacom) have moved to find ways to minimize the number of However, we are seeing a huge increase toward switched fabric-based standards VME cards used and/or move away from in demandwithfor rugged, conductionlike ATCAnow and its derivatives, then cPCI Get Connected technology and companies providing solutions VME completely. The primary reasonGet is Connected cooled and 3Uresource cPCI.forThis is ideal willtechnologies fade away these markets andgoal be isreis a new further size exploration into products, andincompanies. Whether your to research the la datasheetfor fromrugged a company, speak directly with anboth Application or jump to cost reduction. The industrial automation Mil applications, fromEngineer, a placed bya company's the new technical specs.”page, the goal of Get Connected i in touch with the rightand resource. Whichever level of service youhas require for whatever type of technology, market is seeing strong pricing pressure weight cooling standpoint. GMS Get Connected will help you connect with the companies and products you are searching for. from offshore companies in Taiwan, Ko- been aggressively involved in 3U cPCI RTC: Serial switched fabrics continue www.rtcmagazine.com/getconnected rea and Singapore; much more so than in products in the last six months, and has to invade the standard architecture bus defense applications. introduced four new 3U rugged, conduc- arena most commonly as hybrids incorIn defense applications—especially tion-cooled products (CC60x, CCMMA, porating the base specification with a for technology insertion upgrades— CC61x and CC50x). We will be adding high-speed connector added to provide VME still is No. 1. When it comes to four new products in this form-factor the conduit for the serial channel(s). selecting an architecture, VME has the this year. One of the more recent of these is cPCI highest success rate with our customMy chief technology officer, Udi Express, incorporating the base specifiers, such as Lockheed Martin and Ray- Levin, gave these thoughts, which sort of cation of cPCI and its Eurocard formtheon. We certainly don’t see any other sum up the relationship VME has with factor with PCI Express added as the technology replacing VME in the near cPCI: “Every time a new bus technology serial fabric. Is GMS currently workfuture for these customers. In fact with has been introduced, VME was proGet Connected with companies and Get Connected the addition of VXS, which is backward claimed to be “dead.” When cPCI was products featured in this section. with companies mentioned in this article. compatible with legacy VME, many adopted by Telecom (that became so www.rtcmagazine.com/getconnected www.rtcmagazine.com/getconnected

Ad Index

Products

End of Article

September 2006

Get Connected with companies mentioned in this artic

ExecutiveInterview ing on any designs incorporating cPCI Express? Do you think this will take over as the next generation of cPCI? Sharfi: As I have been saying for the last eight or nine years—during the “War of the Buses” era—the next-generation system bus, whether it’s Rapid I/O, PCI Express, InfiniBand or whatever, will once again be an extension of the local CPU bus. Ray Weiss (may he rest in peace) and I were in complete agreement on this issue (a rare event for us). Anyone thinking that the systems interconnect bus is dictated by anything else but the extension of the local CPU bus is misinformed, plain and simple, and has not learned anything from the last 27-30 years in this market segment. If you look at Exorbus, (which is where GMS started 30 years ago), it was an extension of a 6502/6802 bus. When VMEbus came along it was nothing but an extension of a 68000 bus. Multibus was nothing more than an extension of the Intel bus structure. The same is true with the cPCI bus. The cPCI bus is nothing but a local PCI bus extended out to the backplane with “Hot Swap” functions added to it. I believe this trend will continue with the development of new buses. So based on that, PCI Express is the CLEAR winner for now! If you open your PC, the old PCI connectors are no longer there. Now we have the small black connectors, which are PCI Express. The same is true with the

off-the-shelf shippable switches for PCI Express, but I believe that as the market demand increases, we will see such products from several vendors soon. The other statement that I have made publicly for the last eight years—for which we were criticized by many people including our friends at RTC (won’t mention any names)—is that “parallel buses are dead!” This includes VMEbus and cPCI bus. All new technologies are going to be Serial Switched buses. Our favorite bus structure is VITA 41.3. Why? Because every OS has support for 802.11 and many switches are available at a very low cost per node. VITA 41.3 supports fast data transfer rates yet also supports legacy VME protocols such as doorbell interrupt and priority interrupt structure, which makes it ideal for both speed and real-time applications. RTC: We receive a great number of press releases every month—frequently announcing new products. It’s been noted that GMS has been particularly active in new product development turning out new products with unusual frequency for a company of its size. What is your magic for turning out new designs at this rate? How do you target designs to go after? Sharfi: Thank you for the compliment. That is indeed true. GMS has the ability to bring to market products with complexity such as the V469, (quad Pentium M, dual-redundant CPU) in less than

Anyone thinking that the systems interconnect bus is dictated by anything else but the extension of the local CPU bus is misinformed, plain and simple. AGP bus used for high-end graphics applications; it’s now being replaced with 16-Lane PCI Express. As to GMS, we are developing several products using PCI Express hybrid buses. We just introduced our V469, a true dual-redundant server, and we have several other PCI Express products that we are introducing by year-end. We believe that PCI Express is the way to go for now. It is unfortunate that we still do not have 56

September 2006

four months. It would take other companies well over a year to develop such a product. We know this because our customer Lockheed Martin started out working with a competitor. After 12 months the designers were unable to deliver and Lockheed came to GMS. How we do it is really where our intellectual property (IP) lies. GMS uses a librarybased approach to all designs. Our libraries are extensive and very complete,

from schematics to symbols for PCB layouts, software drivers, test drivers, etc. We also have 3D modeling for each component and the MTBF for each component in our libraries. So for us to put a design together based on a variation of current existing product is simple and straightforward. Historically, we used to build a product and then try to market it. Opportunities proposed by customers don’t always align with that approach. Our rapid design response now enables customers to approach us. They tell us what they want built and we deliver within their time frame. So now, every product that GMS is building, we are building to order. This is the big shift in GMS’ business direction. Our customer base has recognized our talent and what we can offer, and know they cannot find it off the shelf. They come to us to develop such products as the V469, CC60X, CC61X, P630 (our hand-held unit) and the CC51X. All of these products are built per our customers’ specifications. Once we fulfill our contractual obligations, we are allowed to sell these products to others. I am truly proud of GMS engineering staff; the average tenure in my engineering department is 14 years. We don’t lose designers; consequently the level of talent just goes up and up. RTC: It seems, based on the press releases we have seen lately on rugged conduction-cooled products from GMS and the new “rugged military look” of your Web site and literature, that GMS seems to have shifted its business direction. Can you comment on that|? Sharfi: Good observation. Indeed GMS has shifted its focus. Ironically this was not by some great master plan. GMS, since its creation, has been very active in the automation business, and we have delivered products at far below the average selling price of others in this marketplace. GMS design and manufacturing processes have enabled us to be price-competitive and continue to win business in the automation marketplace, which is extremely cost-sensitive. Further, in the automation industry, the disciplines of packaging are significantly different. Functionality per square inch is the primary concern; the second criteria is performance/watt per

ExecutiveInterview dollar. GMS focus for years and years has been on performance, low power and low cost, enabling us to still find success in this market segment. Why the shift into the conduction-cooled arena? Our customers have approached us and have invested a significant amount of money in GMS to put our commercial products on conductioncooled platforms. The key element in their decision was our commitment to keep the cost structure within check while providing the highest quality standards possible. Let me give you an example: if you take a look at our 3U conduction-cooled cPCI board vs. the “usual guys” in the business, we provide four times the memory, 16 Gbytes of onboard bootable flash and full PICMG 2.9 at half the price of our competitor. It turns out the defense customers are getting very knowledgeable as to what the cost should be and many are developing their own boards because of unreasonable pricing. RTC: At a recent conference you had a small computer UMPC (Ultra Mobile PC) that you were talking about with some military customers. However, there is already talk in other applica-

departments, police departments and construction. Other unanticipated applications are in the hospital/medical industry (as Mr. Levin’s article in the July issue of RTC details). The key to the success of our rugged UMPC is that we provide the highest horsepower in the smallest footprint with the lowest wattage possible. This is more than any other company has been able to demonstrate, worldwide. RTC: VMEbus is celebrating its 25th anniversary this year and to some, it’s getting a little long in the tooth. However, applications remain strong in the military, which continues to keep VME alive. Do you believe VME continues to have a life outside of the military? If so, where? Do some of the extensions to VME such as VMS and VITA 41 promise to carry VME in some fashion into the future both in and out of the military? Sharfi: As I explained in earlier, VME will absolutely continue to thrive and grow with new extensions and hybrid buses such as VITA 41. However on the commercial side, the price is what is driving customers away, NOT THE

On the commercial side, the price is what is driving customers away, NOT THE TECHNOLOGY! tion areas for the need for very small form-factor PCs. Do you envision new applications for your UMPC in applications such as medicine, transportation, industrial control and others? Which ones? Why? Sharfi: The P630, which you reference, was developed for a customer-specific application for defense use. Since its introduction as a standard product, three other OEMs have approached us for a product variance based on the P630; two of them defense-related, one industrial. Our backlog and the revenue generated from the hand-held units are significantly more than all of the VME and cPCI boards combined (shipments in the thousands). We are seeing applications on the commercial side, such as taxicabs, fire 58

September 2006

TECHNOLOGY! Often people confuse this fact and say things like the technology is not good enough for automation and other commercial applications, thus the market is drying up. Pardon me for saying this but—it’s our own stupidity that drove many customers to PC-based hardware. When we are selling a processor board on VME in the $4,000 range and the customer can get it for $1,200 in other platforms, of course he is going to seek more cost-effective solutions. GMS approach to VME is that our pricing has always been and stays comparable to our CompactPCI line and very comparable to our form-free product line. RTC: The VSO has spent significant effort in assembling its VITA-58 specification for cans. Do you believe such

high-density packaging is the wave of the future both in and outside the military? The “can” concept appears to be still somewhere in the future, as there is yet to be prototype development or any deployed systems. How long do you think it will take for the concept to become reality? Sharfi: VITA 58 is not even fully defined yet. The current definition is much too broad to lead us anywhere. It is a nice wish to have an enclosure standard that is supposed to support any existing or future form-factors defined under VSO, but it is still much too broad. There are some nice ideas that are defined by VITA 58, like installing the unit with no need for any tools: this would be a great feature when one is in the field. I don’t see VITA 58 becoming a reality in less than two years and we should revisit this question once there are more defined specifications. RTC: Playing on the previous question, the Liquid Cooled Embedded Computer (LCEC) consortium also has a group working on a similar concept and with the concept of an LRU (Line Replaceable Unit) as a secondary benefit. Looking at both the PICMG effort and VITA 58, where do you believe the earliest development will occur—in the military with VITA 58 and the LRU requirements or in the communications sector looking for high-density, highperformance systems? Sharfi: I am not a believer in forcing a technology to work. My belief is to lower the power usage so that we can use more traditional, lower-cost packaging techniques. I don’t believe in liquid cooling. I am not saying that I don’t believe that there are applications for liquid cooling, but I don’t think the masses will move on to liquid cooling. Let me a give you a realcase application, not a hypothetical one. In traditional Phased Array radar, there are several hundreds of sensors that connect the ship’s mast to the computers below deck via massive amounts of cabling. Each DSP computer below deck consumes 60-100 watts and there are well over 30 processor boards in some systems. This creates huge cooling problems both for the computers and the room that houses

ExecutiveInterview these computers. One of our customers is taking 1” x 2” processor boards (derivatives of our P50X) and has installed them right behind the sensors. All the processor modules are connected to each other via a network mesh, requiring just a few cables to be brought down below deck. I believe the market is looking to suppliers for creative solutions in the approach to their designs; not to create a problem then solve it with afterthought solutions. Liquid cooling is creating a patch instead of resolving the root cause; that’s box thinking instead of “out of the box.” There will be a few applications that use it but I do not see it becoming used by the masses. RTC: There continues to be increasing activity surrounding the PICMG specification for AMC, and the associated MicroTCA. Now that the MicroTCA spec has just been approved, do you believe AMC and MicroTCA will be major additions to the stable of bus architectures? Why? Why not? Sharfi: I like AMC. I think it is a good solution and we have fixed a lot of the issues we had with PMC. I like the ability to use AMC as a stand-alone, not as a carrier and that it is hot swappable. I think it is great technology and will catch on. Very likely it will be the successor of the 3U cPCI products. We certainly will be developing several products based on this platform. MicroTCA as well as CompactTCA are results of the wake up of the market after creating ATCA with its very large form-factor. Many market segments found out that they could not use large form-factor boards. This forced the community to go back to the “old” 6U form-factor for the CompactTCA and the smaller form-factor for the MicroTCA. I believe that these form-factors will probably be more popular than, or at least as popular as, ATCA for our segment of the embedded market since that segment has many more applications that require the smaller form-factors.

the large numbers of fabrics. It seems that this is an on-going issue and it is hard to see it being resolved by a single definitive standard. Thus, I believe that the majority of the solutions will be proprietary solutions and custom designed as they were in the past with Telecom and Datacom. RTC: OEMs such as GMS develop products to solve customers’ problems. What do you find are the biggest obstacles that designers want suppliers to address with their product offerings and their company?

RTC: An article recently in the Wall Street Journal said that five companies—Cisco, Lucent, Motorola, Nortel and Qualcomm have been working to revise their systems to meet Verizon Wireless requirements. The article goes on to say “Carriers are rolling out new multimedia features and are expected to spend billions of dollars in coming years on new Internet-based technology, known as IMS, for Internet Protocol Multimedia Subsystem.” Do you believe the standard architecture board

Sharfi: This is by far my favorite question. The biggest obstacle—whether we talk about defense or commercial applications—is obsolescence. Customers are tired of getting letters from their suppliers, including GMS, notifying them that the product they have been buying is going End-of-Life, and they have to go buy hundreds or thousands of them and put them on the shelf until they are ready to use them. This is a huge problem in the defense market, costing the taxpayer billions of dollars for stockpiling hardware. GMS came to this realization about four years ago. Since we cannot beat obsolescence we need to design around it. GMS is head and shoulders above anyone else on this issue. We put our processors, not our I/O, on a mezzanine and our I/O on a baseboard. We know

I am not a believer in forcing a technology to work. My belief is to lower the power usage so that we can use more traditional, lower-cost packaging techniques. business will get to participate in this market place? Will platforms be proprietary? ATCA? MicroTCA? Sharfi: Since the first appearance of switched fabrics and the concept of ATCA, there has always been confusion due to

that the processor, the North Bridge, the South Bridge and the memory—as well as power supplies to power such silicon—will become obsolete in five-six years, best case. However, if you look at I/O functions, they are not likely to go obsolete anytime soon like Serial ports, September 2006

ExecutiveInterview Ethernet, SCSI, USB, etc. These are the items that we put on the baseboard. Our P60x processor mezzanine (Pentium M CPU based on 855GME chipset) is a perfect example of this concept. We built this product three years ago and installed it on several VME and cPCI processor boards, making claims that these VME/cPCI boards will be upgraded with new processors to avoid obsolescence. GMS is going to introduce our P70X module (based on Core Duo, 945GM chipset) soon, which gives our customers technology migration without

Sharfi: Many of my comments to the previous question apply here in terms of how to design for obsolescence. VME is a strong platform and has a lot of advantages, especially in real-time embedded applications. To date there has not been a better bus developed. VME was designed for real-time applications and cPCI was not; its lineage is the PC. The requirement for real-time defense applications will go on, but average military systems designs have more than a five to six year system life requirement. If you can get six years out of your VME product life

cPCI average product life cycle is three years or less whereas VME is at least five years. having to redesign the baseboard, cables, chassis, power supply and everything else associated with it. That is the perfect example of planning for obsolescence from the get-go. While on this topic I think we need to give Intel some credit. About five years ago, at the Bus and Board conference in Long Beach, Calif., Intel designers stood up at the podium and made statements about how committed they were to the embedded market. Many of us (especially me) were skeptical about such statements. Well, five years later Intel has demonstrated this BETTER than any other supplier. Intel has gained our trust and the trust of the industry, and is now the leader in our market segment. More NEW Pentium-based designs are being deployed than any other processor. The company’s long product life cycle, roadmaps and communication with suppliers like GMS have won this trust. GREAT Job Intel!! And Thanks! RTC: This is the 25th year of VME and it still is going strong. However, just about all of the products developed for it in the early years are gone. What do you find the current product life cycles to be for the board-level products from introduction to the end of design wins? How do technology upgrades or insertions to a particular product affect its life cycle? Does this differ by architecture? VME vs. cPCI? 60

September 2006

cycle without redesigning, you have done well nowadays. GMS shipped 68030 products over 17 years with the same form-factor and silicon—those days are long gone. When the COTS initiative came along, with all of its benefits, it also killed the life cycle of products. We married the good, the bad and the ugly of the PC market. As to where do you get a longer life cycle? By far you get it from VME. cPCI average product life cycle is three years or less whereas VME is at least five years. For most military real-time applications, the VMEbus offers sufficient performance and reasonable data bandwidth. The basic VME interface has not changed much during the last 25 years, and even with all the proposed changes, it is still basically backward compatible. This makes it easy to upgrade the onboard technology while leaving the VME interface the same. Thus, we can continue to refresh the old systems for as long as the VME interface is available. This can be for many years—actually, no time limit. RTC: The total market for mid- and high-end, modular embedded computers is estimated by some analysts to be in the area of $7 billion in 2006 growing to more than $12 billion in 2009. Do you agree with these estimates? Where do you think the bulk of the growth will happen? Communications? Military?

Industrial Control/Medical? Other? Will any architectures dominate in your opinion? Sharfi: You know how I feel about analysts. Whether we’re talking market analysts or stock market analysts, they always tell you the facts after the fact. Go look at what the market analysts said regarding the growth for the Telcom market and tell me how close they came to their predictions. With that said I definitely agree that there is strong growth in the modular embedded computer business for all of the reasons I mentioned above, such as distributed processing and lower power consumption. I truly believe the defense market has huge needs for small handheld rugged applications. We are talking to some customers, many of whom have requirements in the thousands of units. I do believe that this market segment alone is going to grow significantly. The second area in which I see growth will be industial control and medical applications, where small form-factors and low power will allow users to embed the processor module in the I/O control subsystem. Going to a doctor’s office and having the staff perform an MRI in the office rather than going to the hospital is just one small example of an excellent medical application. Embedded computers will find their way into displays, resulting in smart displays with the processor and I/ O embedded in them. In defense applications, we are going to see disposable computers for electronic surveillance and war fighters’ health monitoring. The possibilities are almost endless.

China

An Emerging Market for Embedded Companies on the Move

ACCELERATE YOUR ROG* Beijing October 17

Shanghai October 20

Shenzhen October 24

Taipei October 27

China and Taiwan’s sizzling industrial computer industry is creating double-digit growth opportunities in the 2006 embedded space alone This October, let the successful Real-Time & Embedded Computing Conferences (RTECC) take your sales team into the emerging marketplace Push your *Rate of Growth beyond the Expectation • Unique Exhibitor Showcase • Open-door Technical Seminars • Select Sponsorships

Now make the call to action with the right showcase for embedded companies on the move! www.rtecc.com To ensure your successful sales experience, contact: Lauren Hintze Event Sales Manager +1 949.226.2014 RTC Group - Western No. America, Asia laurenh@rtcgroup.com

Lauren Rathfelder Event Sales Manager +1 949.246.7055 RTC Group - Eastern No. America, Europe laurenr@rtcgroup.com

Catherine Li China Event Sales Manager (86)10 8288-8222 Ext 7009 ChinaECNet Int’l Media Sales Director catherineli@chinaecnet.com

Software&Development Tools Middleware for Servers

Off-the-Shelf Carrier-Grade Middleware: The Next Logical Step With the growth of ever more complex communications systems, middleware is becoming a vital tool for integration, management and customization through the use of modular software elements that enable high availability, speed development and support product differentiation.

er exploration ether your goal speak directly ical page, the ght resource. technology, es and products

by N iranjan Vaidya OpenClovis

he telecommunications industry gave birth to the catch phrase “Innovation at the speed of the Internet,” so it should come as no surprise that telecom equipment vendors themselves are facing increasing pressure to rapidly bring ompanies providing solutions now complex carrier-grade solutions to market. An evolving ecosystem oftechnologies industry standards and off-the-shelf to ploration into products, and companies. Whether your goal isproducts to research is thehelping latest pplication Engineer, or jump a company's technical page, the goal of Getincreasingly Connected isdevelop to put you drive this tocycle and equipment vendors can ervice you require for whatever type of technology, a large portion of their systems with off-the-shelf hardware, opanies and products you are searching for. erating systems, protocol stacks and carrier-grade middleware. Carrier-grade middleware provides the glue that binds the various hardware and software components in a system (Figure 1). The simplest way of looking at it is that it provides everything above the operating system that is not specific to the “application” of a system. The primary focus areas for middleware are core infrastructure services such as intra-cluster messaging, system and platform management, and, of course, high availability. In addition to providing specific features, the middleware also differentiates a product by defining the system’s overall character. Well-designed middleware can enable the rapid scoping of system requirements, model component and system behavior beGet Connected fore a product is built; simplify the development and re-usability with companies mentioned in this article. of applications across a product family; drastically reduce intewww.rtcmagazine.com/getconnected gration effort and support iterative enhancements across multiple product releases.

End of Article

Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

on.

September 2006

Customer Application/Service

Carrier-Grade Middleware

Carrier-Grade OS Hardware Platform (AdvancedTCA, BCT, other)

Figure 1

Carrier-grade middleware provides the infrastructure necessary for building a manageable and highly available system. It forces a clean separation between the application and the platform, promoting modularity and reuse.

The carrier-grade features of a network element have traditionally been viewed as major market differentiators but they are now considered the norm, and a product is only distinguished by its lack of such characteristics. The bar will only go up in the future. By selecting appropriate off-the-shelf carrier-grade hardware, operating systems and middleware, equipment manufacturers can focus limited resources on their own applicationspecific innovations, while significantly reducing both risk and time getting a product to market.

October 3–5, 2006 Santa Clara Convention Center Santa Clara, CA USA

BE F R A OR EG (U N I se D E S ST Pr S om A EP ER V o Co E T. de $ 15 RT 20 T C0 0 H 80 ! 6)

www.arm.com/developersconference

WORK SMARTER. WORK FASTER. Connect to ARM Technologies and Partner Solutions at the ARM Developers’ Conference

Join ARM, its technology and tools Partners from around the world page 63 ® at the only industry event for developers [ 63of]ARM Powered solutions and: Q

LEVERAGE the industry’s strongest network of systems developers and partners

DISCOVER the ARM connected technologies and tools behind leading industry products

LEARN AND EVALUATE practical solutions for designing with ARM IP at 120 conference sessions, hands-on labs, panel discussions and 2-day expo

EXPAND your network of companies supporting the ARM architecture

WALK AWAY with everything you need for designing advanced digital products

Platinum Sponsors

Gold Sponsors

Silver Sponsors

Media Sponsors

Brought to you by:

Software&DevelopmentTools

Availability

Manageability

CLI

SNMP

LOG

Transaction/Lock Service

Provisioning

Alarm/Notification Management

Chassis Management

Upgrade Management

Diagnostic Management

Fault Management

Name Service

Checkpoint Service

Group/cluster Membership

Availability Management

SAF AIS, DMTF CLP

Object Registry OS Adaption Layer

SAF HPI

HW Adaptation

Communication / Messaging Services Operating System (Kernel) Platform

Figure 2

The pre-integrated components provided by middleware enable the development of a full featured and flexible system, with less expense and faster time-to-market.

Middleware: Tying Together a Platform

The term middleware is often used loosely, so it makes sense to define it in the context of a carrier-grade system. The features provided by carrier-grade middleware can be roughly broken down into four mutually interdependent areas: system management, platform management, high availability and core infrastructure. Particular implementations will naturally provide more or less functionality in each area. In terms of system management, middleware should provide a complete solution to model the various logical and physical entities in a system using an object model or information base. It provides the run-time framework to manage and monitor these entities as well as standard interfaces such as simple network management protocol (SNMP) or command line interface (CLI) to integrate with external element management systems. A number of third-party products in the system management space have traditionally provided these capabilities. Middleware must provide comparable functionality as well as tighter integration with application and availability management capabilities. Middleware also provides the ability to manage a specific platform using industry standard interfaces such as the Service Availability Forum’s Hardware Platform Interface (SAF HPI). Often, the middleware is pre-integrated with standard platforms such as AdvancedTCA or IBM’s BladeCenter/T, enabling a plugand-play experience. A system designer need only write drivers or board support packages for application-specific hardware such as specialized cards or chip sets. Middleware may further simplify the integration of these components into the rest of the framework, providing an additional level of assurance that they will work together reliably. Actual higher availability (“five nines”) is best achieved when a system designer plans for it up front and designs a sys64

September 2006

tem taking various constraints and behaviors into account. Middleware can aid this task by providing a comprehensive modeling framework and a design discipline, which, in addition to modeling and validating the desired system characteristics, also gives a measure of predictability to system behavior. Eventually, a system failure is inevitable and a system that fails and recovers predictably is preferable to one that handles the various fault modes in an ad-hoc manner. The framework can also enable consistent application behavior from one platform footprint to another. Hand in hand with the modeling framework, middleware must also provide the run-time infrastructure to manage the components’ entire life and fault cycle, from start-up to shutdown, and from detection and diagnosis to service recovery and repair. This infrastructure may include an availability management service for managing the life cycle and work assignments of redundant components, a fault service for implementing userdefined repair policies, a checkpoint service for enabling stateful application failovers, a messaging service with high-availability semantics, an upgrade framework for non-service impacting software upgrades and group membership services for enabling the formation of independent redundancy groups. Obviously, this is just a sampling of the services that middleware should provide. It might also provide common application domain services such as an extensible virtual IP framework. A carrier-grade system typically consists of multiple hardware and software components that must work together to become, in essence, a system. As such, they require infrastructure services such as intra-cluster messaging protocols, name services and replicated in-memory databases, all of which go beyond the capabilities of the typical operating system. These services must be provided in a portable manner for easy use of heterogeneous operating systems and processors.

Software&DevelopmentTools All of the characteristics and features provided by middleware must work together to provide a cohesive product experience to both the system designer and the end user. Traditionally, equipment vendors have built many of these features in-house or sourced them independently and integrated them inhouse, often at great expense. Pre-integrated solutions that offer a large percentage of a productâ&#x20AC;&#x2122;s middleware needs are the next logical step in the evolutionary process. An example of some of the features and services that can be included in a middleware implementation are shown in Figure 2. There are two distinct types of pre-integration possible with middleware. One is internal pre-integration among the discrete components that make up the middleware, with, ideally, the discrete components built on a common framework of concepts and modules. The other is external pre-integration with operating system distributions (kernel and libraries) and hardware platforms (chassis, CPUs, switch fabrics, line cards, etc.). Despite standardization, external pre-integration gives rise to a large number of permutations, and most middleware vendors would likely offer a preferred list of tested combinations with others supported on an as-needed basis.

A Question of Character

In addition to enabling the manageability and availability of applications, carrier-grade middleware also imparts design characteristics to a system that truly distinguish it beyond the point features easily visible to the users of the system. It goes without saying that the middleware components must themselves be manageable and highly available, but middleware also impacts the final solution in other ways. A middlewareâ&#x20AC;&#x2122;s range of modeling capabilities can vary from none, to modeling via a programmatic interface, to a comprehensive IDE framework that does everything from importing entity attributes via management information bases (MIBs) to generating code stubs, and perhaps even validating system behavior before the system is built. Modeling frameworks are commonly used in visual programming paradigms at a component design level. Their adaptation to system design can truly surpass traditional design approaches and enable rapid and precise development of a system. Models can be shared, reused and iteratively enhanced, thus promoting commonality across a family of products. Despite such commonality, one size does not fit all. The carrier-grade market ranges from small footprint embedded boards to large multi-chassis systems whose resources (processors, memory, etc.) compare with high-end enterprise servers. It is necessary to customize middleware both in terms of the components that must run on a processor and in the system resources used by each component. This can be a minefield for middleware developers as they trade off conflicting size, speed, scalability and algorithm complexity requirements. A comprehensive modeling framework can include the ability to customize the middleware for various target profiles, additionally supporting portability. A modular middleware architecture supports customization. It also supports high availability by helping to isolate faults at a module level and enabling the system to take module-specific recovery actions. September 2006

Software&DevelopmentTools

To ease the task of integration, middleware may impose a programming model on application components to enable manageability and availability. Many established equipment providers have legacy applications in their systems and this model needs to integrate well with those applications. Middleware can ease the burden of application adaptation by auto-generating code as well as providing default templates. One very significant feature of middleware is that it enables portability. Traditionally, system designers have built platforms and applications together, with potential leakage of one into the other. By defining a clear separation between the application and the platform, middleware forces portability into the design. In addition, middleware can explicitly support portability by implementing appropriate adaptation layers that hide target OS, processor and link specifics where necessary. The modeling framework can also support portability by allowing the system designer to define target profiles with distinct characteristics and adapt the abstract application models to specific profiles when “building” code for the model. Since middleware hides the platform from the applications, system designers may view any potential problem with the system as a middleware issue. Middleware can differentiate itself by the traceability built into its components and by its ability to audit overall system behavior.

The Role of Standards

A number of industry specifications are helping to shape the middleware space. These include the Service Availability Forum’s HPI and AIS (Application Interface Specifications), the PICMG (PCI Industrial Computer Manufacturers Group) AdvancedTCA set of specifications and the SMASH (Systems Management Architecture for Server Hardware) CLP (Command Line Protocol) specification from the DMTF (Distributed Management Task Force). These standards help to assure equipment manufacturers that they can achieve a certain level of functionality and mitigate risk by choosing a standards-compliant implementation. At the same time, it is important to understand the limitations of the standardization process and the scope for innovation available to different middleware vendors. With a few exceptions, standards such as the SAF AIS typically specify the interfaces desired from compliant middleware; they do not (and cannot) specify the internal behavior. It is possible for two implementations to be compliant at the interface level but to provide a very different quality of service, giving a very different character to a system. The standardization process also focuses on defining a common denominator of services that are of interest to the participants. It cannot be all encompassing. Both the middleware designer and the system developer must balance the need for standards with the desire to provide a fully integrated solution. Non-technical factors such as the support provided by a particular middleware vendor can also factor into the process. Does the vendor provide an “off-the-shelf but throw over the wall” solution or are they amicable to a close partnership, with new features developed in a collaborative environment? 66

September 2006

Software&DevelopmentTools

A Range of Choices and Decisions

It is important that equipment designers considering the development of a new product or product family explore the middleware landscape and take advantage of standards and competitive market dynamics. At the same time, it is important that they do so with the proper expectations. A middleware solution can reduce development time and cost by a significant percentage, both up front and over the product life cycle, but it is not a magic wand for solving the designer’s specific problems. A full understanding of system and application needs is still required: knowledge of the middleware model is critical and application development, integration and testing with the middleware framework is still the burden of the equipment vendor. A good middleware solution can force many of these issues to be addressed up front in the development cycle rather than later when the cost to fix them is much higher. A middleware solution needs to cover a broad range of deployment possibilities and it will thus offer numerous features and configuration knobs for each component—typically far more than developed by an internal team. This can help to future-proof a product, and give the ability to enable new features and behaviors with just a few parameter changes. However, a typical application will only need a few of these capabilities and it is best to phase in new features as the development team gains confidence with the middleware’s capabilities and, undoubtedly, its eccentricities.

Regardless of whether equipment vendors view middleware as a differentiator, a value-add or a necessary overhead, they are betting the farm on it. A concern of “not invented here” is natural and due diligence is clearly required. In our experience, system designers fall into two broad categories: those who wish to view the middleware and platform as a black box, and those who view no middleware solution as “perfect” and thus, wish to enhance it with internally developed capabilities. In either case, a prototyping approach may help to gain understanding of the strengths and weaknesses of each solution. It may also help for system designers to explore a collaborative development model with the prospective middleware vendor to allay some of their fears and doubts. Collaborative development can also help with the integration of legacy applications and infrastructure components that the equipment vendor wishes to retain in the new solution. It is important to keep in mind that the middleware itself is a complex, evolving ecosystem and that both systems designer and middleware vendor can benefit tremendously from an open approach to development. OpenClovis Petaluma, CA. (707) 285-2852. [www.openclovis.com].

Vanguard analyzers, PCI express, AMC, XMC, PCI/PCI-X, PMC, cPCI, and VME bus Vortex data recorders for VME, cPCI. Phoenix VME DSP processors plus FPGA, I/O modules 10 GB ethernet XMC, AMC modules

American Predator Corporation

Industrial motherboards

cPCI,VME, and PMC modules

Convection to conduction VME modules

Fibre channel tester/ analyzer 1G, 2G, 4G

Software and custom development services VME ATR’s, liquid cooled, airborne chassis, and commercial enclosures

qpi_v2_4.indd 1

Southern California’s

LEADER

in VME, cPCI,

PCIe, PCI, & PMC

For over 20 years, QPI Sales has been serving the Southern California Region as a Manufacturer Representative Firm specializing in the sale of Computer Bus Boards in the Mil-Aerospace and Commercial Markets.

QPI SALES 1-888-4-VMEBUS(CA) or (714) 966-5360

www.qpisales.com

9/14/06 September 20062:09:43 PM67

Products&Technology PMC Collects TimeMachine Data

Since most processors on Compact PCI- or VME-based systems lack a real-time trace port, some developers have not been able to take advantage of the advanced debugging capabilities of Green Hills Software’s TimeMachine tool suite. TimeMachine lets developers visualize and replay software execution, allowing bugs and inefficiencies to be easily and quickly eliminated. A new card from Green Hills connects to an available PMC slot on the system to provide TimeMachine data. Using an adapter that connects to a free PMC slot, TraceEdge-PMC outputs TimeMachine data to a Green Hills SuperTrace probe, which collects up to 1 Gbyte of data and uploads it to the developer’s workstation. The Green Hills INTEGRITY RTOS comes bundled with the PCI device drivers required to interface with the TraceEdge-PMC adapter. TraceEdge-PMC is available for all PowerPC processors. The SuperTrace probe is also available for a wide range of processors with on-chip trace capability including ARM, PowerPC, MIPS and V850. The TraceEdge-PMC card alone costs $2,000 and the SuperTrace Probe plus PMC card costs $11,900. A MULTI Professional or TimeMachine license is required. Green Hills Software, Santa Barbara, CA. (805) 965-6044. [www.ghs.com].

Cell BE Processor-Based Board Works on PC Workstations

A board based on the Cell Broadband Engine processor product is the first product with Cell Technology for commercially available PC workstations and may greatly reduce the cost of application development for the Cell. The Cell Accelerator Board (CAB) from Mercury Computer Systems is designed to deliver 180 GFLOPS of performance in a PCI Express ATX form-factor suitable for compute-intensive applications such as rendering, ray tracing, video, image and signal processing. When deploying the CAB, users can leverage the Mercury MultiCore Plus Advantage software, which employs middleware that abstracts hardware capabilities and manages the distribution of data across multiple computing elements working in tandem. Available software development tools and libraries include the MultiCore Framework (MCF), optimized Scientific Algorithm Library (SAL) and the Mercury Trace Analysis Tool and Library (TATL). As with Mercury’s other Cell BE processor-based offerings, the CAB supports Linux via a Yellow Dog Linux BSP (board support package) from Terra Soft Solutions. Prototypes of the Mercury CAB are now available in limited volumes. Production availability is planned for the first quarter of calendar 2007. The Cell Workstation Development System includes a CAB, highend dual-processor Opteron workstation running Linux and Windows and the Mercury MultiCore Plus Advantage software. Pricing starts at $7,999 for a single board, with discounts available on higher volumes. Mercury Computer Systems, Chelmsford, MA. (978) 256-3599. [www.mc.com]. 68

September 2006

Data Acquisition Processor Board Simultaneously Samples 16+ Channels

High-channel count data acquisition systems must sample inputs simultaneously on multiple channels. A mid-range data acquisition processor board from Microstar Laboratories can work in multiples, so engineers may easily create systems that sample 16 or more channels at once, at 625 ksamples/s per channel. Two or more boards can be synchronized across a network. The DAP 5380a/526 is powered by a 233 MHz Pentium that runs DAPL, a multitasking RTOS optimized for data acquisition and related control functions. Each board has 16 analog inputs and can acquire 14bit data at up to 800 ksamples/s per channel. The board can sample eight channels simultaneously at 625 ksamples/s per channel. For faster sampling, in four-channel mode the board samples up to four channels simultaneously at 800 ksamples/s per channel. It contains 128 Mbytes of onboard memory for data buffers and uses DMA bus-mastering to transfer data to the PC at up to 3.2 million samples/s. DAPstudio can be used to configure and control the board. The board can also be configured and controlled from LabVIEW, DASYLab, MATLAB, C++, VB and any applications that use DLLs. The board costs $3,995 and DAPstudio costs $199. Microstar Laboratories, Bellevue, WA. (425) 453-2345. [www.mstarlabs.com].

PXIe Data Acquisition Modules Target Test and Control Apps

Two new data acquisition PXI Express modules are the first in the industry, according to the company releasing them, National Instruments. They are designed for high-channel-count data acquisition, highspeed control and manufacturing test applications. The NI PXIe-6259 and PXIe-6251 M Series modules deliver fast analog and digital I/O with a dedicated per-slot bandwidth of up to 250 Mbytes/ s. They feature up to 32 analog channels with 16-bit, 1.25 Msamples/s sampling speed, up to four analog output channels with 16-bit, 2.8 Msamples/s update rates and 10 MHz digital I/O on up to 32 lines. As with other NI M Series DAQ devices, the new PXI Express modules feature the NI-STC 2 system controller, the NI-PGIA 2 amplifier and NI-MCal calibration technology for increased performance, accuracy and I/O. The new modules work with all existing PXI modules and software. Existing code written for the NI LabVIEW graphical development platform, NI LabWindows/CVI ANSI C development environment and NI Measurement Studio for Microsoft Visual Studio can be used with them. The NI PXIe-6259 and NI PXIe-6251 M Series modules are priced from $1,149. National Instruments, Austin, TX. (512) 683-0100. [www.ni.com].

6U PICMG 2.16 SBC Delivers I/O Flexibility and Ruggedized Options

A new 6U CompactPCI Single Board Computer (SBC) from GE Fanuc Embedded systems uses the 1 GHz MPC7447A or 1.4 GHz MPC7448 Freescale processor and extensive I/O ports for improved performance and flexible, rugged functionality. With conduction cooling and ruggedized options, the C2K integrates the Marvell MV64460 System Controller (Discovery III) Bridge chip, which includes a highspeed DDR333 SDRAM controller with 167 MHz interface that can service memory up to 1 Gbyte. For increased I/O expansion, the C2K hosts two 64-bit IEEE1386.1 PMC sites, and the PLX PCI 6254 CompactPCI Backplane Bridge allows the C2K to operate as a system controller or peripheral processor card. The IPMI and hot swap capabilities of the C2K make field operation and maintenance easier by allowing the end user to constantly monitor system status and, when required, change boards without shutting down an entire system. The C2K gives engineers flexibility with multiple I/O, including three Gigabit Ethernet ports, four RS-232/RS-422 ports, four RS-422/RS-485 ports, two 1.5 Gbit/s SATA ports and three high-speed USB 2.0 ports. The convection model also provides one high-speed USB 2.0 port at the front panel and 16 programmable GPIO ports with independent interrupts. Optional configurations are available for customers who want to use the C2K in industrial applications that require extended temperature operations. The C2K is available with VxWorks, Linux or Windows XP support and will be RoHS-compliant. Pricing for the convection model of the C2K Single Board Computer in OEM quantities starts at $2,790. GE Fanuc Embedded Systems, Albuquerque, NM. (800) 875-0600. [www.sbs.com].

PCIe Dual PMC Carrier Card Delivers 8 Lanes and FPDP

A new full-length PCI Express carrier card is reputedly the first to support eight lanes of PCIe I/O and dual FPDP connectors, achieving higher data transfer rates than are obtainable with conventional parallel buses. The ICS-7003 from ICS, part of Radstone Embedded Computing, provides a platform for subsystem blades based on two PMCs. The card includes a PCI Express to dual PCI-X bridge, and each of its PMC sites supports PCI- X (64-bit, 133 MHz), allowing data rates of up to 1 Gbyte/s, or up to 2 Gbytes/s in each direction simultaneously. An intelligent switch matrix provides user-defined Pn4 connections between the two PMC modules, entirely avoiding the host processor and operating system. The switch also provides support for the board’s two FPDP II ports. Installation into 8-lane or 16-lane PCIe systems requires no additional drivers. Pricing begins at $2,964. Radstone Embedded Computing, Billerica, MA. (800) 368-2738. [www.radstone.com].

ETX miniBaseboard: First Extremely Compact ETX 3.0 Baseboard

An “extremely compact” ETX baseboard is designed to support the advanced features of new ETX 3.0-compliant solutions—all within a 130 mm x 155 mm footprint. The Kontron ETX miniBaseboard serves as a universal development platform for ETX 3.0 modules and is completely backward compatible with all ETX predecessors. All of the interfaces onboard are dedicated, therefore accelerating application testing on the target platform with freely scalable processor performance. The result is a significant reduction of both time and expense in bringing the application to life. OEM customers who do not require additional special features can use the Kontron ETX miniBaseboard as a standard product for their end applications. As with all of Kontron’s ETX products, the Kontron ETX miniBaseboard includes long-term availability, robust construction, function-tested design, and is highly future-proof—all thanks to ETX 3.0 compliance. At 130 mm x 155 mm, the RoHS-compliant ETX miniBaseboard is Get (95 Connected and slightly larger than the ETX modules x 125 mm) itwith pairstechnology with. Thus, the companies providing solutions now Kontron ETX miniBaseboard can be integrated as an extremely compact baseGet Connected is a new resource for further exploration board in almost any embedded application, simply and easily. The integrated into products, technologies and companies. Whether your goal communication interfaces onboard the Kontron miniBaseboard are:speak 1x directly is to research the latest ETX datasheet from a company, 32-bit PCMCIA slot,with 4x an USB, 1x 32-bit PCI slot, 4 serial interfaces, and 2x Application Engineer, or jump to a company's technical page, the Ethernet interfaces with and 10/100 Mbit. cardresource. goal10/100/1000 of Get Connected is to put youAinCompactFlash touch with the right Whichever service you require fortowhatever of technology, slot and two IDE interfaces alsolevel areofavailable in addition all of thetype traditionGet Connected willathelp youinconnect with the companies and products ally offered ETX features. Pricing starts $234 small quantities.

Ad Index

you are searching for.

Kontron America, Poway, CA (858) 677-0877. [www.us.kontron.com].

www.rtcmagazine.com/getconnected

Family of RoHS-Compliant CompactPCI Backplanes

A variety of CompactPCI backplanes are now available in Restriction of Hazardous Substances (RoHS)-compliant styles. The RoHS Get Connected with technology and companies providin initiative requires that new electronic Connected is a new resource for further exploration into produc equipment released to theGet European market after 07/01/2006 datasheet contain from no a company, speak directly with an Application Engineer, in touch the right resource. Whichever level of service you require fo more than 0.1% (1000 PPM) ofwith lead, Connected will help you connect with the companies and products mercury, hexavalentGet chromium, polybrominated www.rtcmagazine.com/getconnected biphenyls, polybrominated diphenyl ethers or more than 0.01% (100 PPM)of cadmium. For backplanes, lead-free solder and components must be used. Elma Bustronic has updated dozens of CompactPCI backplane designs to meet the compliance standard. The RoHS backplanes from Elma Bustronic comply with the PICMG basic specification 2.0 R3.0 and PICMG Hot Swap specification 2.1 R1.0. Also, Elma Bustronic offers versions conforming to PICMG 2.16 for Compact Packet Switching and the H.110 Computer Telephony specification 2.5 R1.0. The backplanes feature an 8-12 layer controlled impedance stripline design, superior power distribution, and have virtually zero crosstalk. The RoHS-compliant CompactPCI backplanes are with companies and available inGet 3U,Connected 6U, and 7U heights. Configurations are available in Low products featured in this section. Profile, EasyPlug and EasyCable formats. Both right and left-justified www.rtcmagazine.com/getconnected versions are available, as well as both 32-bit and 64-bit designs. Pricing starts under $100 depending on volume and configuration requirements.

Products

Elma Bustronic, Fremont, CA. (510) 490-7388. [www.ElmaBustronic.com]. Get Connected with companies and products featured in this section. www.rtcmagazine.com/getconnected

September 2006

Products&Technology

Low-Cost Development for PIC Microcontrollers

Microchip Technology Inc., a leading provider of microcontroller and analog semiconductors, today announced that the popular PICkit 2 development programmer now supports in-circuit debugging of selected PIC microcontroller products. Engineers, students and anyone with an interest can now easily begin development and evaluation with PIC microcontrollers for a very low initial investment. The new PICkit 2 Debug Express Kit from Microchip Technology features a 44-pin demo board populated with a PIC16F917 microcontroller, and connects to any personal computer via USB. Its in-circuit debugging features include halt, single step and setting a breakpoint. The main barriers to entering the world of microcontroller code development and embedded programming—the cost and complexity of development tools—are eliminated with a purchase price of $49.99 for the complete kit. Included are 12 tutorials that allow users to learn at their own pace (source code provided), along with Microchip’s free MPLAB IDE integrated development environment and a host of other software—enabling new users to easily enter the world of embedded control. Available now, the PICkit 2 Debug Express (DV164121) is $49.99, which comes complete with Microchip’s 44-pin PIC16F917 Flash microcontroller demo board, and includes the programmer, USB cable and CDs. Microchip Technology. Chandler, AZ. (480) 792-7200. [www.microchip.com].

System Host Board Supports Dual-Core Processing and Graphical Performance

A new PICMG System Host Board (SHB) is designed as a mainstream platform and uses the Intel 945G chipset, which supports the dual-core Intel Pentium D processor, all Intel Pentium 4 processors, as well as Intel Celeron D processors for the LGA775 socket. With up to 10.7 Gbytes/s of bandwidth and 4 Gbyte memory of dual DDRII 533/667 SDRAM, the PCE-5120 from Advantech offers fast system response and support for 64bit computing. An Intel ICH7R I/O Controller Hub provides eight USB 2.0 and four SATAII ports. Software support for SATA RAID 0, 1, 5 and 10 ensures reliable storage and system protection. The PCE-5120’s scalable CPU power and PCIe x16 link to the SHB Express backplane cater to high-performance PCIe graphics cards, and provide 3.5 times more bandwidth than a standard AGP 8X interface. The chipset features the Intel Graphics Media Accelerator 950 for faster graphics and 3D performance. In addition to the PCIe x16 port, the PCE-5120 features four PCIe x1 links to the backplane to enable flexible support for expansion boards. The PCE-5120 has an onboard x1 PCI Express link for single or dual Gigabit LAN using an Intel 82573V controller, which enhances the network data flow with bandwidths up to 500 Mbytes/s. When combined with a SHB backplane, like Advantech’s 12-slot SHB backplane, PCE-5B12-64A1, the result is a comprehensive solution supporting a wide array of high-speed connections and expansion choices. Single unit pricing starts at $614. Advantech, Irvine, CA. (949) 789-7178 [www.advantech.com]. 70

September 2006

Mini-ITX Motherboard Features Intel Core Duo

Dual-core processors can give up to twice the performance for a lot less than twice the heat. A new mini-ETX motherboard from ITOX Applied Computing is equipped with the Intel Core Duo processor and supports the Core Solo, as well as the Celeron M and Celeron M ULV CPUs. The RoHS-compliant G5C100-N-G Mini-ETX motherboard utilizes the 945GM Express chipset paired with an ICH7M I/O controller hub, as well as up to 4 Gbytes of 533 MHz or 667 MHz DDR2 dual-channel memory, dual onboard PCI-Express Gigabit Ethernet controllers and Intel GMA 950 graphics. A single mPGA 479 socket provides support for specified Intel processors with 667 MHz, 533 MHz and 400 MHz system bus speed. For connectivity, the board has two Gigabit LAN ports, four serial COM ports, eight USB 2.0 ports, two PS/2 ports, a VGA port (2048x1536), 1-LVDS DFP Interface (1600x1200), Onboard 5.1 channel Audio with S/PDIF out, two SATA interfaces, an Ultra ATA 100 interface, a CompactFlash Type II socket and an FDD interface. List price is $468. ITOX Applied Computing, East Brunswick, NJ. (888) 200-4869. [www.itox.com].

Family of PCI Express Frame Grabbers for HighEnd Applications

A family of five PCI Express frame grabbers can be used in conjunction with a new EUROCOM 400 COM Express single board computer with dual-core Intel Xeon processor LV 2.0 GHz CPU from American Eltec. COM Express on the EUROCOM 400 provides x8, x4, x2 and x1 PCIe lanes to transport data. The PC_Eye/RGB is a color frame grabber for PCI Express. It supports two RGB cameras in multiplex mode with reset/restart functions. The PC_Eye/ RGB supports the PCI Express (x1 - 1 lane) interface at more than twice the data rate of a 32-bit PCI interface. The PC_Eye/Quadro is designed for use with up to four simultaneously running synchronized cameras. In quadro mode, the images are stored as four separate monochrome images in an 8-bit-per-pixel format. The PC_Eye/Async is a frame grabber for four simultaneous unsynchronized monochrome input channels. It enables the simultaneous recording of four unsynchronized monochrome 40 MHz and 8-bit input signals via four independent AD converters. The PC_Eye/Sec frame grabber was developed specifically for security and monitoring applications. It interfaces to up to 16 standard composite video (CVBS) color cameras conforming to NTSC standards. Pricing varies depending upon model and quantity. The PC_EYE/MONO frame grabber is priced at $682 in single piece quantities. American ELTEC, Las Vegas, NV. (702) 878-4085. [www.americaneltec.com].

Tachometer/Overspeed Module Targets HighNoise Environments

A new tachometer/overspeed module that acquires low-frequency inputs from industrial speed sensors is designed to ensure reliable measurement in high-noise environments. The V365 from Highland Technology is an 8-channel, single-width, 6U VME module that measures frequency and period over a wide dynamic range. Each of eight tachometer input channels includes a differential input amplifier and programmable signal conditioning. The V365’s four programmable overspeed/underspeed blocks have independent single-pull double-throw relay outputs, which may be aimed at any selected tach channel. Each is programmable to trip on static or latched overspeed or underspeed conditions. Relay polarity and startup override control is provided, as well as self-test of the overspeed facilities. The V365 can be connected directly to common transducer types such as variable reluctance or Hall-effect magnetic speed pickups, AC line voltage or alternator windings up to 150V root-mean-squared (RMS), optical pickups, reed-switch or signal-conditioned fuel flow meters and contact closures. The analog and digital signal levels of any selected channel can be monitored at any time via two sub-miniature B (SMB) signal monitor connectors. A third SMB connector provides internally or externally generated self-test signals for in-crate testing. Price is $5,400. Highland Technology, San Francisco, CA. (415) 551-1700. [www.highlandtechnology.com].

JTAG/Boundary Scan Software Uses Altera’s USB Blaster

A software option in the System Cascon JTAG/Boundary Scan software from Göpel electronic enables the download cable of Altera’s USB Blaster programming hardware to be used as a native JTAG/ Boundary Scan Controller throughout the entire product life cycle. Using the existing USB-Blaster set up, configuration data can be applied immediately through the SYSTEM CASCON software suite in conjunction with other test and programming procedures, providing an effective solution without the need to purchase new hardware. Integrating the SYSTEM CASCON software suite with Altera’s USBBlaster download cable allows developers to use the USB-Blaster to execute Boundary Scan test procedures. The USB-Blaster can also be used to program flash devices in-system and configure Altera FPGAs via the Jam Standard Test and Programming Language (STAPL), JEDEC standard JESD-71, Simple Vector Format (SVF) specification, or IEEE-Std.1532 specification. Additional SYSTEM CASCON software support for Boundary Scan chain designs with Multi-Drop devices is automatic, including all scan router devices by Firecron, National Semiconductor and Texas Instruments. All test and in system programming (ISP) procedures generated with Altera’s USB-Blaster download cable are fully cross-compatible with other JTAG/Boundary Scan Controllers by Gopel electronic, including the recently introduced hardware platform SCANFLEX. Support for the USB-Blaster download cable is included for no additional charge in SYSTEM CASCON beginning with release 4.3. Customers can purchase the USB-Blaster download cable through Altera distributors. Göpel electronic, Jena, Germany. +49-3641-6896-39. [www.goepel.com].

Host Interface Board Extends PCIe Bus at x8 Speeds

In high-performance applications based on PCI Express, more downstream PCIe boards can mean slower speeds. A new host interface board (HIB) from One Stop Systems installs in a PCIe x8 or x16 slot and operates at 20 Gbits/s. It is designed with re-driver technology, boosting the high-speed signal over a PCIe cable to a downstream board. The downstream system acts as if it resides in the host system. The HIB2 extends the host bus over a high-speed PCI SIG-specified data cable. This connects to the upstream port of the company’s chassisor rackmounted 1U PCIe switch, further increasing the extendibility of the single host to over 100 add-in boards. The HIB2 also supports a downstream Connected with This technology and link in non-transparent mode toGet a second host system. configuracompanieswith providing now links tion allows host-to-host communication up to solutions two 20 Gbit/s Get Connected is a new resource for further exploration for a total bandwidth of 40 Gbits/s. into products, technologies and companies. Whether your goal The HIB2 is availableis in host mode and target mode, a target to research the latest datasheet from using a company, speak directly backplane and external PCIe card. It lists for $433. with an Application Engineer, or jump to a company's technical page, the

Ad Index

goalEscondido, of Get Connected is to put you in touch with the right resource. One Stop Systems, CA. (760) 745-9883. Whichever level of service you require for whatever type of technology, [www.onestopsystems.com].

Get Connected will help you connect with the companies and products you are searching for.

www.rtcmagazine.com/getconnected

Analyzer Supports 5 Gbit/s Per Lane PCIe

A new PCI Express protocol analyzer captures and analyzes second-generation PCI Express bus traffic at data rates of up to 5 Gbits/s per lane. The PE Tracer Gen2 Summit x16 Get Connected with technology and companies providin from LeCroy locates Get Connected is a new resource for further exploration into produc errors faster and redatasheet from a company, speak directly with an Application Engineer, cords and displays all in touch with the right resource. Whichever level of service you require fo traffic, even at high Get Connected will help you connect with the companies and products www.rtcmagazine.com/getconnected lane widths and data rates. The PE Tracer Gen2 Summit x16 works with the company’s PE Tracer application, using its triggering, filtering and error-reporting tools. The analyzer’s real-time monitoring tools analyze response and latency of transactions, data throughput and link utilization. Features provided in PE Tracer software include real-time statistics, protocol traffic summaries, detailed error reports, scripting and creation of user-defined test reports. Ethernet and USB ports are provided for connection and configuration flexibility. The PETracer Gen2 Summit x16 supports spread-spectrum clocked traffic, lane swizzling for flexibility in board configurations, auto-link with companies and sensing forGet linksConnected of varying width and support for multi-link operations products featured in this section. where PCI www.rtcmagazine.com/getconnected Express ports are bifurcated into narrower links. Raw Mode Recording records bytes as they come across the link, allowing debugging of PHY layer problems and combining the features of a logic analyzer format with a decoded protocol analyzer display.

Products

LeCroy, Chestnut Ridge, NY. (845) 425-2000. [www.lecroy.com]. Get Connected with companies and products featured in this section. www.rtcmagazine.com/getconnected

September 2006

Publisher’s Letter September 2006

On Independence…

n July fourth this year, when our country celebrated its independence, I was fortunate enough to be representing our magazines at the Kennedy Space Center as the nation put on, what was for me, the biggest Fourth-of-July fireworks display in the world—the launch of Space Shuttle Discovery (STS-121). While the technology behind this momentous event was staggering—especially after the tragedy of the Columbia—it brought to mind how important our independence really is both as individuals and as a publication. While the U.S. Constitution guarantees free speech and freedom of the press, these are often taken to extremes and have oft times become an excuse for sensationalism, editorial irresponsibility, grandstanding and overall deplorable behavior, their true mpanies providing solutions now meaning—the thought rather than just the words—comes to bear loration into products, technologies and companies. Whether your goal is to research the latest most often when those freedoms are threatened. plication Engineer, or jump to a company's technical page, the goal of Get Connected is to put you The quote, disapprove of what you say, but I will defend rvice you require for whatever type of “I technology, to you theare death yourfor.right to say it,” often attributed to Voltaire, but nies and products searching actually written by Evelyn Beatrice Hall in her 1906 book, The Friends of Voltaire (written under the pseudonym Stephen G. Tallentyre) probably best defines current thinking. And while irresponsible journalists have taken this idea as a right to do as they please, and have often done damage, the final judgment is by the readers: they can choose to read or not to read the publication. Recent national events have brought some of these concepts to the forefront—such as the publishing of certain sensitive information that could be used against this country by terrorists— however, it is not my objective here to pass judgment or even to broach that subject. However, even the trade press has threats to its own indeGet Connected pendence. organizations and consortia are continuwithCompanies, companies mentioned in this article. www.rtcmagazine.com/getconnected ally looking to get more editorial exposure than perhaps their importance in the industry demands. It’s the job—the responsi-

r exploration her your goal speak directly cal page, the ht resource. echnology, s and products

End of Article

Get Connected with companies mentioned in this article. www.rtcmagazine.com/getconnected

September 2006

bility—of independent publications to bring readers correct and unbiased information on all significant product and technology developments—not just those of special interests. When that independence is threatened, everyone loses. As Editorial Director of The RTC Group’s publications, I strive to see that our editors enjoy that independence—even from me. Frequently, there are many issues that threaten that independence—some advertisers want, and may even threaten to get, preferential editorial treatment. Trade groups with myopic industry views threaten and even make written demands on publications for special consideration. And organizations, hungry for publicity, rant and rave and make veiled threats if their special interests aren’t given preferential treatments.

Couldn’t Be More Wrong

Many times these come from young, inexperienced marketing/communications people. Eventually they learn. However, some come from directors of mature organizations that somehow believe if they can muscle the trade press into giving them some level of exclusive treatment it will further their cause. They couldn’t be further from the truth or more wrong. For those publications that knuckle under, it becomes obvious to readers that the publication no longer has its independence. And while it takes time, readership drops off and eventually advertisers fail to see the benefit of advertising ,and the publication will succumb. If RTC readers wanted a one-company or one-organization view of the industry, they would go directly to that company’s or that organization’s Web site. But what RTC offers—and what our readers want—demonstrated by the unparalleled success of the publication—is a complete view of leading technology and companies that are leading in technology and product develop-

Publisher’s Letter

While the U.S. Constitution guarantees free speech and freedom of the press, these are often taken to extremes and have oft times become an excuse for sensationalism, editorial irresponsibility, grandstanding and overall deplorable behavior, their true meaning—the thought rather than just the words—comes to bear most often when those freedoms are threatened. ment. RTC’s editors combined have over 100 years of experience in looking at and identifying significant trends in the industry. We’re not always 100% right, which is why we look at—and publish information on—all technologies and product families. For example, in not-too-distant history, journalists in our industry were pressured by various groups to support Futurebus+. I, for one, wrote a lot on the technology and products, but also covered in detail what was competitive to it and why there were obstacles to its acceptance in the market. At the time, I was severely criticized by some factions for my views. Let’s look at more recent events. RTC published Interconnect Strategies as a supplement to both it and its sister publication COTS Journal. We gave fair technical and product information to all approaches—Ethernet, RIO, ASI, PCI-E and others. It was clear, and we pointed out, that the market could not support all of these approaches. However, all were given equal hearing in our publication. The market has made and continues to make its deci-

sion—perhaps, in part, because of our independent coverage. Single-company and single-organization publications have popped up on the landscape from time to time, only to publish a few issues and fade away. The trade press cannot function when fettered with rules and restrictions. A trade publication that caters unduly to one group or company will be abandoned by its readers—and rightly so. Readers of the trade press look for leading products and technology, not a puppet of one company or organization regardless of how it’s disguised—as an independent publication, a “media partner” or whatever. It is for this reason The RTC Group’s publications are reviewing all industry relationships and their possible undue influence on our content.

Warren Andrews Associate Publisher

The RTC Group – No longer Welcome at Bus and Board The RTC Group has been a regular participant in the annual Bus and Board conferences since their inception. This year, however, our editorial staff was disinvited by Mr. Ray Alderman, the Executive Director of the VME International Trade Association (VITA). After exhaustive discussion of the conditions he said were necessar y for our attendance, Mr. Alderman relented and said he would consult the VITA Board of Directors concerning our participation. Mr. Alderman then returned with a set of conditions for The RTC Group’s participation, which John Reardon, CEO of The RTC Group was forced to decline because he refused to impose the kinds of editorial restrictions on his staff that were contained in the agreement. Mr. Reardon, surprised by this unprecedented move, was further confounded when VITA announced a “media partnership” with a competitor shortly thereafter.

September 2006

is to research the latest datasheet from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connected is to put you in touch with the right resource. Whichever level of service you require for whatever type of technology, Get Connected will help you connect with the companies and products you are searching for.

www.rtcmagazine.com/getconnected

Advertiser Index Get Connected with technology and companies providing solutions now Get Connected is a new resource for further exploration into products, technologies and companies. Whether your goal is to research the latest datasheet from a company, speak directly with an Application Engineer, or jump to a company's technical page, the goal of Get Connected is to put you in touch with the right resource. Whichever level of service you require for whatever type of technology, Get Connected will help you connect with the companies and products you are searching for.

www.rtcmagazine.com/getconnected

Company

Page

Website

ACT/Technico........................................................................................................ 23........................................................................................... www.acttechnico.com ADLINK Technology, Inc......................................................................................... 21....................................................................................www.adlinktechnology.com Advanet Technologies............................................................................................ 17.......................................................................................... www.advanettech.com

Products

End of Article

Application Acceleration Seminar........................................................................... 33............................................................................................. www.aaseminar.com ARM Developers Conference.................................................................................. 63....................................................................................................... www.arm.com BitMicro Networks, Inc............................................................................................ 6..................................................................................................www.bitmicro.com BittWare, . ............................................................................................................ 27.................................................................................................www.bittware.com

Get Connected with companies and

Get Connected

productsInc.. featured in this section. Dynatem, ....................................................................................................... 39................................................................................................ www.dynatem.com with companies mentioned in this article. www.rtcmagazine.com/getconnected

www.rtcmagazine.com/getconnected www.embeddedplanet.com Embedded Planet.................................................................................................. 26................................................................................... EnGenius Technologies.......................................................................................... 16.......................................................................... www.engeniustech.com/datacom GE Fanuc Embedded Systems...............................................................................4,18............................................................................. www.gefanuc.com/embedded General Micro Systems, Inc................................................................................... 75............................................................................................... Get Connected with companies mentioned in thiswww.gms4sbc.com article.

www.rtcmagazine.com/getconnected Get with companies and products featured in this section. Hunt Connected Engineering Ltd............................................................................................. 66................................................................................................. www.hunt-rtg.com www.rtcmagazine.com/getconnected

Hybricon Corporation............................................................................................. 19................................................................................................ www.hybricon.com Hypertronics Corporation........................................................................................ 8........................................................................................... www.hypertronics.com Interactive Circuits and Systems............................................................................ 46....................................................................................................www.ics-ltd.com Interface Concept.................................................................................................. 65................................................................................... www.interfaceconcept.com Kontron America..................................................................................................48,76.............................................................................................. www.kontron.com Logic Supply, Inc................................................................................................... 66............................................................................................ www.logicsupply.com National Instruments............................................................................................. 10.......................................................................................................... www.ni.com Octagon Systems..................................................................................................2,3................................................................................... www.octagonsystems.com Performance Technologies...................................................................................... 7..........................................................................................................www.pt. com Phoenix International.............................................................................................. 6.................................................................................................www.phenxint.com QPI Sales.............................................................................................................. 67................................................................................................ www.qpisales.com Real-Time & Embedded Computing Conference....................................................29,43.................................................................................................. www.rtecc.com Real-Time & Embedded Computing Conference China/Taiwan................................. 61..................................................................................................... www.rtecc.com Red Rock Technologies, Inc................................................................................... 65...........................................................................................www.redrocktech.com SBE, Inc................................................................................................................ 57...................................................................................................... www.sbei.com Sealevel Systems.................................................................................................. 51................................................................................................ www.sealevel.com Technobox, Inc...................................................................................................... 52............................................................................................. www.technobox.com Thales Computers................................................................................................. 42................................................................................... www.thalescomputers.com Themis Computer.................................................................................................. 24...................................................................................................www.themis.com Trenton Technology................................................................................................ 28..................................................................................www.trentontechnology.com VadaTech.............................................................................................................. 13............................................................................................... www.vadatech.com VME Showcase...................................................................................................... 54............................................................................................................................

RTC (Issn#1092-1524) magazine is published monthly at 905 Calle Amanecer, Ste. 250, San Clemente, CA 92673. Periodical postage paid at San Clemente and at additional mailing offices. POSTMASTER: Send address changes to RTC, 905 Calle Amanecer, Ste. 250, San Clemente, CA 92673.

September 2006