Page 1

The magazine of record for the embedded computing industry

April 2013

Development Tools Target ARM Will Software Modeling Replace Source Code? Keeping Up the Health of Distributed Systems An RTC Group Publication







Windows 8 for Embedded: A First Look

38 EN50155-Compliant DC-DC Converters for Railway Applications

40 Avionics Module Saves Space, Weight, Power and Cost


43 3U CompactPCI Serial Processor Board for Modular, High-Speed Applications



6Editorial Intelligent Systems: Chasing the Definition of Emerging Tech Trends 8

Industry Insider Latest Developments in the Embedded Marketplace

10 & Technology Newest Embedded Technology Used by 38Products Industry Leaders Small Form Factor Forum Custom, but without the Custom

Technology in Context


Atom, Core, ARM: What Works Where

Windows Embedded

vs. Core Processors: Finding Windows Embedded Legacy 12 Atom 28The the Right Fit Continues with Windows Embedded 8 Standard Scott Fabini, RadiSys


What Fits Where: High Performance or Low Power Processor Architecture Platforms Jack London, Kontron

TECHNOLOGY CONNECTED Advances in Industrial Networking


Discover Problems in Your Distributed System Before Itâ&#x20AC;&#x2122;s Too Late Ronald Leung, Real-Time Innovations

John R. Malin and Sean D. Liming, Annabooks

TECHNOLOGY DEPLOYED Development Tools for ARM Architecture Offers Challenges Along with Features to 34ARM Help Meet Them Shawn A. Prestridge, IAR Systems

Industry watch Software Modeling

46The End of an Era for Source Code Peter Thorne, Cambashi

Digital Subscriptions Available at RTC MAGAZINE APRIL 2013


APRIL 2013

Solid or Spin...


we go both ways

PRESIDENT John Reardon,


Ruggedized VPX Drive Storage Module Whatever your drive mount criteria, everyone knows the reputation, value and endurance of Phoenix products. The new VP1-250X, compatible with both solid state or rotating drives, has direct point-to-point connectivity or uses the PCI Express interface with the on-board SATA controller. f controlle It is available in conduction cooled, conduction with REDI covers (VITA 48) and air cooled conďŹ gurations. conďŹ guration Leading the way in rugged COTS data stortechn age technology for decades, Phoenix keeps you on the leading edge with very innovative products!

We Put the State of Art to Work


Art/Production ART DIRECTOR Kirsten Wyatt, GRAPHIC DESIGNER Michael Farina, LEAD WEB DEVELOPER Justin Herter,

Advertising/Web Advertising XXXQIFOYJOUDPNt714-283-4800 PHOENIX INTERNATIONAL IS AS 9100 REV C / ISO 9001: 2008 CERTIFIED

Untitled-1 1

2/28/13 9:52 AM


Billing Cindy Muir, (949) 226-2021


MSC Embedded Inc. Tel. +1 650 616 4068

COM Expressâ&#x201E;˘ - MSC C6C-A7 Ultimate graphics and video performance The MSC C6C-A7 module is based on AMDâ&#x20AC;&#x2DC;s Embedded R-Series platform delivering high-performance processing coupled with a premium high definition visual experience in a power efficient solution. Supporting OpenCLâ&#x201E;˘ it can boost the computing performance using the graphics engines for parallel processing.

AMD Embedded R-Series Accelerated Processing Units R-460L quad-core, 2.0/2.8 GHz R-452L quad-core, 1.6/2.4 GHz R-260H dual-core, 2.1/2.6 GHz R-252F dual-core, 1.7/2.3 GHz



Untitled-4 1


Â&#x192; AMD Radeonâ&#x201E;˘ HD 7000G Series graphics Â&#x192; Up to 16 GB DDR3 SDRAM Â&#x192; MicroSD card socket, bootable Â&#x192; Three DisplayPort/HDMI/DVI interfaces Â&#x192; VGA and  LVDS/Emb. DisplayPort Â&#x192; Four independent displays supported Â&#x192; DirectX 11, OpenGL 4.2, OpenCL 1.1

3/1/13 12:17 PM

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, Editorial Office Tom Williams, Editor-in-Chief 1669 Nelson Road, No. 2, Scotts Valley, CA 95066 Phone: (831) 335-1509

Published by The RTC Group Copyright 2013, 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.


Intelligent Systems: Chasing the Definition of Emerging Tech Trends


or some time now, we have been bandying about the term “intelligent systems.” After observing this process for awhile, I am becoming convinced that the expression intelligent systems is a sign of an effort to try to put a name on a number of diverse phenomena in order to get a handle on them, to control and use them. In this sense, then, intelligent systems is a rubric that is being applied to a number of technology trends, and we have not yet come to a general agreement as to exactly which of these trends fit under that definition and to what degree. And a more important question: does intelligent systems (IS) merely describe the phenomena or does it represent a practical attempt to harness and exploit these technology developments in a meaningful way? I’m hoping, of course, that it is the latter. Recently the Gartner Group identified what it thinks are the top ten technology trends for 2013. These include mobile devices and applications, cloud computing, the Internet of Things, Big Data and advanced analytics, among other things. The report itself did not try to define such a thing or group of trends as IS, but I think that some of these trends can, to differing degrees, be grouped under that concept, and strategies can be developed to actually make use of them in a coherent and perhaps “intelligent” way. So no one yet has a firm claim on exactly what defines intelligent systems, but here are at least some observations. The three trends I would say are absolutely essential to a world we could describe as intelligent systems are the Internet of Things, Big Data and cloud computing. The Internet of Things refers to what will be billions of connected devices that interact with the world. These will range from small sensors and actuators, consumer products like automobiles and appliances, automation controllers, surveillance systems and more. Since all of these devices are connected to the Internet, they all produce data. That data combined with data generated at higher levels by human and non-human users who make use of it, is passed out of IT and monitoring systems as Big Data. Technologies at a higher level than that of the “things” will be needed to deal with the volume, the speed and the complexity of Big Data. These engines will perform data aggregation, filtering, formatting and correlation at high speeds, often using what I would call a sub-trend, which can be thought of as high-speed processing on the fly. With the volumes and speeds of Big Data,



Tom Williams Editor-in-Chief

it is no longer possible to subject it to batch processing; it must be handled as it flows in order to be subjected to the third major trend: advanced analytics. Analytics can make use of both live data from the world of things as well as data that has already been acquired, interpreted and stored. And new analysis on new combinations of data can lead to ideas on how to use that data that may not even have been conceived of originally. Another development that could be considered a result of the previous trends is nonetheless a central aspect of what is coming to be called IS, and that is cloud computing. Although much of Big Data may be proprietary and stored behind security, a large amount will also be public and available via the cloud. This has implications in both the enterprise and the public/social sphere. Enterprise applications, data and cloud services are available to a company’s customers and employees via mobile devices such as tablets and smartphones as well as from any office. In addition to the enterprise cloud, we can expect to see the personal cloud grow to be the place where more people keep their personal content rather than on a PC. As a growing number of phones, tablets, appliances, cameras, automobiles, entertainment systems and homes have access to the Internet, their data will be stored and used from the cloud using whatever device and/or embedded app is appropriate. We are already seeing tablets used for access to enterprise systems, industrial controls and medical systems in addition to the huge number of personal apps. So it might be possible to consider intelligent systems as an infrastructure through which Big Data is generated, communicated, analyzed and ultimately used in myriad applications both commercial and private. We can therefore look at all the technologies that are used to maintain this infrastructure as the domain of intelligent systems and classify the various devices such as microcontrollers, small operating systems, network processors, wireless networks, app development tools and all the way up to large server farms as a technology realm we can confidently describe as intelligent systems. It is said that a name can give one power over things. This may be such a case. As we gradually come to agree on what we are to call this convergence of developments, we gain an understanding of what we expect of it and learn how to use it and then how to develop it further. It is truly an exciting thing to observe and participate in.


INSIDER APRIL 2013 PICMG Enhances Hardware Platform Management (HPM) Specification PICMG has announced the adoption of two new specifications in the hardware platform management (HPM) series that standardize connections to in-shelf LANs for the mandatory management controllers on xTCA boards and modules. Hardware platform management focuses on fundamental aspects of the xTCA hardware, including: collecting inventory data, budgeting power, tracking and responding to temperature events, and brokering highspeed fabric connections for xTCA systems. xTCA includes the AdvancedTCA, MicroTCA and AdvancedMC open module frameworks. The new specifications further improve the performance, robustness, high availability and serviceability of xTCA platforms by taking standardized advantage of existing in-shelf LANs for hardware platform management purposes. HPM.2 defines how xTCA management controllers can connect with the LAN facility(s) that already exists in xTCA shelves, such as the Gigabit Ethernet Base Interface that connects all the boards in an AdvancedTCA shelf. These LAN connections are dramatically faster than the mandatory 100 Kbit/s Intelligent Platform Management Bus (IPMB) links in the base xTCA architectures. The extra speed is especially beneficial for large transfers, such as firmware upgrade images using protocols standardized by the existing HPM.1 specification. HPM.3 complements HPM.2 by standardizing a vendor-independent method for assigning Internet Protocol (IP) addresses to HPM.2-capable management controllers via the Dynamic Host Configuration Protocol (DHCP). A large, distributed xTCA application can include hundreds or thousands of management controllers, and all LAN-attached controllers must have unique IP addresses. HPM.3 can substantially reduce the challenges of implementing these address assignments in a reliable and consistent fashion, even for a mixed vendor configuration.

Kontron Launches HighSpeed CompactPCI Initiative

Kontron has announced the launch of its High-Speed CompactPCI Initiative. Kontron will extend its support to include two CompactPCI standards, CompactPCI Serial (CPCI-S.0) for 3U form factors with PCIe, GbE, USB and SATA over backplane, as well as CompactPCI Serial Mesh (based on PICMG 2.20) for 6U form factors with 10GbE over backplane. This initiative gives the CompactPCI bus architecture a next generation product enhancement providing CompactPCI applications with an enormous performance boost, and extending the investment security of CompactPCI technology for another decade. The scope of applications using 3U CompactPCI Serial (CPCI-S.0) is extremely broad and varied. It ranges from multi-processor systems for



the computation of complex imaging processes to highperformance video or radar data recording through flexible SATA/RAID architectures, and from high-bandwidth wireless communication with parallel working radio modules WLAN, UMTS, LTE, right up to powerful multi-monitor systems in control rooms. Sample applications for the new modular CompactPCI Serial Mesh (based on PICMG 2.20) systems are in high-availability telecommunications and datacommunications applications for carriers and also with government and security companies. Applications include secure wireless systems, radar and sonar applications, and complex computational algorithms such as imaging processing. Now, even with smaller space requirements, existing system installations are provided with a significant increase in data throughput

when new long-term generation systems are being developed, which, in the future, could be scaled up to 40 GbE.

PC/104 Consortium Adopts PCI/104-Express and PCIe/104 Specification Update

The PC/104 Consortium has announced that the PCI/104-Express & PCIe/104 Specification, with provisions for PCI Express Generation 2 and Generation 3, was approved in February by member vote. The additions and enhancements to the specification align it with performance improvements to PCI Express 3.0. Transfer rates of up to 2 Gbyte/s, 8 Gbyte/s 16 Gbyte/s and 32 Gbyte/s are now possible on its x1, x4, x8 and x16 links. The connector technology and footprint are preserved in the specification. Layout examples

for the different PCI Express links, USB 3.0 and SATA were added to assist developers. “This revision demonstrates the commitment of the PC/104 Consortium members to keep the base specifications relevant as improvements are made to the standards used with PC/104,” stated Jeff Munch, chair of the PC/104 technical committee. “Improvements such as these are necessary to ensure that our members are offering the highest caliber of product possible.” Products based on this revision of the specification are already rolling out as PC/104 suppliers take advantage of the higher performance interconnect capability with PCI Express Gen 2 or Gen 3. This opens up new possibilities for data-intensive applications with fewer required PCI Express links.

Irida Labs and Tensilica Partner for Computer Vision Applications

Tensilica and Irida Labs have announced that they are partnering to enable availability of Irida Labs’ computer vision software portfolio on Tensilica’s new IVP imaging/video DSP (digital signal processor). Irida Labs has joined Tensilica’s Xtensions partner program and will enhance the support of new joint customers with their extensive computer vision expertise. IVP is an imaging and video data plane processor (DPU) that is targeted for the complex image, video and gesture recognition signal processing functions in mobile handsets, tablets, digital televisions (DTV), automotive, video games and computer vision

based applications. The IVP DSP has a unique instruction set tuned for imaging and video pixel processing that gives it an instruction throughput of over 16x the number of 16-bit pixel operations compared to that of the typical host CPU with single-issue vector instructions. In addition to its raw instruction throughput advantage to host CPUs, the imaging specific compound instructions supported by IVP give it a higher peak performance of 10 to 20x and much higher energy efficiency. IVP’s rich instruction set has more than 300 imaging, video and vision-oriented vector operations, each of which applies to 32 or more 16-bit pixels per cycle. Irida Labs is a platformindependent leading technology provider of software and silicon intellectual property (IP) for embedded computer vision. Its product portfolio includes embedded vision software and silicon IP for high throughput applications such as video stabilization, face detection and recognition, lowlight image enhancement, pedestrian detection and traffic sign recognition addressing consumer electronics, mobile devices and automotive markets. Founded in late 2007, the engineering team is based in Greece, with worldwide sales support.

GE Video Technology Used by NASA to Study Hurricanes and Wildfires

GE Intelligent Platforms has announced that the company’s video processing technology had been deployed on board NASA’s Global Hawk as part of the agency’s Hurricane Severe Storm Sentinel (HS3) Mission. The Global Hawk aircraft is unique in its research capabilities because of its long range and flight duration,

providing extraordinary capabilities for scientific and commercial ventures. The rugged full motion video (FMV) compression appliance, the GE daq8580, provides visual situational awareness on the missions, which study hurricanes and wildfires. The missions are a joint partnership between NASA Dryden Flight Research Center (DFRC) and the National Oceanic and Atmospheric Administration (NOAA), and see the autonomous Global Hawk deployed to conduct unprecedented atmospheric research initiatives. The daq8580 is a rugged multichannel FMV compression appliance for processing, server and storage applications in harsh, constrained environments. It is designed to address the challenges of processing, transporting and storing full motion video through video encoding, and can interface with a wide variety of analog and digital I/O and process standard video formats up to 1080p30 as well as computer resolutions up to 1600x1200. It provides exceptional compute power for video compression/decompression, video switching, format conversions, scaling, blending and many other video processing functions while enabling multichannel video compression and decompression for over 100x reduction in bandwidth without sacrificing video quality. The Global Hawk aircraft can reach altitudes above 60,000 feet and cover more than 20,000 km in extended 30 hour missions. It has been involved in several science campaigns each with specific information gathering objectives. The scope of these missions has varied from high altitude monitoring of ozone depleting molecules, to the study of large cyclones in the arctic influencing weather patterns.

Mentor Graphics Expands Automotive Linux Infotainment Business

Mentor Graphics has announced that it has expanded its automotive business unit by purchasing certain assets from MontaVista. This establishes Mentor Graphics as a major commercial provider of Linuxbased automotive in-vehicle infotainment (IVI) solutions. Once integration is complete, the Mentor Graphics automotive infotainment solution will feature the combined capabilities of the MontaVista Automotive Technology Platform (ATP) and the Mentor Embedded Infotainment Base Platform, and will be paired with advanced development tools including Sourcery CodeBench and Sourcery Analyzer. Mentor Graphics has led the standardization of Linux for automotive IVI applications, with executives serving on the boards of directors and actively contributing to numerous technical working groups, including SAE International, AUTOSAR and the GENIVI Alliance. With over 45 million automobiles sold each year, auto manufacturers are keenly aware of the impact of IVI systems as a key differentiator with consumers. This includes the need to integrate technologies such as voice recognition, touch screens, web browsing, GPS and connectivity between automobile and smartphones—all dependent on embedded software.

ZigBee Alliance Debuts Battery-Free Option with New Green Power Feature

The ZigBee Alliance has announced an eco-friendly way to power ZigBee products via

its new Green Power feature. Products, including switches and dimmers, can now be easily powered with available energy harvesting sources rather than using a battery or AC mains power, creating a nomaintenance, environmentally friendly solution. Green Power is an optional feature for the ZigBee PRO feature set that is part of the enhanced ZigBee 2012 specification recently ratified and released to Alliance members. It significantly expands the capabilities of ZigBee PRO, further strengthening its leadership position as the global standard for wireless sensor and control networks and the Internet of Things. With ZigBee PRO Green Power products, consumers and businesses can add ZigBee devices to many more areas including locations where power is unavailable, not allowed for safety reasons or for historical preservation purposes. Product manufacturers can implement ZigBee into more products with confidence, knowing ZigBee is backed by a thriving, innovative and competitive ecosystem. As a part of the extensive Alliance test and certification plan development process, it selected golden unit development kits from GreenPeak Technologies, Philips, Schneider Electric and Texas Instruments. These implementations will serve as golden units against which future ZigBee Certified products using the ZigBee 2012 specification will be tested. This testing process ensures compliance with the standard so that manufacturers can be assured of consistent communications. Testing services were provided by NTS, TRaC and TÜV Rheinland. RTC MAGAZINE APRIL 2013



FORUM Colin McCracken

Custom, but without the Custom


ustom. It means a design that’s exactly what it needs to be in order to meet an application’s requirements. It also means a design that requires a lot of non-recurring engineering (NRE) expenses for research, development, testing and certification. If only we can get the former (purpose-built) without as much of the latter (price tag). Consider the medical device market. Medical imaging applications continue to attract custom R&D investments. Radiography, ultrasound, computed tomography and MRI systems produce detailed graphics to display on ever larger monitors. Patients and doctors need not pay attention to the rocket science “behind the curtain.” The rocket science involves all disciplines from design tools to ASICs to board-level designs to algorithms implemented in firmware (FPGAs and microcontrollers) or DSPs or even GPUs running OpenCL code. Capturing waveforms, signal processing and calculating 2D and 3D graphical maps are distinct areas that come together to round out a full custom system design. It goes without saying that custom engineering is expensive. In a competitive marketplace, time-to-market and time-to-revenue are still everything, and project managers and executives alike scour the planet for special resources and building blocks to get a leg up. As medical device manufacturers continue to grow by acquiring the latest start-up companies, getting through FDA certification quickly means either the survival or demise of startups and their breakthrough technologies. Advancements in embedded processors and ecosystems, along with FPGAs that have SERDES PHYs to hang on PCIe lanes, are accelerating the move toward modular architectures. “Modular” means comprised of individually upgradable building blocks. It opens the door for commercial sourcing of items like compute/control processor modules that run popular operating systems. For line-powered medical devices, the big computeron-module winner so far is COM Express. Predominately x86oriented, this standard from PICMG is adding value within the custom medical electronics community—for design and for operations/logistics alike. Using an off-the-shelf processor module allows OEMs to get to market quickly without re-inventing the wheel by putting processor, chipset, RAM, LAN, etc. down on the main board. Instead, the main board becomes a custom carrier board with



an off-the-shelf CPU subsystem plugged into it. COM vendors are quick to tout the benefits of modules, including future upgradability and multiple price/performance points with a common carrier board; a broader range than is possible with socketed processors alone. There are two additional benefits of modules that aren’t mentioned often, yet are “priceless” to medical device manufacturers. The first is still within the realm of upgradability, but not after 5-7 years in production when the processor goes end-of-life (EOL). Rather, after 2-4 years of R&D and while still in product development, it’s imperative to upgrade to the latest processor technology just prior to submitting FDA paperwork and clinical trials and production. Sometimes this move is to squeeze a few hundred MHz more out of the processor. Sometimes it’s to reduce the thermal design power (TDP) for a given clock speed. A drop of 20-40 percent of processor power consumption can simplify the thermal design (cooling solution), or simply add a greater margin before the processor throttles down its speed. But the most common reason to do this is to re-start the 7-year lifecycle clock of processor availability. Medical system OEMs want to be in production for 10 years. At least, they want the ability to stay in production without having to tie up cash in stockpiling EOL inventory, and/or having to tie up engineering resources to re-qualify the latest processor module sooner instead of later. Processors and chipsets don’t come in pin-compatible replacements after 2-4 years, so using a module can prevent a board re-spin and debug cycle. The second benefit of modules that is rarely mentioned has to do with the high cost of the carrier board components for imaging, whether DSPs or FPGAs. In production, any boards that don’t pass functional test need to be debugged or scrapped if not fixable. If the processor and chipset are also on the same board, the failure / scrap rate and cost go up significantly. It’s better to keep them on separate boards. COM Express and other processor module standards make it possible to get the benefits—and then some—of a full custom embedded design without the full custom sticker shock. The latest quad-core processors with excellent integrated graphics lend themselves to the performance and long-term availability needs of the medical imaging market.


Your search is over! CompactPCI • COM Express • VME • PrPMC/XMC • VPX • Custom X-ES delivers the latest 3rd generation Intel® Core™ i7 processor solutions on the widest range of standard and custom form factors in the industry. With our proven record of meeting aggressive schedules, you can count on X-ES to deliver Intel Core i7 processor solutions on time, and with unparalleled customer support. Call us today to learn more.

You need it, we have it! That’s Extreme.

Extreme Engineering Solutions 608.833.1155

ploration your goal k directly age, the source. ology, d products

Technology in


Atom, Core, ARM: What Works Where

Atom vs. Core Processors: Finding the Right Fit There is a broad range of performance between the Intel Atom and Core processers. However, it is not just the performance that must be considered before choosing a processor for an applicationâ&#x20AC;&#x201D;power, size and weight requirements need to be considered as well. by Scott Fabini, RadiSys


oday we are witnessing explosive aging systems, session border controllers, growth in consumer technology, test and measurement, oil and gas equipwhere the computing power of a lap- ment and more require intensive processtop or desktop is now migrating progres- ing delivered by a Core processor. At the sively into ultrabooks, tablets and smart- other end of the scale, applications such phones. These computing platforms are as home security, automation and conbeing optimized for performance in low- trols and more can leverage the low-power er power applications, enabling this tech- Atom processor. With Intelâ&#x20AC;&#x2122;s range of new nological trend toward smaller, lighter Atom and Core processors, embedded apandsolutions more efficient machines. Applying plications can now target a specific marnies providing now ket segment this technologies trend to and thecompanies. embedded market, ion into products, Whether your goalwe is to research the latest with performance and price point. The for many system deare seeing our current systems becomation Engineer, or jump to a company's technical page, the goal of Get Connected is tochallenge put you you require for whatever type of technology, signers becomes choosing the right proing smaller and lighter with a range of and productsoptimized you are searching for. cessor to meet their needs. solutions. This will enable Adopting the optimal processor for embedded systems to be feature and perthese products, or designing products that formance matched to the size, weight and scale between Atom and higher-perforpower that the specific embedded applimance Core solutions, can be a daunting cation demands, driving market growth task. The equation can be simplified by in multiple segments. starting with the basics of power, size and There are a range of embedded apweight, CPU and GPU performance, I/O plications that use Atom and Core processors, seeking the right balance of perfor- requirements and price of the application mance, power, size and cost. At the high to match the best processor solution, as end of the scale, applications such as im- shown in Table 1.

End of Article Get Connected

with companies mentioned in this article.



Typically, Atom solutions are optimized for power, but performance and

APRIL 2013 RTC MAGAZINE Get Connected with companies mentioned in this article.

I/O are still key considerations. Conversely, Core solutions are optimized for performance and I/O, but with power as a contributing factor. For applications that demand CPU performance but are more flexible on power and price, Core solutions are generally a better fit. The distinguishing factor that really stands out when considering Atom solutions in particular is the low power draw of the CPU. When talking about a CPUâ&#x20AC;&#x2122;s power draw, it is referred to as Thermal Design Power, or TDP. TDP represents the maximum power the CPU will draw when running real-world applications. Modern Atom CPUs range from 13W down to 3.5W TDP, whereas Core CPUs range from 17-47W TDP. The low-power aspect of Atom solutions can be especially important for cordless battery powered applications, such as portable medical devices or robotics. Batteries are typically rated in watt-hours, and the lower the watts the system draws, the more hours the battery will last. Alternatively, a system drawing lower power can have a smaller battery with a lower watt-hour

technology in context

Size and Weight

Weight is another important facet when choosing between Atom and Core processors. In motorized embedded applications such as unmanned vehicles and robotics, a lower weight will allow the vehicle to travel longer distances. Applications that are not battery powered can also benefit from a low-power solution based on Atom or Core technology. A lower-power solution requires a smaller footprint or lower height profile for the heat-sink and fans used to cool the device. This might enable a fixed enterprise telecom application to fit in a 1U server height rather than a 2U. Aluminum or copper heat-sinks are typically heavy, therefore a smaller solution also enables a lighter solution. Atom solutions are also physically smaller at the chip-level, therefore modules based on Atom processors can often fit into a smaller form factor. In COM Express solutions, for example, Atom-based products fit into the 95 x 95 mm Compact form factor, whereas the current higherpower Core solutions fit into the 95 x 125 mm Basic form factor. The Atom achieves this remarkable level of power-performance

Atom vs. Core CPU Performance/Watt Performance Test 7.0 Benchmark 90




80 70 60 50 40 30 20 10 0 N2600 Atom 3.5W

N2800 Atom 6.5W

D2550 Atom 10W

3517UE Core 17W

3555LE Core 25W

3612QE Core 35W

3615QE Core 47W

Figure 1 CPU Performance/watt scales well transitioning from Atom to Core. Performance/watt provides a clear scale for selecting the CPU processing power required.

Atom vs. Core CPU Performance/Watt 3DMark06 Benchmark 200 180





rating, thereby enabling a smaller physical solution. The Core series of CPUs is the workhorse processor of many embedded applications, with technology analogous to a modern laptop computer. There is an excellent level of scalability within the family, with offerings available at 17W, 25W, 35W and 47W power bands. At each power band, there is another level of granularity, with Core i7, i5, i3 and Celeron options often available. Along this scale, i7s are optimized for performance whereas Celerons are optimized for price. Core solutions can also be offered in battery-powered applications, as well as fixed devices requiring low power.

140 120 100 80 60 40 20

0 N2600 Atom 3.5W

N2800 Atom 6.5W

D2550 Atom 10W

3517UE Core 17W

3555LE Core 25W

3612QE Core 35W

3615QE Core 47W

Figure 2 As more applications require signal or image processing, the use of embedded GPUs helps to improve overall performance without major power increases.

through optimization of several key areas. It is important to understand these optimizations when selecting between the Atom and Core CPU families. Beyond the more physical constraints of

size, weight and power, it is important to remember that the CPU is there to run the application. It must have the performance and I/O to meet the requirements. RTC MAGAZINE APRIL 2013


technology in context

Figure 3 Atom on the left and Core i7 on the right; COM Express modules from Radisys. The Atom solution is smaller but both meet COM Express standards and can be interchanged in an application.


When evaluating solutions it should be noted that, in general, a Core series CPU will typically out-perform an Atom CPU under a GHz-to-GHz comparison. This is because the Atom solution is optimized for low power, and to keep the power of Atom chips low, cache sizes are typically smaller (512 Kbyte-1 Mbyte instead of 2-3 Mbyte), memory speed slower (DDR3-1333 vs. DDR3-1600) with less capacity (4 Gbyte vs. 8-16 Gbyte), and some of the pipelining within the CPU is simplified. Atom CPUs optimize these factors in a manner that is balanced across a wide variety of applications. This enables scalable solutions that are not bottlenecked by one or more of these aspects as the application scales to the low-power devices (Figure 1). Core solutions typically offer the leading edge of Intel’s “tick-tock” model that promises to follow every microarchitectural change (“tock”) with a die shrink of the process technology (“tick”). These advances provide the latest architecture and transistor technology to bring out the best performance at a given power. Higher cache sizes, faster and higher capacity memory, dual-channel memory and deeper parallelization are all aspects focused on performance. Core i7s unleash higher core counts and faster frequencies; whereas Celerons slow things down to make them available at a lower price. Additionally, Core CPUs offer ECC protection on memory, which is important



for many applications ranging from enterprise telecom to defense applications. Core solutions also offer some special vPRO CPU features such as VT virtualization support, AMT remote manageability, TXT trusted execution technology and Turbo mode, which allows the CPU to boost performance when the thermal environment allows for it. Atom CPUs and the lower-tier Celeron and i3 CPUs typically shed these features in the interest of either power or price. For applications that demand CPU performance but are more flexible on power and price, Core solutions are generally a good fit.


The grey zone between Atom and Core processors comes with the addition of more real-time 3D graphics rendering to the application. Atom CPUs use PowerVR GPU technology, which is the same technology commonly used in modern smartphones, tablets and navigation systems. Such systems can achieve respectable graphics performance at very good screen resolutions. For embedded applications such as Human-Machine Interfaces (HMI) running a simple 3D user interface, this technology is more than sufficient. Even moving-map displays and portable ultrasound can be scaled onto Atom solutions, but results may vary based upon the amount of information to be presented. These are areas where the value of a scalable “common platform” solution en-

abled with both Atom and Core platforms could really shine. Such a common platform in portable ultrasound, for example, could enable manufacturers to offer a “value product” based on Atom technology for areas that require very basic imaging capability, and a “premium product” based on Core technology for areas where the quality of the image and high drawrates for real-time imaging are required. Core series CPUs have received significant improvements to GPU capability in recent years, with third and fourth generation Core solutions offering a real breakthrough in performance. The transition to an architecture based on scalable execution units (analogous to CPU “Cores,” but for GPUs) has provided a significant jump in GPU performance for imaging applications such as portable ultrasound. It also enables more vibrant and detailed displays for mapping and video-streaming applications in public safety and defense. The scalable GPU architecture also allows for a more granular offering in terms of price-performance (Figure 2). There are Core i7 solutions available with 16 EUs, alongside Celeron solutions with a modest 4 EUs for a lower price point. This level of CPU/GPU scalability is perfect for embedded programs looking to optimize across a broad cost/ performance curve. Core series products also support the latest graphics libraries, such as DirectX 11, OpenGL 3.1 and OpenCL 1.1, whereas Atom products tend to use prior-generation libraries (DX 9/10, OpenGL 2.0, no OpenCL


As mentioned earlier, Atom is optimized for the needs of lower-power and smaller form factor applications. Atom CPUs are optimized with a smaller subset of I/O such as PCI Express, SATA and USB 3.0. This is reflective of the applications that use these lower-power devices. It makes little sense to attach a 150W x16 PCI Express GPU to a 3.5W CPU, so Atoms are typically limited to 3-4 lanes of PCI Express. Similarly, only two SATA lanes are generally provided, sufficient for an HDD, SSD or optical drive, which is typical of small form factor embedded systems. Core series CPUs offer more expansion options, typically enabling the x16

technology in context

PCI Express plus another 7 x1 lanes. These lanes can often be broken up to x4, if required. Six SATA lanes are typical, as well as multiple USB 3.0 lanes. Clearly if significant I/O expansion is required, a Core platform will make the most sense, but a scalable solution supporting both Atom and Core solutions can provide that extra level of scalability. This is often the case in enterprise telecommunications applications, which use a Core based solution with many PCI Express ports interfacing to multiple Ethernet controllers as their “premium” platform. Then they will also offer a “value” platform utilizing an Atom CPU with just a few Ethernet ports. Similar to the portable ultrasound example above, the transition between Core and Atom based solutions provide a scalability that customers really value. This scalability can be achieved in a single design, using a COM Express modular solution so that the Atom vs. Core decision is a configuration option instead of a completely separate design.

Scaling between Atom and Core

Unfortunately, Atom and Core solutions are not pin-compatible with one another. An SBC designer must typically choose between the two before starting the design. Additionally, the “tick-tock” model and Moore’s Law means a better solution will probably be available next year. Wouldn’t it be nice to be able to offer one design that can scale between the two, or upgrade to the next version once it is available? Many customers offering long-life embedded products in the medical, defense and industrial automation sectors have unlocked real value by offering products that provide that very strong assertion using RadiSys COM Express solutions. COM Express is an industry standard solution that offers the Core CPU complex of processor, memory, chipset and network controller in a single modular solution. The I/O is broken-out to a high-density and reliable connector. COM Express is more than a “mezzanine card” with limited I/O capability. Essentially all of the I/O of Core and Atom processors are made available through the connector: PCI Express, SATA, USB, LPC, Audio,

Atom vs. Core for Optimal Scalability Product Code


CEHM76-1047UE-0 CENM10-D2250-0 CENM10-N2600-0

Product Family





COM Express Type COM Express R 2.0 COM Express R 2.0 COM Express R 2.0 COM Express R 2.0 95x125mm 95x125mm 95x125mm 95x125mm Type 6 Type 6 Type 6 Type 6 Processor/Chipset Processor Family

Intel 3rd Generation Intel 3rd Generation Intel ATOM Core i7 Celeron

Intel ATOM

Processor Name




















Processor Tdp





Memory Capacity





Memory Implementation

Dual Channel, 2 SODIMMs

Dual Channel, 2 SODIMMs

Single Channel, 1 SODIMM

Single Channel, 1 SODIMM

Intel Graphics Engine


HD Graphics

GMA 3650

GMA 3600

GPU Execution Units

12x Execution Units 6x Execution Units 4x Pixel Pipeline

4x Pixel Pipeline

GPU Frequency Base-Turbo

650-1000 MHz

350-900 MHz

667 MHz

400 MHz

Independent Displays





PCI Express

7 PCIe x1

7 PCIe x1

3 PCIe x1

3 PCIe x1

PCIE x16

1 PCIe x16 Gen3, configurable

1 PCIe x16 Gen3, configurable



USB 2.0/3.0





SATA 2.0/3.0







TABLE 1 Atom vs. Core comparisons—Processor speed, memory bandwidth power consumption and I/O capabilities, must be considered for each application.

SMbus and I2C, and video interfaces like HDMI and LVDS are all accessible. It is your “platform Core” that you can build your system around. And by using COM Express modules with a common standards-defined pin-out, you can enable a scalable solution and swap between Atom and Core based “value” and “premium” offerings (Figure 3). Atom and Core platforms each provide different benefits, and it’s important to evaluate the power, size and weight, CPU, GPU I/O requirements and price for each application. In some instances a scalable common platform solution

that supports both Atom and Core platforms is the ideal choice, enabling one platform to target multiple different segments while reducing development costs. Companies such as RadiSys offer COM Express solutions, which are ideal for a scalable platform that utilizes Atom and Core processors to provide differentiated applications that help deliver the best product to market. RadiSys Hillsboro, OR. (513) 615-1100. [].



Technology in


Atom, Core, ARM: What Works Where

What Fits Where: High Performance or Low Power Processor Architecture Platforms Meeting the demands of embedded applications for low power and low cost while still meeting performance requirements can seem daunting at first. Fortunately there is a wide range of choices and considerations to narrow the selection that can lead to the right result. by Jack London, Kontron


he advancements in processor technology over the last decade are nothing short of amazing. And, the speed at which they are developed is equally impressive. Not only has performance and throughput increased, but designers have found a way to integrate so much functionality and interface options that enable the transition to more intelligent systems development. It would seem that every new embedded system needs to be lightning fast and that developers are sourcing primarily the latest high-performance processor architectures. However, this doesnâ&#x20AC;&#x2122;t give a true sense of overall market demandâ&#x20AC;&#x201D;not every application has high-performance compute-intensive requirements. Sometimes performance and throughput take a backseat to size, weight and power design considerations. So what processor architecture fits best and where? With a diverse array of embedded systems requirements today, the answer clearly is system dependent.

Market Needs Drive the Range of Processor Options

High-performance embedded computing applications such as image processing in automation and medical ap-



Figure 1 Designed to bring leading-edge performance, low power consumption and low heat dissipation to a broad range of applications, Kontron CompactPCI processor boards support an extensive range of interfaces with SATA 6 Gbit/s and USB3.0 support, three independent graphics outputs as well as up to 5x Gigabit Ethernet interfaces connected via PCI Express.

plications, embedded cloud computing and digital signal processing in communications, as well as signals intelligence in military and aerospace platforms, all share a common demand in terms of highest possible signal processing performance,

throughput and graphics processing. At the same time, this demand is frequently coupled with strict requirements in regard to power efficiency to deliver a level of performance per watt that matches the needs of size, weight and power (SWaP)-

PC/104, PC/104 Express & ISM Showcase

Featuring the latest in PC/104, PC/104 Express & ISM technologies PC/104-Plus Watchdog Timer Board

High-Speed USB/104 Module Provides 96 Lines of Optimized Digital I/O

Temperature measurement, monitor, and alarm Fan status and speed control PCI/104 power monitor / limit alarm interrupt Opto-isolated input to trigger reset General purpose opto-isolated input, two outputs Two general purpose 8-bit A/D inputs External fused 5V and 12V power Light sensor for enclosure security

ACCES I/O Products, Inc. Phone: (858) 550-9559 Fax: (858) 550-7322

High-speed USB 2.0 device Twelve or six 8-bit ports independently selectable for inputs or outputs All I/O lines buffered with 32 mA source, 64mA sink current capabilities Jumper selectable I/O pulled up to 5V for contact monitoring, down to ground, or floating Resettable fused +5VDC outputs per 50-pin connector PC/104 module size and mounting compatibility

ACCES I/O Products, Inc. E-mail: Web:

Phone: (858) 550-9559 Fax: (858) 550-7322

ADLQM67PC – Industry’s Only PC/104 2nd Gen Intel Core Quad Platform

USB Wi-Fi Modules 802.11b/g/n Compliant

Intel® Core™ i7 Gen2 Quad and Duo Core 2.1GHz – 2.2GHz Up to 8GB DDR3-1333 DRAM Type 1 Bottom-Stacking PCIe/104 V2.01 Gen2 protocol 2x SATA 6Gb/s with RAID 2x GLAN Ethernet 2x RS232 COM Ports, 8x USB2.0 Ports Video - VGA/DVI/HDMI/DisplayPort/ LVDS Standard -25C to +70C, -40C to +85C Option

ADL Embedded Solutions Inc. Phone: (858) 490-0597 Fax: (858) 490-0599

E-mail: Web:

Radicom Research, Inc. Phone: (408) 383-9006 Fax: (408) 383-9007

Schroff® 450/40 ATCA Chassis Supports Next Generation ATCA Board Requirements

Schroff Phone: (401) 732-3770 Fax: (401) 738-7988

Minimize time-to-market for critical, high-availability applications 2, 6, and 14 slot backplanes 40 Gbps (10 GBASE-KR) transmission rate – bussed or radial IPMB Increased power availability of 450W per slot Enhanced cooling capabilities of 400W per slot in the front Improved cooling capabilities of 50W per slot in the RTM Both integrated AC or DC Redundant Power Entry Module options available in all slot counts E-mail: Web:

E-mail: Web:

USB 2.0 hot swappable interface Compatible with USB1.1 and USB2.0 host controllers Up to 300Mbps receive and 150Mbps transmit rate using 40MHz bandwidth Up to 150Mbps receive and 75Mbps transmit rate using 20MHz bandwidth 1 x 2 MIMO technology for exceptional reception and throughput 2 U.FL TX/RX antenna ports Wi-Fi security using WEP, WPA and WPA2 Compact size: 1.0” x 1.0” x 0.25” (Modules) Windows 2K, XP, Vista, Win7 support Linux 2.4/2.6 support RoHS compliant E-mail: Web:

Fanless, Extended Temperature Atom™ Powered PC/104-Plus SBC

Module: PPM-C393-S

1.66GHz N455 Intel® Atom™ processor Runs Linux, Windows® and other x86compatible operating systems Up to 2GB of DDR3 SODIMM supported Simultaneous LVDS and CRT video Intel® Gigabit Ethernet controller Four serial COM ports (two RS-232, two RS-232/422/485) PC/104-Plus and PC/104 expansion Long-term product availability

WinSystems, Inc. Phone: (817) 274-7553 Fax: (817) 548-1358

E-mail: Web:

technology in context

Figure 2 Offering single core ARM Cortex A8 technology performance of up to 800 MHz in an exceptionally small 82 mm x 50 mm footprint, SMARC modules based on Texas Instrumentâ&#x20AC;&#x2122;s AM3874 Sitara ARM processors deliver the computing power required by the diverse range of smart mobile devices.

constrained applications that characterize many embedded deployments. One prime example of increased performance levels, lower thermal design power (TDP), improved high-end embedded graphics performance, optimized security and broad scalability are the latest embedded computing platforms based on the 3rd generation Intel Core processors. A wide selection of standards-based embedded form factors has integrated these processors making them well-suited to streamline the development of a broad array of high-performance embedded applications. Part of the reason is Intelâ&#x20AC;&#x2122;s new 22-nanometer (nm) 3-D tri-gate transistor technology, which offers several architectural improvements that deliver the feature set required for these challenging and complex demands. On the other end of the spectrum, a growing number of military, industrial automation/HMI, digital signage and medical applications have mobile, harsh environment and space-constrained requirements that demand extremely low power consumption. These embedded systems need ultralow-power embedded platforms that offer operation at three watts or below in a slim profile design. OEM developers also need these computing platforms to provide scalable and flexible design solutions that deliver the latest interface and graphics support



along with performance that can deliver the same functionality as consumers have come to expect from their smart mobile devices. A leading processor architecture solution for these types of embedded designs is ARM. Many developers have seen that ARM has advanced to provide an open systems approach that supports a wider range of interfaces and features. ARM processors have been proven to provide an attractive platform for low-profile, high-density embedded applications while also offering long-term availability and scalability that allows efficient development for multiple product generations.

Core Processor Performance Advantages

Offering significant advantages for a range of compute-intensive applications, the third generation Intel Core processors provide up to 20% enhanced computing power and up to 40% increased performance per watt compared to designs based on the 2nd generation Intel Core processors. With the third generation Intel Core processors, Intel is introducing the tri-gate transistor, in which the transistor channel is raised into the 3rd dimension. Adding a third dimension to transistors allows Intel to increase transistor density to 1.4 billion transistors on a die size of 160 mm2 and offer significantly greater capabilities.

Embedded computing platforms that implement the new processors enable OEMs to build applications with increased processing density and I/O bandwidth. Enabling designers to utilize the power of the latest quad-core Intel processors, suppliers have integrated them now on small form factors such as COM Express, AdvancedMC and 3U VPX to meet improved size, weight and power requirements. Additional improvements, such as extended Intel Advanced Vector Extensions (AVX) and SSE instructions as well as the support for OpenCL 1.1, provide developers efficient tools to reduce the development effort and time-to-market for parallel computing applications. Together with the 3-D tri-gate transistor technology, Intel also introduced a new graphics architecture that offers up to twice the HD media and 3-D graphics performance compared to its predecessor. Further standardized platform advancements offer integrated Intel HD Graphics 4000, a feature that delivers 30% more execution units than the previous generation, and natively supports three independent digital display interfaces, enables sophisticated graphics-intensive applications such as infotainment and digital signage with an immersive user experience (Table 1).

Putting Core Performance to Work

Offering up to 40% increased performance per watt compared to platforms based on second Generation Intel Core processors, this increase in power efficiency allows applications with tight thermal envelopes to take advantage of the parallel performance of up to four CPU cores and eight threads. This not only enables highly efficient small form factor applications, such as unmanned aerial vehicles (UAVs), but, due to the high level of integration, also allows consolidating multiple computing systems onto one single platform through virtualization. This is a key feature in Intelâ&#x20AC;&#x2122;s Vpro feature suite, which is available on the Core series platforms, and results in reduced hardware costs. The decreased system count also results in higher MTBF values of the consolidated installation and helps to save valuable space for SWaP optimized high-performance embedded computing applications. Power-hungry applications have a solution in the latest Core-based platforms that

technology in context


Architecture Improvements


Intel HD Graphics 2500/4000

Intel HD Graphics 2000/3000

Unified Shader Architecture



Execution Units (EUs)

6/16 EUs

6/12 EUs

Dedicated Math box



Media processing



Targeted OS Optimizations

Windows 7/Windows 8 Windows 7/Vista/XP

Independent Displays



Up to 1350 MHz

Up to 1350 MHz

Direct Support



Open GL Support

Open GL 3.1

Open GL 3.0

Shader Model Support

SM 5.0

SM 4.1

Dynamic Frequency Scaling

Yes (mobile and DT)

Yes (mobile and DT)

Maximum Resolution

2560 x 1600

2560 x 1600

HDMI (V.1.4 with 3D Support)



3D Performance Core Frequency

SENSORAY embedded electronics experts Made in USA

Model 953-ET

PCI/104 Express 4-Channel A/V CODEC

t4 NTSC/PAL video input/outputs t4 stereo audio inputs/outputs tH.264 HP@L3, MPEG-4 ASP, MJPEG MJPEG video; AAC, G.711, PCM audio tUltra-low latency video preview concurrent w/compressed capture tFull duplex hardware encode/decode tText overlay, GPIO t-40° to +85C° extended temp Info at

TABLE 1 Architecture and 3-D feature comparison of the third and the second generation Intel Core processor.

provide Intel Turbo Boost 2.0 technology, which increases the clock speeds of both the processor cores and the graphics unit independently. This automatically shifts processor cores and processor graphics resources to accelerate performance, tailoring a workload to give users an immediate performance boost for their applications whenever needed. Depending on the load, the actual speed can be increased by up to 40%. Supporting faster interfaces makes it possible to achieve high-speed system requirements. For telecommunication applications, the memory controller now supports 1600 MHz to connect to DDR31600 memory. The same is true for the processor’s 16 PCIe 3.0 lanes, which is the next evolution of the general-purpose PCI Express I/O standard. Furthermore, third generation Intel Core processors also deliver an improved integrated graphics unit that provides support for three independent displays that enables multi-screen applications without the need for an additional dedicated graphics controller. These new features make embedded platforms equipped with the 3rd generation Intel Core processors a suitable solution for applications in which a huge amount of data has to be processed in a limited thermal envelope. The application areas that are best able to benefit from advanced technologies and improved graphics performance

are situational awareness systems such as radar, sonar, image processing, video surveillance with recognition and computeraided diagnostics (Figure 1).

Low-Power ARM Platform Design Benefits

Let us now look at where low-power processor architecture solutions fit in the equation. Today’s ARM-based solutions can deliver significant benefits by meeting a wide range of application design requirements. Of particular importance to spaceconstrained, low-power applications is the less than 1 watt operating power for ARM processors that also supports extended industrial temperature ranges. Thus ARM is well-suited for very small portable handheld devices as well as for all larger devices where power consumption cannot exceed a few watts. ARM processors, too, have demonstrated exceptional performance with dual/ quad core CPU offerings. ARM-based platforms also feature long product life of 7-15 years. ARM’s ultra-low-power operation is achieved due to SoC architecture, which consolidates all of the necessary processing, I/O interfaces and graphics processing into a single piece of silicon and eliminates the need for a supporting chipset and/or additional bridge components. Moreover, ARMbased platforms can require fewer pins and fewer power circuitry components thus re-

Model 812 Cable included

PCI Express Video+Audio Capture

t8 Channel frame grabber t8 Channel mono audio t8 Composite video inputs t8 Digital I/Os for camera control tD1/VGA/CIF/SIF t240 fps for NTSC/200 fps for PAL tPCI-Express x 1 interface Info at

Model 826

Measurement & Control

t6 advanced 32-bit counters t 16 analog inputs, 16-bit, 300 ks/s t8 analog outputs, 16-bit, 900 ks/s t48 digital I/Os with edge capture tSupports incremental encoders, pwm/pulse generation, pulse/ frequency/period measurement tWatchdog timer Info at

SENSORAY. com | 503.684.8005

Untitled-5 1



3/26/13 12:40 PM

technology in context ducing cost. Plus, ARM-based module platforms allow embedded OEMs to implement simplified passive cooling and other effective thermal management methodologies needed in mobile, compact designs. A new standard for ARM/SoC-based building blocks has been recently ratified. The Smart Mobility ARChitecture (SMARC) specification from the Standardization Group for Embedded Technologies (SGET) defines two extremely flat ultra-low-power Computer-on-Modules (COMs) form factor sizes. The standardization of SMARC modules puts in place a comprehensive ecosystem to support scalable ARM-based building blocks. Crucial in streamlining development of low-power systems is the breadth of ARM processors from Texas Instruments, NVIDIA and Freescale. With SMARC modules, embedded computing suppliers are now leveraging the long list of features in ARM processors into a new generation of platforms. Providing various levels of high performance and low power, and from high end to low, ARM-based SMARC COMs are the building block solutions that help OEMs develop a strong portfolio of products using a common hardware platform.

Where Low-Power Module Building Blocks Work Best

The primary design considerations for ultra-low-power systems can be narrowed down to performance, integrated graphics support, thermal design power

(TDP) and cost. At the top end of the ultra-low-power performance offerings are SMARC modules based on the NVIDIA Tegra3, which offers up to 1.2 GHz performance and graphics support for LVDS and HDMI, making it a compelling choice for graphic-intensive applications that require low power, low profile and the ability to withstand harsh environments. Taking it down a notch, SMARC COMs that integrate ARM Cortex A9based quad core Freescale i.MX 6 and the NVIDIA Tegra3 processors support higher-end applications that require significant CPU performance coupled with advanced graphics capabilities. Freescale i.MX 6-based COMs enable efficient application design for applications that have extremely compact, fanless thermal and superior graphics performance requirements. SMARC modules based on the single core Texas Instruments AM3874 Sitara processor are best suited to more mainstream applications that have lower cost limitations. These ARM Core A8 modules still provide excellent graphics support with extremely low TDP, but also deliver optimized cost to performance ratios making them good for harsh operation applications that fit on the lower end of the performance spectrum (Figure 2). The application areas for SMARC modules include industrial automation tablet-based systems, outdoor POS digital signage, HMI systems and portable data terminals such as a healthcare pa-

tient monitors. These applications have requirements for features such as wireless connectivity, touchscreen displays, video, integrated camera or bar code scanner support. There are numerous end-user advantages of low-power ARM modules for these systems. It gives users the ability to use the application for a longer period of time without a charge, eliminates the need for a thermal cooling solution, and permits a more portable solution in terms of a lower and thinner profile system that is ultimately more user-friendly. Advancements in processor architectures and the standardized embedded computing platforms that integrate them are ensuring the diverse requirements are met. The functionality of compute-intensive applications that need the highest possible signal processing performance, throughput and graphics processing are well-supported with the latest third generation Intel Core processors. Likewise, mobile, harsh environment and space-constrained applications that require extremely low power consumption have a ready solution with new SMARC modules that tout 3 watts or below power consumption in a slim profile design. High performance and low-power-based computing platforms are born from the specific market application needs that demand them. Kontron Poway, CA. (888) 294-4558. [].

Intel速 AtomTM Processor


Untitled-1 1


3/26/13 12:30 PM

Where Apps Meet Security and the Internet of Things. The Amphion Forum brings together 400 thought leaders from academia, business, government and technology for an exclusive summit to discuss the threats and opportunities presented by the unprecedented proliferation of connected devices.

Washington Marriott at Metro Center

Washington DC

Mobile App Security

Enterprise App Deployment

Critical Infrastructure Security

Medical Mobility

Register Now

May 16â&#x20AC;&#x201C;17 Use code EB030713 to register for only $295

Smartphones in Defense

Home Automation

Factory Floor Apps

Secure App Gateways

In-Flight Devices

As mobile computing platforms pervade industrial, government, and enterprise network systems, security among connected devices has become a tipping-point topic for enterprise IT, lines of business and app developers alike. The Amphion Forum delivers cutting-edge content and invaluable insights, along with terriď&#x192;&#x17E;c networking opportunities.



connected Advances in Industrial Networking

Discover Problems in Your Distributed System Before It’s Too Late In a distributed system, components running on different OSs and applications written in different languages all work together as a single reliable system. As the complexity of these systems increases, it becomes more important to gain visibility and expose potential problems before they jeopardize the production system. by Ronald Leung, Real-Time Innovations


istributed systems are composed of many sub components, often times built separately by different teams, and are implemented and deployed in multiple phases. There are many ways to facilitate communications between different components, but to truly design systems in a loosely coupled manner where you can easily plug in additional components without affecting existing deployed applications remains a very difficult problem. Once the system is deployed, the need to have visibility of what’s happening will certainly arise, and without a flexible architecture to insert additional tools for monitoring the system, this would become an impossible mission. The term “distributed systems” refers to myriad variations. One example is the technologies in a passenger car that assist driver safety. Different radars and sensors are located around the car to monitor the surroundings, and these data need to be transmitted in real time to other components for analysis, which triggers further actions to yet other components to help the driver avoid collisions. Other examples of distributed systems include



military systems inside navy ships and unmanned vehicles, medical imaging systems, asset tracking systems and air traffic control systems. These systems may look very different from a high level, but beneath the hood they all share many similar characteristics. All of these are large scale systems composed of lots of separate processors and embedded devices, and it is absolutely critical that all components in the system work together in real time and be highly reliable. Getting the system up and running properly is only the first problem. Administering and maintaining the system to make sure it is behaving as you expect without interrupting its components adds additional challenges. Far too often the need to add additional tools and components is considered only as an afterthought, and by the time you realize you need these tools, adding them into the deployed system requires major and costly changes. For instance, between two communicating nodes, how do you intercept and visualize the live data flow, without interrupting either node? How do you verify that the attributes of nodes that are

meant to communicate have compatible parameters? How do you keep a record of all the data on the wire, and possibly replay the same or a subset of the data for testing or debugging purposes? And with components that are designed to be loosely coupled and can run independently, how can you manage and administer all these components in one place?

Building on Top of a Proven Architecture

One of the key criteria for a flexible architecture is the ability to plug new modules into the existing system without modifications to any deployed components. Data Distribution Service (DDS) is an open and growing standard maintained by the Object Management Group (OMG) that is designed specifically to address the challenges of a loosely coupled, flexible, yet seamlessly integrated distributed system. When applications need to send data and messages between each other in a distributed environment, one way is the “message-centric” approach where the infrastructure typically has very little infor-

technology connected

mation about the message. The infrastructureâ&#x20AC;&#x2122;s job is to deliver these indistinguishable messages equally. In contrast to that is a â&#x20AC;&#x153;data-centricâ&#x20AC;? middleware, where the infrastructure understands the data format. It is similar to a relational database, where the database is fully aware of the schema of the data. It knows the definition of all the tables, all the columns, primary keys and foreign keys, and you can have triggers where, upon the occurrence of a certain event, the system can automatically react by taking a corresponding action. In a data-centric approach, the data model is also explicit and well defined, so that all applications fully understand the meaning of the data. The open Data Distribution Standard specifies both the API interface and the wire protocol. This allows applications written in different languages and running on different platforms to interoperate effortlessly. DDS implementations from different vendors can also interoperate, as long as the implementation is coded according to the standard specification. Various vendors, including IBM, have provided their own implementation of DDS and participated in demos to show interoperability across vendors.

Figure 1 Screenshot of Analyzer checking for entities that have incompatible settings.

Automatic System Analysis

With Data Distribution Service, the middleware infrastructure takes care of the discovery and communications between all available applications. While a lot of this work is done automatically, there are many ways to customize the behavior of the communication through Quality of Service (QoS) values of each data publisher and subscriber. There are QoS values for configuring reliability, durability and filtering, among many other behaviors that you can configure. Configuration is done by simply specifying the appropriate value through an XML file or through code in your application. Having these QoS values significantly simplifies development, improves

Figure 2 Analyzer displaying detailed explanations for mismatched entities.



technology connected

Figure 3 WireShark analyzing data packets sent by an RTI Connext application.

efficiency, and also makes your system very flexible. You can change many behaviors in the communication by simply updating these QoS values without changing your application code. To illustrate with a specific example, one of the many QoS values available is the Deadline. The Deadline QoS specifies the maximum time between data samples. You can set it on a data writer, which declares a new data sample will be published at least every x seconds. You can also set it on a reader, which specifies that a reader wants to receive one data sample at least every y seconds. Consider a case where you have a sensor application with a data writer that sets the deadline value to 1 second. This lets other components know that it agrees to send out a new data at least once every second. It can send a lot more frequently than that, but after it sends out one sample, before the next second on the clock ticks in, it promises to send at least one more data sample. For the control application that is listening for data from the sensor application, you can specify a deadline on the reader of say 10 seconds. This means that the reader requires a new data every 10 seconds. Now it becomes obvious that the QoS value of one application may have impact in another application, and the values need to be compatible for them to communicate. Using the current example, if a writer promises to write a data every one second, and the reader wants a data only every other 10 seconds, the values are compatible since the writer publishes data more frequently than the reader requires. And since the middleware is aware of these QoS values, it will do all the work to match readers and writers that have compatible QoS values. Same for the opposite situation: If the values were incompatible, the middleware would also know not to attempt sending data between these applications and waste bandwidth.

Leveraging Appropriate Tools

Figure 4 Monitorâ&#x20AC;&#x2122;s system overview displaying a summary of the entities on the network, and highlighting components that have warnings or errors.



RTIâ&#x20AC;&#x2122;s comprehensive tool suite is used in all phases of development. In initial phases, a common issue that prevents communication is incompatible settings between components. To quickly see all the participants and whether the QoS values are compatible, the Analyzer of-

technology connected

fers the QoS Match Analysis. Analyzer will discover all participating entities on the network and generate an automatic report that highlights the number of readers and writers that are reading or writing a particular data type and whether they are matched or mismatched to each other. The example in Figure 1 discovered four mismatches and you can drill down into the mismatched entities to see why. Double clicking on the entity reveals the exact QoS that didnâ&#x20AC;&#x2122;t match. In Figure 2, it highlighted the Deadline QoS that are incompatible. The writer has a default value of Infinite, where the data reader actually requires a data sample every 100 ms. Since the values are incompatible, there will be no data between these data writers and readers. There are many other QoS values that can be tuned, and it is not easy to keep track of all the values without having the right tools. Using the Analyzer tool to run a quick report makes it much easier to identify the problematic values.

Real-Time Publish Subscribe Protocol that DDS uses, and you can use it to see low-level details in each data packet that is flowing on the network (Figure 3).

Deep Monitoring

It is important to monitor your system to make sure it is running without any problems. In addition to the data that you can retrieve from simply tapping onto the wire, additional insights require deeper

monitoring and instrumentation. RTI provides an optional monitoring library that, once enabled, can collect various statistics from readers, writers and DDS entities, and then publish these data using DDS. The companion RTI Monitor application can be used to see the data collected by the monitoring library. There are many different views and panels that will help you understand your system. On the left panel in Figure 4, you

Wire Tapping

Since the data model is explicitly defined, the architecture is well adapted for additional applications or tools to tap onto the wire. The tools simply subscribe to the same data topics, and no changes are required in the data publisher or subscriber. One use case for wiretapping is to visualize the data that is on the wire. RTI offers a plugin for Excel that allows you to bring in data from the Connext Data bus right into a spreadsheet. You can install it on top of Microsoft Excel. This Excel plugin would detect all the available topics. You simply select the data topic and data fields that you are interested in, and then you can specify which cell location in an Excel worksheet to display this data. Once you have set up a cell to receive data, the values in the cells will be updated automatically as new data arrives, and you can use the data like any other cells in a spreadsheet. For example, you can plot the values in a bar / pie chart, and the charts will also be updated dynamically based on the latest value in the cells in the spreadsheet. Wiretapping is also necessary for advanced debugging. The popular network analyzer WireShark supports the Untitled-2 1


4/3/13 10:01 AM RTC MAGAZINE APRIL 2013

technology connected processes, and you can see the number of writers and readers in this process. The view is also color coded to highlight errors or warnings.

Centralized Administration and Logging

Figure 5 Administration Console displaying log messages from an application.

can see a system tree displaying the hosts in your system, the processes, the topics, the data writers and the data readers. On the right panel you see a system

overview. Near the bottom is a diagram where you get a graphical view of what your system looks like. In this example you can see there’s one host with two

If the application is running entirely on one host or in some sort of server, traditional log files serve as a convenient way to track critical information during run time. However, in a distributed system, gathering and collecting log files from all the components becomes a big hassle. To solve this problem, RTI has created a Distributed Logger API, which allows the application to publish log messages remotely. The API is very simple. You can log messages of different error severity, and they can be published on the network so that you can view them in a centralized location. To view live updates of these distributed log messages in one place, you can use the Administration Console. The different color-coded messages reflect each message’s error severity (Figure 5). In addition to log information, it is also difficult to see system information

Low Power, High Performance ARM Solutions P









CSB1724-88F6282 1.6Ghz 88F6282






1GByte 16-Bit Wide DDR3-800 Dual PCIe x1 and Dual GIGe



Dual SATA Gen 2 and Dual USB

1Ghz Dual Core MV78200

1.33Ghz Quad Core MV78460

1GByte 64-Bit DDR2-800 w/ECC

2GByte 64-Bit DDR3-1333 w/ECC

PCIe x4 and Dual x1

PCIe x4 and Dual x1 Gen 2

Dual SATA Gen 2 and Dual USB

Dual SATA Gen 2 and Dual USB

Quad 10/100/1000 Ethernet

Quad 10/100/1000 Ethernet

For more information Tel: 401-349-3999 Email: Web:


Untitled-1 1


Coming soon… CSB1626T6-MV78460 COM Express Type 6 Dual 10G XAUI/SFI Option 3/28/13 1:29 PM

technology connected from each host in one place without proper tools that collect live statistics from these hosts. The Administration Console displays the CPU and memory usage of each remote host. In Figure 6 you can see one hostâ&#x20AC;&#x2122;s CPU is at 14%, and the other one is at 0%. You can also see the amount of free and used memory of each host summarized in one single table. The information is also color coded to help easily spot systems that may need attention. As the name may suggest, the Administration Console serves as the centralized administration tool for components running remotely. It can monitor and manage the state of other services. One of these services is the routing service, which allows you to route data traffic between different networks and different transports. It also allows you to apply transformations to the data. The Administration Console provides a dashboard of all applications and running services. One thing to note is that all of the services also use the distributed logger. Whenever any service or any application that uses the distributed logger logs a message, it will appear in this one single view. Administration Console gives you

Figure 6 Administration Console displaying the status from various hosts.

a summary of the number of applications and services that are having problems, and it will indicate the warnings and the errors by marking it with corresponding error or warning icons, so you can easily spot them and drill down into the detailed log messages to figure out the problems. To design a distributed system that is reliable, flexible and easy to maintain, the key is to start with the right architecture.

Systems unavoidably need to evolve at some time, and the flexibility in the underlying architecture allows you to easily add additional tools to help monitor and discover problems in your system. Real-Time Innovations Sunnyvale, CA. (408) 990-7400. [].

Professor Artilaâ&#x20AC;&#x2122;s Tips on Embedded Computing

w w Untitled-3 1


// +1 (714) 671-9996

Authorized Distributor


3/28/13 1:31 PM RTC MAGAZINE APRIL 2013

technology in

systems Windows Embedded

The Windows Embedded Legacy Continues with Windows Embedded 8 Standard The much anticipated release of Windows 8 has arrived and brought with it a lot of speculation over what features an embedded version might have. Windows Embedded 8 Standard (WE8S) builds on the Windows Embedded Standard 7 tools and Microsoftâ&#x20AC;&#x2122;s latest OS designed to reach beyond the desktop to mobile, touch-friendly devices. by John R. Malin and Sean D. Liming, Annabooks


n the past we would have been referring to Windows 8 as the latest desktop version of the popular Microsoft Windows operating system. However, this release of Windows has been architecturally designed not only for desktop and laptop PCs, but also for the needs of touch screen systems. Windows 8 has a whole new set of APIs, the WinRT APIs, that were designed to support the new Modern UI touch-centric graphic user interface used on the new Surface devices. Figure 1 shows the dual application support architecture. Windows 8 has also been ported to ARM processors to support mobile devices, but only Modern UI / Surface applications are supported by that port. The ARM version is called Windows RT. The Modern UI / Surface applications are available via the online store. The WinRT API is limited and the applications built with them run in an individual memory sandbox. Embedded developers would not likely create Modern UI applications because of the limitations. If you take away the Modern UI blocks in green from Figure 1, you will have the same desktop



operating system of previous versions. Consider current Modern UI to be the first generation of a new programming API. Improvements will be made over time. WE8S stands firmly on the shoulders of Windows NT Embedded, Windows XP Embedded and Windows Embedded Standard 7, leveraging the full functionality of the Windows 8 Pro desktop operating system and providing Lockdown Features that include the Embedded Enabling Features of WES 7 along with some new features and improved configuration settings to address the needs of embedded devices. WE8S, like its predecessors, leverages off-the-shelf hardware, Windows 8 applications and Windows 8 device drivers. Any application or driver that runs on Windows 8 can run on WE8S as long as you have the right features in the image. One of the biggest questions is about ARM support. Windows Phone 8 and Windows Embedded 8 Handheld are now running the Windows NT kernel on an ARM processor. Multicore ARM processors now have enough horsepower to run Windows RT on the new Surface tablet. With all the mobile devices running ARM

processors these days, it would seem obvious to include ARM support for WE8S; but Windows Embedded 8 Standard does not include support for ARM. Will there be support for ARM in the future? Microsoft is controlling the devices Windows RT will support. Considering the history of Windows CE, it might be that Microsoft wants better control of the non-x86 operating system.

Development Process and Tools

WES 7 made a big change in the development process from Windows XP Embedded. WE8S improves on the WES 7 development process, and with these improvements come some terminology changes to go along with yet another product name change. WES 7 answer files are configuration files in WE8S. Distribution share in WES 7 is now a catalog in WE8S. WES 7 packages are now called modules in WE8S. The modules are the biggest improvement since you can now create your own modules. The high-level development process stays the same. Two paths for OS installation are supported. One path provides a

tech in systems

quick method for deploying an operating system build to the target hardware, but it lacks a lot of capability to customize the features. This method uses the Image Build Wizard (IBW) DVDs. The second path provides a more advanced method with feature control, which uses Image Configuration Editor (ICE) to create a custom IBW disk. ICE allows you to select individual modules to customize the OS support in the image. ICE also allows you to automate the installation of the operating system. Custom drivers, applications and settings can be set up in a configuration file to build a custom IBW disk that installs everything on the target. The automation helps to remove any human error for projects that would have had to manually install and set up these items The modules are broken down differently than WES 7 packages. A straight conversion from a WES 7 answer file to a WE8S configuration file is not 1:1. Every WE8S image is built on the Embedded Core module, which has a starting size of 2 Gbyte for 32-bit support, thus images are bigger in WE8S. The Embedded Core is not broken down as much as it could be. Modern UI is the standard shell that is part of every image. The shell is bypassed with the new Shell Launcher module so you can launch custom applications. Since Module UI shell is in the image, .NET Framework 4.x is also in the image, thus the reason for the increased footprint.

Modern UI / Surface Apps

Desktop Apps



C, C++, C#, VB

Java Script

WinRT APIs Communications & Data

Graphics & Media

HTML JavaScript

C / C++ native


Internet Explorer



Devices & Printing

Application Model Windows Kernel Services HAL x86 / amd64

Figure 1 Windows 8 architecture.

Custom Modules

WES 7 moved away from the component concept of Windows XP Embedded to packages. The packages were signed CAB files, and the benefit was the ability to better patch service a system in the field. The biggest drawback was the lack of custom package support, so you could not use the same tools for patching OS updates to patch custom applications and drivers. To add custom applications and drivers in WES 7, a distribution share had to be set up, but the process was confusing. New to WE8S is Module Designer (Figure 2), which allows you to create custom modules for applications and drivers. Module Designer is a wizard that walks through the process to create a module. You can add the applications and driver files, set

Figure 2 Module Designer.

the file paths, add dependencies on other modules and create custom commands. Compared to Component Designer in XP Embedded, Module Designer is the simplest method to create custom selectable elements. Custom modules mean that you have a single solution to patch images in the field using DISM.

One Servicing Solution

Servicing an image is an important part of maintaining the lifecycle of the product. Servicing can take place online, with the image running, or offline using

a WIM file. Online servicing can be easily carried out using DISM and an update/ service configuration set created in ICE. Deployment Image Servicing and Management (DISM) was introduced in WES 7 and provides servicing support for the image, both in the factory and in the field. For WE8S, DISM has been updated to capture OS images and replaces ImageX. To get the latest OS patches and updates, Windows Embedded Developer Update (WEDU) is now supplied with the WE8S tools. The WEDU UI has been improved, and this new version supports RTC MAGAZINE APRIL 2013


Tech In Systems

Applications User-mode Kernel-mode

I/O Subsystem Filter Manager File System - NTFS

Metadata Bypass?

Overlay Cache Sector X

Unified Write Filter Volume Manager Partition Manager Disk Driver Sector 1

Sector 2

Sector X

Figure 3 Unified write filter.

updating WES 7 Distribution Shares, as well as WE8S Catalogs. If you want to create a custom recovery or patch media, you can build a custom Windows Pre-Installation Environment 4.0 (WinPE) disk. ICE allows you to build the base WinPE disk, and you can add features to the WinPE disk such as .NET Framework 4.x to create an advanced user interface.

Beyond the Desktop: Lockdown Old and New

WE8S introduces many changes in the area of Embedded Enabling Features starting with the name. Lockdown Features is the new name for the Embedded Enabling Features. The old write filters, Enhanced Write Filter (EWF), File Based Write Filter (FBWF) and the Registry Filter, are still available in WE8S, but they are being deprecated in favor of a new write filter solution called Unified Write Filter (UWF). UWF combines the best features of the previous filters into a single solution. To achieve the write-through capability of FBWF and still be configured for EWF, UWF is a sectorbased write filter that stores sectors in the overlay cache (Figure 3). The choice of overlay cache can be RAM or a special cache file that exists on the boot disk. The disk overlayâ&#x20AC;&#x2122;s advantage is to save on RAM by using disk space. The UWF cache file will be cleared on system reset just like RAM, so any cached writes will be lost on re-


Untitled-6 1


4/3/13 10:41 AM

boot, whether the cache was RAM or disk file. The UWFMGR.EXE utility is used to control the state of EWF and perform some administrative functions. Whereas for programming support, the EWF and FBWF came with custom APIs and had a header file and library file to support C++ programming, UWF takes a different approach with Windows Management Instrumentation (WMI) APIs for programming support. This means you can control the UWF with PowerShell scripts or any programming language that supports WMI APIs. In addition to the new write filter there are some other filters available. The keyboard filter that was introduced as a download for WES 7 SP1 is now a standard feature in WE8S, and it can now be configured in ICE instead of through the Group Policy Editor after the operating system is installed on the target. MessageBox default reply and the Dialog Box Filter have been combined into a single solution called the Dialog Filter that is more useful than previous versions. The Dialog Filter lets you identify dialogs, windows and processes that are to be blocked, given a specific response, or be exempted from filtering. Since the Modern UI shell (Surface Shell) supports touch screen gestures, there is a new gesture filter. The Gesture Filter is configured in ICE and allows you to select from a single gesture to filter to any combination of eight defined gestures.

tech in systems

Keil MDK-ARM MDK-ARM™ is the complete development environment for ARM® Cortex™-M series microcontrollers.

Figure 4 Embedded lockdown manager.

Finally, UWF, Keyboard Filter and Dialog Filter are being managed by a single tool called the Embedded Lockdown Manager (ELM), shown in Figure 4. ELM can connect to the local system or remote systems. ELM provides a graphical view of the current settings that you can change on the fly. You can also export a PowerShell script file for the current settings. The script file can be put into a module for preconfiguring the OS upon installation. Not everything was brought over from WES 7. The USB and SD card boot media options have been removed.

Standard vs. Industry

Also being introduced is the next release for Windows Embedded POSReady called Windows Embedded 8 Industry (WE8I ). The Industry version addresses the requirements for system integrators building point of sales systems. The lockdown features that are in WE8S are also available in WE8I, but the development process is completely different. WE8I doesn’t use build tools like ICE to automate the installation, but installs directly from a DVD like Windows desktop. Manual installation of applications and drivers is required. WE8I is intended for one shot installation for short lifecycle products, which POS system integrators need. There may be some confusion on WE8I being used for other industries, but when managing long lifecycle products

like medical, gaming and government system, WE8S is the preferred solution. Being able to automate the build process and remove human error is important to highly regulated industries and controls support costs over a longer time period.

Activation Required

All of the new features in WE8S make it a much better embedded solution than Windows Embedded Standard 7 (WES 7), but there is one very serious drawback: activation is required. This means that every image that you ship must connect to the Internet and register with Microsoft. Even if networking is not used in your product, the image must be activated. For embedded systems, activation is about as useful as a screen door on a submarine. The real concern is what happens during the lifecycle of the product. Embedded systems have long lives. With all the updates, there have been instances when Windows can get deactivated, like when there is a major hardware change. Failure to reactivate within a given time period results in a screen overlay dialog that reminds the user that activation is required. It doesn’t change the performance of the system. To some customers, this dialog might be a flag that there is a defect in the product and also an indication that Windows is in the system, when the OEM tried really hard to hide the fact that Windows is there. When a system is activated a unique signature is

Untitled-5 1

ARM DS-5 DS-5™ Professional is a full featured development solution for all ARM Powered® platforms.

+1 800 348 8051 © ARM Ltd. AD364 | 01.13



1/31/13 11:57 AM

Tech In Systems

sent to the system. Since activation has to happen on each system, any systems that require a CRC check like gaming and some medical devices will not be able to use WE8S. There was a threat to have activation in WES 7, but it was removed after many customers balked at the idea. This time activation is required, which can impact manufacturing costs and field upgrades. During the WES 7 attempt, there was

talk of BIOS signing or a license server that could be used as a workaround. With UEFI becoming the standard, something in the firmware might be possible, but the short-term solution looks to be a volume license implementation that is used for the desktop.

Additional Support

Windows is not a real-time operating system, but there are a couple solutions

from TenAsys that add real-time capability. The first is INtime, which adds a realtime kernel to run side-by-side with the Windows kernel. The INtime SDK lets you write real-time applications that have direct access to hardware and interface to Windows applications via semaphores and mailboxes. If you already have an investment in a real-time operating system, the second solution is TenAsysâ&#x20AC;&#x2122; eVM, which allows you to run in an embedded virtual machine. eVM limits the legacy impact with an efficient shared I/O layer. WE8S contains all the security features of Windows and includes the lockdown features for further security. There is a point of vulnerability that is not addressed: USB ports. Many viruses can come through different USB devices like flash drives, custom keyboards, digital pictures frames, etc. To control what gets connected to a system, Sofa King Software developed SecureBus, a USB filter solution that lets you set up a list of devices that can connect to the system. Unauthorized devices will be blocked and their drivers will not load. WE8S builds on the WES 7 development process to add new capabilities to best service systems in the field. If activation was not an issue and there was better module breakdown of the Embedded Core, WE8S would be the most complete embedded solution based on the Windows desktop ever developed. One thing has not changed. Microsoft still uses the shared success model to differentiate from other operating systems. Since WE8S is Windows 8 broken down into modules, you can build and test your applications and driver on the Windows 8 desktop before ever making the investment in WE8S. Annabooks Irvine, CA. (714) 970-7523. []. TenAsys Beaverton, OR. (503) 748-4720. []. Sofa King Software [].


Untitled-3 1


3/26/13 12:35 PM

technology deployed Development Tools for ARM

ARM Architecture Offers Challenges Along with Features to Help Meet Them The ARM processor architecture offers a number of hardware features, which, in conjunction with tools that are able to take advantage of them, can greatly enhance understanding, bug avoidance and bug fixing during the development process. by Shawn A. Prestridge, IAR Systems


mbedded developers who create systems using the ARM architecture face challenges similar to developers who use other architectures. Some of these challenges involve initializing the device and its peripherals while others involve the ability to write to peripherals that are attached to the microcontroller core. Moreover, debugging software written for any device can be a daunting affair if the application is complex. Fortunately, ARM Ltd. has worked diligently to foster an ecosystem of partners and services that mitigate many of the difficulties in developing embedded software, both on the software development and the debugging aspects of product development. This ecosystem helps a developer write an application quickly and efficiently, all while minimizing the potential for defects in the code. Semiconductor companies generally have different ways to initialize and utilize parts of the core, such as interrupts and condition registers. Because of this, it can be difficult for an embedded developer to learn how to use these parts of the



microcontroller effectively and correctly, especially if their organization is using parts from several different silicon vendors. ARM has a unique solution to this problem via their Cortex Microcontroller Software Interface Standard (CMSIS) interface. CMSIS defines an API that an embedded developer can use to set up timers, ITM registers, the Nested Vectored Interrupt Controller (NVIC) and many other things. This has a two-fold benefit for the developer. First, by using this API, a developer can quickly set up and access these features of an ARM Cortex core. Second, a project can be moved to a different ARM Cortex part by simply switching the target in the project settings of IAR Embedded Workbench for ARM and recompiling the code. While some register names and the startup code may need to be changed as well, this should be a trivial undertaking. The ability to do this is particularly important when an organization decides to do a follow-on project that expands the capabilities of the previous application, because CMSIS makes it easier to move

to a more powerful part. By reducing the learning curve of how to use these features of the microcontroller, CMSIS helps to shorten time-to-market for a project. While CMSIS makes it easy to set up the device and use some of its functionality, it is generally not sufficient in and of itself to help an embedded developer write an application. The fastest way for a developer to get started with their application is to start with a well-crafted and non-trivial example project. This is why the IAR Embedded Workbench for ARM features examples for many popular evaluation and development boards, almost all of which are written by the application engineers of the semiconductor companies whose parts are on the boards. A developer can choose the example application that most closely resembles their project and use that as the basis for their code. By doing so, the developer can be assured that all of the tool options are set up correctly and that they are initializing and using the peripherals on the microcontroller in the manner prescribed by the semiconductor company. Because many companies produce ARM parts, the competition to sell microcontrollers is fierce, and application engineers work hard to produce easy-touse applications that the developer can employ wholesale in their project. Examples of this include USB bootloaders, web servers, audio recording/playback and others. Oftentimes these examples are a large percentage of what the developerâ&#x20AC;&#x2122;s project needs to do, which can drastically reduce time-to-market. Itâ&#x20AC;&#x2122;s also possible to cobble together two or more example applications into a single application that checks more boxes off the specification list for a project. For more complex projects, a realtime operating system (RTOS) becomes an invaluable asset to a developer. In such projects, the application needs to perform many different tasks in the system on a set schedule. As the number of tasks increases, so does the complexity of the scheduling algorithm needed to ensure that all of the tasks complete on time. Additionally, more tasks create a more complicated timeline of execution for the code and make it difficult to figure out how interrupts should be serviced. An RTOS

Technology deployed

can make these headaches disappear by making code deterministic in terms of being able to control the scheduling of different tasks. Once individual tasks have been written for an RTOS, the developer only needs to play with the relative priorities of the tasks to make sure that each one is processed in a timely and efficient manner. Moreover, an RTOS has the advantage of placing the microcontroller in a sleep mode when it has nothing to do, which is much more power-efficient than traditional â&#x20AC;&#x153;superloopâ&#x20AC;? approaches that either try to do all the tasks in one big loop, or sit in a tight polling loop waiting for an event to occur. IAR Embedded Workbench for ARM does kernel-aware debugging for many popular RTOSs so that a developer can see the states of tasks, events, queues, semaphores, etc. of their RTOS-based code in dedicated debugger windows. Additionally, these RTOSs often integrate with other middleware components such as GUI packages and stacks (TCP/IP, USB Host, USB Device, USB OTG, Bluetooth, etc.) to increase the functionality of the underlying system in a straightforward manner that developers can incorporate quickly into their designs. These additional packages in some cases come with middleware-aware debugging from IAR Systems C-SPY software debugger that is included with IAR Embedded Workbench for ARM. Having this information at-aglance makes it easy to see when things go awry in the code and thus speed up the debugging process. Bugs are inevitable in code, but the earlier in the development process they can be found, the cheaper (and generally easier) they are to fix. This is just one of the reasons that many companies are adopting coding standards to help ward off some of the more common sources of defects in applications. One such standard is MISRA C, which stands for the Motor Industry Software Reliability Association (See Checking Rules for C: Assuring Reliability and Safety, RTC, March 2013). This standard was originally targeted to improve the safety and reliability of systems that were to be put into automobiles, but other industries are finding it useful because it forces developers to write code in a safer and more consistent man-

BAD example if ((x == y) || (*p++ == z)) { /* do something */ }

GOOD example int temp = *p++ == z; if ((x == y) || (temp)) { /* do something */ } Figure 1 MISRA C helps a developer catch subtle defects at compile-time so that they are found before testing begins.




4 watchpoints

6 breakpoints

Live core access

PC Sampler ETM Trigger Interrupt trace CPU statistics



Software trace 32 channels


Time stamping

Trace port

ETM Instruction trace Figure 2 The ARM CoreSight debug module provides access to a wealth of information about the state of the microcontroller during application execution.

ner. More importantly, its 140+ rules can help developers find bugs at compile time rather than out in the field. One example is Rule 33, which states that the righthand side of an && or || operation cannot have side effects (Figure 1). The problem is that if the left-hand side is true (in the case of an OR operation), or false (in the

case of an AND operation), then the righthand side is never evaluated so the sideeffect is never performed. This is a rather subtle defect that might take quite a long time to find in testing or in the field, but MISRA C allows a developer to find it at compile time. IAR Embedded Workbench for ARM has a built-in MISRA C checker RTC MAGAZINE APRIL 2013


technology deployed

Figure 3 Power debugging enables developers to see the power consumption of different functions so that they know where to focus their optimization efforts.

that can be used to examine the application code for MISRA C compliance. Additionally, tools such as Gimpel’s Lint can be used to do static-checking of the source code for potential problem areas. Many companies spend most of their development cycles in the test-and-fix phase. While using well-tested code such


ISCP_Ad.indd 1


as CMSIS and RTOSs can stave off bugs, the fact remains that there will be defects in the code when the test-and-fix phase begins, and that phase will still be a significant portion of the project’s schedule. It is therefore imperative to have good debugging tools to quickly diagnose, find and correct code defects. ARM has a de-

bugging interface called CoreSight, which allows a developer unparalleled insight into what is happening in the core as it executes application code. Figure 2 shows a high-level view of the Cortex-M CoreSight debug module. Each part of the CoreSight interface gives the C-SPY Debugger the ability to monitor different aspects of the microcontroller’s execution. Some of the most commonly used features include the embedded trace macrocell (ETM), which is a live instruction trace of the code. This enables developers to see not only where they are in the code execution, but also the path they took to arrive at that point. The instrumentation trace macrocell (ITM) provides the ability to pass data such as variable values, the stack pointer or condition codes from the application to the C-SPY debugger in real time. Each ITM and ETM packet is time-stamped so that the developer can measure how long code execution takes with a fairly high degree of accuracy (in the case of ETM), and when data was sent from the application to CSPY (in the case of ITM).

4/3/13 3:23 PM

Technology deployed Additionally, ETM gives a developer the ability to do code coverage and function profiling, the latter of which helps you to understand where the application is spending most of its time. Both of these provide quite a bit of information to the developer so that bugs can be quickly spotted and the code optimized for the highest possible performance. There are many debugger probes that use the features of CoreSight to allow the developer to quickly find and fix problems in their application. However, there are some probes such as IAR Systemâ&#x20AC;&#x2122;s I-jet that leverage the debug interfaces to provide more information about the system. I-jet allows a developer to do Power Debugging wherein the PC is sampled periodically and the resulting ITM packet is tagged with the value of the instantaneous power being supplied to the board via Ijet. This allows the developer to see how much instantaneous power is being consumed by the application and tie that information to the source code so that areas where the power spikes can be examined and possibly mitigated. Figure 3 shows how this is displayed in IAR Embedded

meds1301_rtec_ad.indd 2

Workbench for ARM. More importantly, this technology allows the developer to create a power profile for their application wherein they can see the minimum, maximum and average power consumption. This profile can help the developer understand how long a particular battery could power the application, or try out different hardware settings or coding algorithms to see if they can lower the overall power profile. Rather than guessing that a certain change would lower the power consumption, the developer can know if it does and more importantly, quantify by how much. The ARM architecture is perhaps the most popular embedded platform used today because of its flexibility in the peripherals it supports as well as the number of semiconductor companies that produce ARM cores. These companies produce sundry mixes of peripherals so that developers can find just the right core for their needs (and perhaps their future needs). The peripherals can be controlled via the CMSIS interface, which provides a simple, easy-to-use interface for developers and abstracts the implementation details so that developers can switch from one

ARM part to another in a relatively painless fashion. ARM also fosters an ecosystem with RTOSs and other middleware components to help developers create complex systems quickly. System defects can also be avoided by including coding standards such as MISRA C or Lint to spot potential problems at compile time. Debugging is also easier with an ARM core because of the CoreSight debug unit that allows developers access to key components of the microcontroller so that they can follow closely the execution of their application to quickly locate and remove bugs. The data from the debug unit can also be utilized in new ways to provide additional information to the developer such as the power profile of the application. All of these components work in concert with one another to make the entire development cycle of an ARM-based design as quick and defectfree as possible. IAR Systems Uppsala, Sweden. +46 18 16 78 00. [].


1/8/13 6:58 PM RTC MAGAZINE APRIL 2013

products &

TECHNOLOGY COM Express Type 6 Module with Three High-Performance Graphics Display Interfaces

Dual Channel PC/104 CAN Module Provides 1000V Isolation

A high-performance COM.0 R2.0 Type 6 module features an Intel Core i7/i5/i3 processor supporting Intel HD Graphics integrated on the CPU with three independent displays. A PCI Express x16 Generation 3.0 bus is available for discrete graphics expansion or general purpose PCIe (optionally configured as 2 x8 or 1 x8 + 2 x4). The Express-IB from Adlink Technology targets applications in government, military, medical, digital signage and communications and is attractive for customers with advanced processing performance and graphics requirements looking to reduce development time by outsourcing the base design of their system and focusing on application functionality. The Express-IB supports Intel Advance Vector Extensions (Intel AVX v1.0), with its improved Floating Point Intensive Applications, and also offers the benefits of increased bandwidth provided by USB 3.0. New with the COM Express Type 6 module are three Digital Display Interface (DDI) ports supporting HDMI, DVI and DisplayPort outputs, in addition to legacy VGA and dualchannel 18/24-bit LVDS displays. The Express-IB also offers Gigabit Ethernet, up to four USB 3.0 ports, four USB 2.0 ports, two SATA 6 Gbit/s ports and two SATA 3 Gbit/s ports (RAID 0/1/5/10) and support for SMBus and I2C. The module is equipped with AMI EFI BIOS supporting remote console, CMOS backup, hardware monitor and watchdog timer. The Express-IB features the Intel Core i7/i5/i3 processor supporting Intel Hyper-threading Technology (4 cores, 8 threads) and up to 16 Gbyte of DDR3 dual-channel memory at 1333/1600 MHz on dual stacked SODIMM sockets to provide excellent overall performance. Intel Flexible Display Interface and Direct Media Interface provide high-speed connectivity to the Mobile Intel QM77 Express Chipset.

A PC/104-compliant, dual channel, isolated Controller Area Network (CAN) module with Windows and Linux drivers employs high-speed isolated data couplers and power supplies to provide 1000V protection between the two NXP SJA1000 CAN controllers and the network interface. This makes the PCMCAN-2-ISO from WinSystems suitable for operation in high-voltage renewable energy, high-speed industrial control, or unpredictable automotive applications, while it maintains low EMI and low-latency operation. Each CAN channel can provide isolated +5 VDC power or receive isolated +5 to +12 VDC power from the interface for additional flexibility. The +5 VDC power supply includes overvoltage, overcurrent and short-circuit protection. The PCM-CAN-2-ISO is compliant with CAN specifications 2.0A (11-bit ID) and 2.0B (29-bit ID). CAN is an industry-standard, serial, asynchronous, multi-master communication protocol for connecting electronic control modules, sensors and actuators in automotive and industrial applications. The PCM-CAN2-ISO supports transfer rates to 1 Megabit per second and has jumper selectable termination resistors so it can be used in various topology systems. WinSystems offers this board in other offthe-shelf configurations. The PCM-CAN-2 is a dual channel, non-isolated unit. The PCMCAN-1 is a single channel, non-isolated unit. The PCM-CAN-1-ISO is a single channel, isolated unit. Special OEM configurations are possible. All configurations will operate over the industrial temperature range of -40째 to +85째C The isolated dual channel PCM-CAN2-ISO lists for $269, and the non-isolated dual channel PCM-CAN-2 lists for $229.

ADLINK Technology, San Jose, CA. (408) 360-0200. [].

EN50155-Compliant DC-DC Converters for Railway Applications A new family of board mount DC-DC converters has been developed for railway applications. The VQB100R and VHB150R series modules from CUI are designed to comply with the EN50155 standard, which specifies input, EMC, mechanical and environmental requirements. The internally potted and encapsulated design provides increased system reliability through added protection from environmental factors such as dust, moisture, shock and vibration. Offered in industry standard quarter brick and half brick footprints respectively, the 100W VQB100R series and the 150W VHB150R series are highly efficient, reaching levels up to 92.5%. Though the modules are designed primarily for railway applications, the VQB100R and VHB100R series also target designs that may experience high transients, including telecom systems and battery-powered equipment. The DC-DC converters provide a 3:1 input range of 66~160 VDC and output voltage options of 5, 12, or 24 VDC. They are designed to deliver a case operating temperature range of -40째 to 100째C and provide 2250 VDC I/O isolation. Protections for over voltage, over current, short circuit, under voltage lockout and over temperature are included. Additionally, all models offer remote on/off control, remote sense, and carry CE and UL/cUL 60950-1 safety certifications. The VQB100R and VHB150R series are priced starting at $119 in quantity 100. CUI, Tualatin, OR. (503) 612-2300. [].



WinSystems, Arlington, TX. (817) 274-7553. [].


Power Supply Meets Increasing Need for Down-Hole Power in Oil and Gas Exploration Ultra-reliable power supplies enable companies to power AC equipment up to five miles “down-hole” in even the most adverse field conditions. The Model BL1500 power supply from Behlman Electronics provides the highest levels of support for oil exploration, drilling, evaluation completion and intervention. Although the Behlman BL1500 has been helping oil-industry companies succeed for seventeen years, it has been continuously improved to keep up with the leading edge of exploration technology. As a result, no matter how old or new the exploration system, BL1500 is the appropriate choice for delivering ultra-reliable down-hole power. The Behlman BL1500 unit is powered from 115 VAC, 47-63 Hz and provides two ranges of high-voltage AC at 60 Hz. Units can be stacked for increased power. Protective circuits include input, short circuit, constant current and thermal protection. BL1500 power supplies have an RS-485 interface, allowing the unit to be controlled and monitored remotely from a central station. It operates from 32 to 131°F (0 to 55°C), and has a high-strength 19-inch rack mount chassis 3.5” high x 21.5” deep (48.25 cm wide x 7.87 cm high x 54.61 deep). Behlman Electronics, Hauppauge, NY. (631) 435-0410. [].

Portable Data Acquisition System for Rugged, Remote Environments

21.5-Inch High Definition Rugged Video Mission Display

A highly accurate transient recorder technology is now available in a portable, ruggedized system for in-the-field measurements. Available with up to 24 channels and sample rates up to 240 MS/s, the flexible TraNET PPC from Elsys Instruments enables precise data acquisition from multiple points simultaneously. The new system combines Elsys’ high-speed, LAN-controlled instruments with a robust industrial PC for flexible data acquisition in a number of harsh and mission-critical environments, including power plant maintenance, electric traction engine testing, ballistics and explosive testing. The LAN connectivity enables reliable, standalone operation in remote applications as well, such as structural health verification on bridges and buildings, seismic activity monitoring or stray voltage detection from defective power lines. Channel configurations of 4, 8 and 12 are available in addition to the 24, depending on application needs. Each channel provides up to 128 Mbyte of acquisition memory. Different Elsys PCI/PCIe-compatible TPCX and TPCE digitizers comprise the heart of the TraNET PPC, depending on specific application needs. The modules offer a typical measurement precision of ±0.03% with transfer speeds of up to 2.5 Gbyte/s on the PCIe-compatible modules. These digitizers offer many features such as single-ended and differential inputs, large input voltage and offset ranges, advanced trigger capabilities with an external trigger, programmable anti-aliasing filters as well as ICP input for piezo sensors and digital inputs. The new TraNET PPC offers several different recording modes to address specific data requirements. Scope mode, similar to an oscilloscope, enables quick configuration of acquisition parameters for simply visualizing and analyzing single-shot events. Continuous mode acquires and writes data onto the internal hard disk at a high speed with no loss of data or dead times between events. Block mode divides the available onboard data acquisition memory into absolute or relative time-stamped blocks and is especially useful for capturing signal bursts where only relevant data needs to be recorded. The most sophisticated mode, Event Controlled Recording (ECR), provides “intelligent streaming” for troubleshooting and long-term monitoring applications. ECR uses smart trigger logic that allows adjacent channels to overlap, ensuring all relevant data is captured and eliminating dead time between triggers. In addition, every channel can acquire waveform data independently on trigger command as well as synchronously with associated channels. Pricing for a TraNET PPC starts at $13,500.

A rugged mission display for airborne platforms comes as a next generation video display with touchscreen and is designed for the most demanding helicopter and fixed wing applications. Featuring a wide array of digital and analog inputs, the AVDU5500 from Curtiss-Wright Controls Defense Systems easily connects to the market’s leading electro-optical turrets, either directly or via any of Curtiss-Wright’s Skyquest VMS video distribution units. The AVDU5500 delivers full HD 1080p resolution and provides market-leading features and performance designed specifically for the demands of defense and law enforcement surveillance, search and rescue applications and the extended environmental characteristics demanded by airborne platforms. The AVDU5500 combines an array of video viewing options and visibility features. For example, the popular built-in “quad screen” option enables operators to view up to four independent live video images simultaneously from any of the multiple video sources sent to the display. The AVDU5500 uses advanced optical bonding techniques to ensure maximum visibility in bright sunshine conditions and improved ruggedness. As standard, the AVDU5500 utilizes a dual LED backlight for Night Vision Goggle (NVG) filtering purposes. When required, the display can be switched into NVG mode, which switches off the standard white backlight and turns on the NVG filtered backlighting, conforming to MIL-STD-3009 NVIS B. This gives the user full color, high brightness imagery usable with direct sunlight during daylight operations, as well as perfectly filtered imagery for NVG operations at night. All other display bezel lighting is also NVG filtered as standard. The AVDU5500 is designed to provide airborne operators with the greatest amount of flexibility and control. The display’s 5-wire resistive touchscreen can be customized to operate with any of today’s leading digital moving maps. A wide array of I/O options, including USB, Ethernet and RS-422/232, enable the AVDU5500 to interface with peripheral equipment. Its eight “hard” bezel keys located at the bottom of the display, control core display functions such as power on/off, brightness and channel selection. With a single keystroke, user-definable, touchscreen-operated “soft” keys can be displayed around the perimeter of the display overlaid on the video. The functions controlled by these soft keys can be defined at the factory and include a range of standard options including Skyquest video recorder control, Picture-in-Picture selection and Quad View.

Elsys Instruments, Monroe, NY. (845) 238-3933. [].

Curtiss-Wright Controls Defense Solutions, Ashburn, VA. (613) 254-5112. []. RTC MAGAZINE APRIL 2013



2U RAID Array Is NEBS Certified for Telecom Applications A new 2U RAID array has been certified to Network Equipment Building System (NEBS) Level 3. The 12-drive array from One Stop Systems supports up to 48 Tbyte data storage using twelve 4 Tbyte SATA drives. It connects to the host server with either PCIe x8 or SASx4 connectivity. The chassis includes dual redundant 500-watt power supplies, two removable blowers for superior cooling, and a removable NEBS filter and filter cover. The RAID array boasts 2700 Mbyte/s data transfers from server to storage on a single PCIe x8 connection. A single SASx4 connection to the server provides 1900 Mbyte/s data transfers. PCIe is suitable for storage applications that require extremely fast read and write transfers. Because there is no software conversion from PCIe on the motherboard to another protocol, latency is reduced, providing extremely fast data transfers. Two SASx4 inputs cable from one or two servers to the two SAS connectors on the rear of the RAID array. Both can be input connections from two servers or one can be an input and the other an output to connect another RAID array, thereby doubling the storage capacity and increasing the performance. NEBS is the most common set of safety, spatial and environmental design guidelines applied to telecommunications equipment in the United States. Although not a legal requirement, it is an industry requirement. NEBS Level 3 has strict specifications for fire suppression, thermal margin testing, vibration resistance (earthquakes), airflow patterns, acoustic limits, failover and partial operational requirements (such as chassis fan failures), failure severity levels, RF emissions and tolerances, and testing/certification requirements. This makes the 2U RAID array ideal for data center applications. While NEBS certification is only required for telecommunications central office equipment, certification means that compute systems that pass the strict requirements are held to higher standards. The 2U RAID array lists for $5,899 and is available immediately

Visualization Tool Enhancements Simplify Defect Detection New enhancements to the CodeSonar software architecture visualization tool from Grammatech include a new tree map view designed to allow users to easily see the hierarchical structure of the code in a very information-dense form. The view uses colorization to show the density of defects in modules so users can easily identify the most problematic parts of the code. The call graph is organized by module structure. Users can drill down to see a greater level of detail, choose different layouts such as treemap, circuit, cluster, flow, radial and other layouts, and attach persistent notes to the diagram. Transitions such as zooming or layout changes are fluid and real-time. With CodeSonar visualization, users can start at individual functions to gain insight

One Stop Systems, Escondido, CA. (877) 438-2724. [].

Avionics Module Saves Space, Weight, Power and Cost A multi-protocol embeddable avionics module is specifically designed to save valuable space, weight and power as well as deliver greater cost-effectiveness and higher reliability in avionics labs, simulators and embedded applications. Featuring both MIL-STD-1553 and ARINC 429 protocols on a single XMC form factor board, the RAR15-XMC from GE Intelligent Platforms is used in a broad range of avionics applications. Featuring advanced API software for Windows 7, Vista, XP (32- and 64-bit), Linux, Integrity and VxWorks that reduces application development time, standard features of the RAR15XMC include 8 Mbytes of RAM and 64-bit message time tagging. Also featured are extensive BC and RT link-list structures, error injection/detection, automatic/manual RT status bit and mode code responses, along with advanced BC functionality. The RAR15-XMC bus monitors provide superior error detection and 100% monitoring of fully loaded buses. Four dual-redundant MIL-STD-1553A/B Notice II channels, ten ARINC 429 receive channels and eight ARINC 429 transmit channels are provided by the RAR15-XMC. Onboard firmware and large data buffers, together with the advanced API, contribute to a very high level of flexibility in monitoring and generating ARINC bus traffic. Simultaneous Scheduled and Burst Mode (FIFO) messaging is supported on all ARINC 429 transmit channels. Each ARINC 429 receive channel provides simultaneous dedicated and buffered mode storage, along with label/SDI filtering. Conduction-cooled and conformal coating are standard on the RAR15-XMC, and these, together with optional extended temperature capability, enable it to be deployed in the harsh environments typical of avionics applications. General Electric Intelligent Platforms, Huntsville, AL. (780) 401-7700. [].



from a bottom-up perspective, annotate nodes and edges with additional information, and overlay the visualization with information on defects and source-code metrics. CodeSonar visualization also includes other programunderstanding and navigation features, and supports sharing of diagrams between team members. Other features allow users to search the graph for functions of interest and navigate to and from the source code of the program. In addition, the architecture is extensible, so users can generate graphs from other sources, feed them in an XML format to the tool, and see them in the same user interface. CodeSonar visualization works in standard web clients. CodeSonar has long been the softwareanalysis tool of choice for embedded developers working on mission-critical applications such as satellites, avionics, industrial controls and medical devices. Companies outside the safety-critical space use CodeSonar to improve software reliability and security. This includes organizations developing software for wireless devices, networking equipment and consumer electronics. Grammatech, Ithaca, NY. (607) 273-7340. [].


Fanless Medical Computer Pushes Mobile Functionality The first of a new TOPAZ line of medical computers from Adlink Technology features rich I/O connectivity, wireless support and duty-specific performance processors with long life support. The TPZ-1300 from Adlink Technology delivers ruggedized reliability, making it suitable for mobile functionality in environments such as telemedicine, nursing information systems (NIS) and medical administration. The Adlink TPZ-1300’s customization for healthcare IT operations provides highly available functionality, with MTBF exceeding 250,000 hours for 24/7 operation. A fanless and cable-free structure ensures extended durability for long-term usage, easy-to-clean housing and noiseless operation to satisfy stringent medical application requirements. An anti-bacterial coating option is also available. The TPZ-1300 provides robust construction, withstanding up to 5G vibration and 100G shock, easily meeting the needs of telemedicine applications such as telecardiology, telepathology and nursing workstation use. The TPZ-1300 is equipped with the Atom D2550/ N2600 processor and also provides abundant I/O function from six USB ports, three GbE ports, four COM ports, one mini PCI-E slot and four digital inputs and outputs to optimize communication and control between multiple devices. The TPZ-1300 also supports multiple OS, including Windows 7, Windows 7 Embedded, Windows XP, Windows XP Embedded and WinCE 7.0. Adlink Technologies, San Jose, CA. (408) 360-0200. [].

Low-Power Development Tools Come with Supercapacitor-Charged Demo Board

3U OpenVPX Low-Latency JPEG 2000 Codec Module Offers Flexible Configuration

Today’s highly integrated, high-performance and low-power processors demand accompanying tools that can enable configuration of the underlying silicon components to meet specialized application demands including low power consumption. Silicon Labs offers a diverse portfolio of Precision32 MCUs based on the ARM Cortex-M3 processor. This portfolio is supported by a set of development tools that enable designers to optimize their designs for the lowest power consumption without compromising performance. The company’s complimentary Eclipse-based IDE and AppBuilder software for Precision32 MCUs includes tools that enable developers to estimate power consumption and receive configuration guidance to minimize system power. The Power Estimator tool gives developers a top-level graphical view of how a Precision32 MCU uses power in active and sleep mode. The tool enables developers to adjust power usage at the onset of a project even without having development hardware. Power Estimator automatically updates the system design with configuration changes, allowing developers to optimize each mode for the lowest power. A companion tool, Power Tips, provides software configuration guidance that helps developers minimize current consumption. Power Tips automatically appears within AppBuilder when the cursor hovers over a configurable setting. This simple ability to see power optimization tips while configuring the MCU saves considerable development time. In addition to the tools, Silicon Labs supplies a Low-Power SiM3L1xx development board: Roughly the same size as an ID badge, this compact development board showcases the power efficiency of SiM3L1xx MCUs. The board contains an ultra-low-power SiM3L1xx MCU, segmented LCD, supercapacitor, LED and photodiode sensor, debug interface and USB port. The board can display information continuously on the LCD for up to three days after a quick 90-second charge of the supercapacitor through a USB cable. Using the USB connector and debug interface, developers can connect the board to a PC and use the Precision32 IDE for software development. The board offers a “demo mode” that shows typical power consumption for various MCU operating modes on the segmented LCD. Developers can also download an iPhone app that lets them enter text such as names and phone numbers and then send the information to the board using specialized image patterns on the iPhone’s LCD that are received by the board’s photodiode sensor.

A 3U OpenVPX JPEG 2000 Codec module comes in a compact, rugged form factor and is powered by a Xilinx Kintex-7 FPGA and two JPEG 2000 compression engines. Flexible video configuration options (bitrate, frame rate, etc.) allow the user to optimize the VCP-2864 from Creative Electronic Systems as needed. The VCP-2864 can be combined with the VCP-8166 H.264 / AVC compression / decompression XMC module to form a single-slot solution for JPEG 2000 and H.264 compression and decompression, as well as raw capture for HD video and still images. The VCP-2864 features multiple SDI input channels compatible with SD and HD signals. The video coding functionality provided by the VCP2864 is designed to guarantee smooth real-time, low latency coding up to HD formats. JPEG 2000 compressed video is available from the processor board via PCI Express (one channel if HD, two channels if SD). Input signals can be duplicated and non-compressed video is available via the PCI Express interface for real-time processing. Scaling down of frame rate, resolution, bitrate and frame cropping are programmable. The board can be configured either in capture / compression or in decompression / output mode. An XMC site is available for expansion with the VCP-8166 H.264 / AVC Codec XMC. The RTM-6487A0 Rear I/O Transition Module for Video Boards (3x micro HDMI, 8x SMB, 2x VGA, 1x CameraLink, 2x mini Stereo Jack, 1x RJ45, 1x mini USB) provides I/O options for the VCP-2864. A standard element of the CES OpenVPX family, the VCP-2864 is compatible with the other 3U VPX boards from CES, including the ETS-8227 multi-protocol switch, the RIOV-2440 single board computer and the FIOV-2310 FPGA processor board. An Advanced Board Management Controller (aBMC) is implemented for VITA 46.11 support, configuration management, event logging and other supporting tasks. It is fully compatible with the CES Configuration, Load and Monitor (CLM) tool. The module has options for air-cooled and conductioncooled operating environments. Software support packages are available for Linux and VxWorks operating systems. A video API library is provided. The video API and board libraries provide the necessary interface for the configuration and control of the board functionality. The VCP-2864 and VCP-8166 combination is unified under a single interface.

Silicon Labs, Austin, TX. (512) 416-8500. []

Creative Electronic Systems, Geneva, Switzerland. +41.22.884.51.00. [].




Gang Tester Provides Significant Increase in Throughput A new TAP Transceiver for the test and programming of up to 16 electronic assemblies in parallel comes with an integrated system power supply that helps to increase the throughput for UUT (unit under test) with a single TAP (test access port) by factor 16. This results in significantly improved production efficiency and reduced investment costs. Thanks to the TAP interfaces’ programmability (voltage, delay, protocol, etc.) and the TAP-to-UUT assignments’ individual configurability, the system can be flexibly customized to fit nearly any environment and application. The SFX-TAP16/G-RM from Goepel Electronic is available as rackmount version in 19’’ technology (1U) and provides 16 parallel TAPs. Its architecture supports all modern Embedded System Access (ESA) technologies, including boundary scan, processor emulation and chip-embedded instrumentation, enabling the basic paradigm change for test and programming methods without mechanical probe access (non-intrusive). This new Scanflex module was developed specifically for Gang applications, whereby various test and programming strategies can be freely combined. The new transceiver can be flexibly adapted to the UUT characteristics; its bandwidth reaches from concurrent test/programming of 16 UUTs with a single TAP to two UUT with eight TAP each. In addition, the TAPs can be individually programmed in many parameters. Goepel Electronic, Jena, Germany. +49-3641-6896-739. [].

Third Generation Intel Core i7 combines with Serial RapidIO on AMC boards Concurrent Technologies released and demonstrated the latest additions to their MicroTCA product family—the latest AMC processor boards featuring either dual or quad core third generation Intel Core i7 processors based on 22nm process technology combined with IDT’s Tsi721 PCIe to RapidIO bridge to provide a highly efficient interface to a Serial RapidIO (SRIO) backplane. Additionally, support for SRIO has been added to the previously released Fabric Interconnect Networking Software (FIN-S), which provides standardized uniform APIs across a fabric interconnect. The result is a high-bandwidth and low-latency solution that utilizes about 5% of the CPU’s bandwidth and provides a realistic opportunity for users of PowerPC/SRIO products to transfer to Intel processor technology. Applications for this hardware and software solution are expected within both the telecom and defense markets, especially where a large number of computing nodes are required to intercommunicate at high data rates. Depending upon I/O requirements, system designers have a choice of two high-performance processor boards—the AM 93x/x1x or the AM 94x/x1x. Users can take advantage of the boards’ processing power, high-performance backplane fabric connectivity and the peer-topeer networking performance of SRIO to develop large multiprocessing systems. These processor boards are particularly well suited for MicroTCA-based telecommunication applications such as IPTV, digital media servers, media gateways, broadband, Long Term Evolution (LTE) or LTE-Advanced, wireless base stations as well as in test systems for wireline and wireless networks. All variants of AM 93x/x1x and AM 94x/x1x support Fabric Interconnect Networking Software (FIN-S), which provides high-performance Ethernet emulation over SRIO along with other application level interfaces. FIN-S enables plug and play of any TCP/IP-based application across the SRIO fabric, thereby reducing the development effort and time-to-market for the user by providing a very high degree of application portability and reusability. The main features and benefits of using FIN-S on SRIO include more than 60% practical application level bandwidth when compared to 10 GigE as well as more than double the performance with half the CPU utilization compared to open source SRIO Ethernet emulation solutions. TCP/IP Application level bandwidth of more than 1.8 Gbyte/s is achieved with Gen 2 SRIO 5 Gbaud/s x4 links. This solution also boasts “jitterless” TCP/IP packet latencies of less than 10 microseconds. Concurrent Technologies, Woburn, MA. (781) 933-5900. [].



New Low-Cost Boundary Scan Bundle for Technology Entry A special package is now targeted for beginners in JTAG/Boundary Scan or users with cost-sensitive projects. PicoTAP Designer Studio from Goepel Electronic is a complete boundary scan test system including hardware and software and offering an extremely reasonable price-performance-ratio. In addition to a Mixed Signal I/O module, the bundle contains the world’s smallest boundary scan controller, the PicoTAP, which is powered via USB and can be plugged directly into the I/O module. The hardware/software bundle contains a PicoTAP controller, a CION Module FXT-96/A and a SYSTEM CASCON Basic/ SX Development Station. Because of the included CION I/O module, analog and digital peripheral ports can also be tested. Additionally, relays and Opto I/O are available to flexibly optimize test coverage. Two package versions are available so that users may choose their most beneficial version. The bundle’s performance level can be extended from pure test (base version) to additional applications such as in-system programming of small Flash and PLD or memory and cluster test. Numerous hardware and software options are additionally available so that the test unit’s performance may grow with the application and investments are protected. Boundary Scan (IEEE Std. 1149.x) is an access method for the test and programming of complex circuits without mechanical probe access (non-intrusive). Boundary Scan is part of the Embedded System Access (ESA) strategies and is based on design-integrated test electronics. ESA technologies include techniques such as chip embedded instruments, processor emulation test, in-system programming or core assisted programming. They are currently the most modern strategies for validation, test and debug as well as programming of complex boards and systems. They can be applied throughout the entire product life cycle, enabling enhanced test coverage at reduced costs. Goepel Electronic, Jena, Germany. +49 3641 6896 739. [].


Ethernet “Anything I/O FPGA Card” A general purpose FPGA-based programmable industrial I/O card provides a 100 BaseT Ethernet host interface. The MESA 7I80HD from Mesa Electronics is a low cost, general purpose, FPGA-based programmable I/O card with 100 BaseT Ethernet host connection. The 7I80HD uses 50-pin I/O connectors with interleaved grounds and I/O module rack compatible pinouts. The 7I80 is compatible with all of Mesa’s 50-pin daughter cards. Open source FPGA firmware configurations are provided for hardware step/dir generation to 25 MHz, PWM generation, analog servo control, absolute (SSI and BISS) and incremental encoder counting, real-time remote I/O, timing, event counting and high-speed serial communication. All 72 7I80HD I/O bits are 5V tolerant. A jumperselectable PTC protected power option allows 1A of 5V power to be supplied to the external daughter cards. Fourlayer construction is used to minimize radiated EMI and provide optimum ground and power integrity. A series of daughter cards is available for industrial motion control, CNC retrofit, high-speed, real-time I/O, Analog I/O, RS-422 interfaces, encoder counting and other applications. The 7I80HD is available with two FPGA sizes, a XC6SLX16 (the 7I80HD-16) and a XC6SLX25 (the 7I80HD-25). Quantity hundred price of the 7I80HD-16 is $108; the 7I80HD-25 is $122. Mesa Electronics, Richmond, CA. (510) 223-9272. [].

3U CompactPCI Serial Processor Board for Modular, High-Speed Applications As part of its High-Speed CompactPCI Initiative, Kontron has announced its first 3U CompactPCI Serial (CPCI-S.0) processor board. The Kontron CPS3003-SA comes equipped with third generation Intel Core i7 processors and offers PCI Express Gen 3.0, USB 3.0, SATA 6G and Gigabit Ethernet over backplane. This paves the way for completely new application scenarios and can greatly boost performance in comparison to classic CompactPCI installations. At the same time, the new CompactPCI Serial class processor board is very flexible. For OEMs who still rely on classic CompactPCI boards, Kontron offers the CPS3003-SA as an option in combination with a CPCI extension module. This means that both CompactPCI Serial boards as well as classic CompactPCI boards can operate together in a hybrid system. Kontron’s new CompactPCI Serial processor board is available in multiple versions and scalable from the 1.7 GHz dual-core Intel Core i7-3517UE processor to the quad-core Intel Core i7-3612QE. For memory-hungry applications, it offers up to 16 gigabytes of ECC DDR3 SDRAM. The Mobile Intel QM77 Express chipset already provides numerous interfaces by default so that the processor board delivers a high performance density in the smallest of spaces. The whole spectrum of serial point-to-point interfaces is accommodated in just three units of height: peripheral boards, which are especially dataintensive, can be connected via two PCI Express Gen 3.0 fat pipes with x8 or x4 lanes. Additionally, five PCIe x1 lanes are available. Hard disk carriers can be connected via four SATA ports, two of them via SATA 6 Gbit/s ports. Plus, there are two USB 3.0 as well as six USB 2.0 ports routed to the backplane. For networking purposes, there are two Gigabit Ethernet ports, which can be routed to the front panel or to the backplane. Additionally, OEMs are presented with two additional USB 2.0 ports and two DisplayPort connections on the front. Furthermore, the CPS3003-SA provides the option to offer rear I/O via the P6 connector, which adds two USB ports (1x USB 3.0 and USB 2.0), a third independent DisplayPort and two serial ports. The Kontron CPS3003-SA supports Linux, Microsoft Windows 7, Windows Embedded Standard 7, Windows XP and VxWorks. Highly integrated Board Support Packages cover all the hardware components integrated on the board. Kontron, Poway, CA. (888) 294-4558. [].

Ad Index Smart Panels Provide High-Value Cost/Performance with Multiple OS Support

Get isConnected with technology and A new series of Smart Panels based on a Cortex-A8 processor companiesAccelerator. providing solutions now and and Integrated PowerVR SGX Graphics The SP-7W61 Get Connected is a new resource further SP-1061 from Adlink support multiple OSs—more than otherfor panel PCexploration into products, technologies and companies. Whether your goal products in the same price range. Along with multiple display size and is to research the latest datasheet from a company, speak directly low power consumption options, these high cost/performance (C/P) with an Application Engineer, or jump to a company's technical page, the products can be implemented a slim and chassis depending onresource. goal of Get into Connected is to thin put you in touch with the right requirements, and can be applied a variety of fields, including comWhichever level ofinservice you require for whatever type of technology, Get Connected munications, medical and industrialwill help you connect with the companies and products you are searching control, with the support of soft-for. ware The SP-7W61 and SP-1061 are both based on the TI-Sitara AM3715 Cortex-A8 processor running at 1 GHz and integrated PowerVR SGX Graphics Accelerator. Get Connected with technology and companies prov Unlike most current embedded Get Connected is a new resource for further exploration into pro systems with limited OS support, datasheet both models support multiple op-from a company, speak directly with an Application Engine in touch with the right resource. Whichever level of service you requir erating systems, including Linux Get Connected will help you connect with the companies and produc 2.6.37, Android 2.3.4 and Windows Compact 7, to meet the ments of a variety of applications in fields such as communications, medical and industrial control, human machine interface (HMI) and transportation. The SP-7W61 (7” 16:9) and SP-1061 (10” 4:3) have a low-power, slim, fanless mechanical design and a high C/P Panel PC module using powerful and efficient components. Compared with other x86 HMI or open frame products, both the SP-7W61 and SP-1061 surpass typical limitations, successfully keeping power consumption to less than 5.9 watts, which is half the typical rate. Their competitive product size makes them easy to implement into slim and thin chassis types, such as those for HMI, control panels or wall mount controller. To maximize product value, Adlink also Get Connected with companies and provides full support on software customization based different platforms. A virtual machine products featured in thison section. or SDK is provided with related documentation for different platforms, so users can easily set up the software environment.


ADLINK Technology, San Jose, CA. (408) 360-0200. [].

Get Connected with companies and products featured in this section.




MicroTCA Carrier Hubs Now Support MTCA.4 Spec Two MicroTCA Carrier Hubs (MCH) have been upgraded to include support of the MTCA.4 specification. Offering the MicroTCA enhancements for rear I/O and precision timing, the AM4904/AM4910 from Kontron are aligned with the requirements of high-speed data acquisition and processing applications. Target segments include physical research centers and many other high-bandwidth applications that require increased serviceability packed into a compact form factor and feature high levels of performance, bandwidth and availability. By supporting four different high-speed fabric variants, including GbE, sRIO, PCIe and 10 GbE, the new Kontron MicroTCA Carrier Hubs build the backbone of practically all data-intensive MicroTCA-based high-performance applications. Besides deployment in physics applications, other typical scenarios can be found in telecom markets—including 3G, LTE and network test equipment—and non-telecom markets such as military, medical, test and measurement, as well as in image and video processing applications. To cover the individual application requirements, the modular Kontron AM4904/AM4910 MCHs come in four different versions. The Kontron AM4904-BASE is a pure Gigabit Ethernet switching version; the Kontron AM4904-PCIE has a combined PCIe mezzanine card and the Kontron AM4904-SRIO features sRIO. For high-bandwidth demands, the Kontron AM4910 provides 10 Gigabit Ethernet switching. All the versions provide central management and data switching functionality on a full-size, single-width Advanced Mezzanine Card (AMC) form factor. Equipped with a powerful 600 MHz PowerPC 405EX processor for MCMC functionality and switching management, the MCHs enable highly efficient, redundant system architectures with up to 12 AMCs, two cooling units and four power units. To provide support for various telco clocks, all Kontron AM4904/AM4910 MCHs feature a clock implementation and a PCI Express fabric clocking distribution. They also include high-performance features such as wire speed, enterprise-class Ethernet switch with Layer 2 switching and Multi-Cast support including full 4K VLANs and Link Aggregation, Quality-of-Service support (QoS). Additionally, a packet classification engine for flexible Access Control Lists is featured. A full set of standard management interfaces including a user-friendly CLI, SNMP, IPMI and web access enable the easy initialization and control of the MCH’s switching and management functionality. External systems or shelf managers can be connected to the Kontron MCHs via the Ethernet front panel ports. Furthermore, the Kontron MicroTCA Configuration Management Software OMVIU is available for remote monitoring and control of MicroTCA systems. Kontron, Poway, CA. (888) 294-4558. [].

Super Wide-View TFT LCDs with Integrated LED Driver Circuits Are Sunlight Readable Three new TFT-LCD products are a 7.0” WVGA and two 8.4” SVGA TFT-LCD displays. All displays feature Super Wide Viewing (SWV) technology and two of them are super high brightness for sunlight readable outdoor applications. Super Wide Viewing technology delivers true color and the best optical performance. It enables people to view images with vivid color (no grey inversion) from any angle. Designed with the latest high-efficiency, long-lifetime LED backlight, the LCD panels achieve brightness levels as high as 1200 nits (cd/m2) and provide an attractive solution for medical, automotive and any outdoor application requiring sunlight readability. The LED driver circuit is integrated into the LCD module, with full dimming function, so no additional components are required to drive the backlight. The backlight itself is specially designed to provide lower power consumption by using the latest technology for LED chips and light guides. The new TFT-LCDs are designed with standard LVDS interface and offer a wide operating temperature range of -30° to 85°C.









800 x 480

800x 600

800 x 600


1000 nits

600 nits

1200 nits

Viewing Angle (U/D/L/R)




Contrast Ratio








Operating temp

-30 ~ 80°C

-30~ 80°C


LED driver





April 2013

April 2013

April 2013





Kyocera offers long-term product availability targeted to meet the needs as a preferred and reliable supplier for industrial, automotive and medical customers. Kyocera Display America, Plymouth, MI, (734) 416-8500. [].







Design Automation Conference







Be a



t of the dA ar

JUNE 2-6, 2013

a r s! Â&#x2021; Au s t



Where IC Design and the EDA ecosystem learns, networks, and conducts business. DAC is the only conference focused on Electronic Design and Embedded Systems and Software (ESS).



stay up to date:

Sponsored by:

DAC.COM In technical cooperation with:

ploration your goal k directly age, the source. ology, d products



Software Modeling

The End of an Era for Source Code For decades, source code has been central to the work of software engineers. Coding—that is, writing and modifying the text that represents the software—is, for most software engineers, the point at which something “real” is created. But this is changing. by Peter Thorne, Cambashi


oftware development engineers are adopters have transitioned to “models” front and center of the revolution in from which they generate code. The the capabilities of smart products. models are the central representation of Embedded software in products ranging the software. Depending on the applicafrom agricultural machines to consumer tion, software engineers may or may not electronics is improving price-perfor- have to work with source code. Even so, mance, flexibility, time-to-market and, source code will remain important for in many cases, quality. Software running decades. But the step-by-step change in in data-centers configures, connects and the status of source code, from something that is written by people, to somesupports these nies providing solutions now smart products to deliver ever more innovative functionality. The thing that is generated automatically by ion into products, technologies and companies. Whether your goal is to research the latest machines, started and is not going result surge intechnical expectations. Every ation Engineer, or jumpisto a a company's page, the goal of Get Connected is tohas put you you require for whatever type of technology, to stop. product with a power supply should conand productsnect you are for. Software engineers deliver an essentosearching the network. Every connected tially intangible product—software. It is product should combine its capabilities the initial configuration of a processor’s with other connected systems to deliver memory. It causes a processor to do what new capabilities. it’s supposed to do. So it is not surprising that software In many ways, this “software” is an design and development engineers are electronic equivalent of the holes in the under pressure to deliver more with less. cards of the punched card system devised The good news is that there are plenty of new tools available to help. Software by Joseph Jacquard to control a machine engineering is at a cusp of adoption of to manufacture patterned textiles. And, a new generation of tools. Many early like manual production of punched cards for textile manufacture, there was a time when manual interaction with program Get Connected with companies mentioned in this article. memory was a plausible way to create software—select a memory location, use

End of Article


APRIL 2013 RTC MAGAZINE Get Connected with companies mentioned in this article.

switches to set the values at that location. But that was long ago, and it was slow and error-prone. Software engineers quickly developed tools to help.

From Concept to Executable

The path from initial concept to final software now contains many different artifacts that provide a view or representation of software. Paper-based drawings of block diagrams and flow charts (the high-level view) were used even when software was being developed on switches (the low-level view). Tools now keep track of all the artifacts used by today’s software engineers; they manage change, allow developers to find old versions, drive workflow, report status and so on. These are of vital importance to software development projects, but here we are going to concentrate on the actual representations of software. “Assembly” languages represent software by translating text into processor instructions, but they require the software engineer to create text for every individual instruction.


Pioneering work from the 1950s onward implemented the “high-level” language concept. In a high-level language, each text combination may define tens, hundreds or thousands of instructions. A compiler translates high-level language text into processor instructions. Coding in high-level languages is easier and more productive than setting switches or writing assembly language. The creators of high-level languages were often motivated by some aspect of software development, and the desire to have a language with which they could implement and test a new concept. There are clues in some of the names. Simula introduced object-oriented techniques in the 1960s because its designer saw that as the right way to articulate software for discrete event simulation. Fortran started out for formula translation. Algol had the expression of algorithms in its heart. But, whatever the motivation, the high-level language concept has been spectacularly successful. Early high-level languages were not immediately suitable for development of software for embedded systems. There were practical objections (“There’s no way to address physical memory locations!”), and performance problems (“The compiler generates too much code”). So creation of software that had to perform in real time and control external physical devices continued to depend on assembly languages. The introduction of C was the watershed that allowed embedded system developers to embrace high-level languages wholeheartedly. C spans a range from high-level constructs—enabling productivity and structure—to low-level control—enabling, for example, device control and interrupt handling. Of course there was no overnight transition. But even reluctant developers would usually find some “unimportant” part of the code that they were willing to rewrite in C. With the introduction of the object-orientated capabilities of C++, the momentum grew.

Figure 1 Software development can, as in this screenshot, courtesy of IBM Rational, depend more on creation and manipulation of system model diagrams, and less on source code editing.

Figure 2 This screenshot from Sparx Systems illustrates how model-based development helps visualize context as well as generate code.

Of course, languages are not the only artifacts that represent software. In the same era that high-level languages were being developed, tools

developers were striving to find ways to translate high-level representations, such as functional block diagrams, into software. RTC MAGAZINE APRIL 2013



Figure 3 This Altova screenshot shows the use of models to write a commercial application.

This idea has taken much longer to bear fruit. For example, the formal conventions of structured analysis techniques (popular in the 1970s) made it possible for data and functional architecture diagrams to define software in the same way a circuit diagram defines an electrical device. But, at that time, it was still necessary for a software engineer to handcraft code to implement the functions and control hierarchies defined by the architecture diagrams. There have always been some examples of high-level constructs being used for direct definition or generation of software. Examples include ladder logic tools for configuration of programmable logic controllers in industrial automation, and data tables generated from state transition diagrams to implement control structures. But this was not the mainstream of software development.



And So to Models

In the early 1990s, Visual Basic (VB) offered a way to develop a user interface containing parameterized predefined controls using just drag-and-drop graphical interaction. Even though the code behind the user interface was still created from high-level language text, the drag-anddrop capability proved to every software engineer that high-level languages were not the whole answer. The drag-and-drop concept, with different technology, is available in many forms for current web and mobile phone app development. When introduced, VB was a solution for a specific type of problem. More general concepts for definition of software structure and behavior were put together by the group often referred to as “The Three Amigos”—Booch, Jacobson and Rumbaugh. Their specification for the Unified Modeling Language (UML) led

to the adoption of UML by the Object Management Group in 1997, and eventually its adoption as an ISO standard in 2005. UML also provided the foundation for SysML—a combined extension and simplification of UML to create diagramming and modeling suitable for systems engineers as well as software engineers. UML and SysML tools have been in use for some time for high-level design and architecture development. Use of these tools allows expression, discussion and development of well-defined highlevel models of software and systems. The growing availability of code generation capabilities is now converting these models from high-level design documentation into the central representation of software. Commercial examples of these tools include Altova UModel, Atego Artisan Studio, IBM Rational Rhapsody and Sparx Systems Enterprise Architect. Figures 1, 2 and 3 show example screen shots. The idea that it is possible to generate software without handling source code has been demonstrated in other environments as well. For example, LabVIEW from National Instruments, and MATLAB/Simulink from MathWorks, both provide graphical system development environments in which there are code elements associated with graphical blocks. The lines that connect blocks define execution and communication of data between these code elements. ASCET from ETAS addresses automotive applications, as does TargetLink from dSpace. This approach creates a new vision. Software engineers do not have to work with the text of source code. The work of a software engineer turns into manipulation of diagrams. Software is obtained automatically from the diagrams.

The Simulation Branch

Simulation tools offer another branch of model-based development for embedded software developers. Tools such as LMS Imagine.Lab and those based on the open Modelica specification (for example, Dassault Systemes / Dymola; ITI / SimulationX; MapleSoft / MapleSim; MathCore / MathModelica; OSMC / OpenModelica; Tongyuan Software and Control / MWORKS), offer multi-domain modeling and simulation in which soft-


ware is one of many simulated technologies that can provide control and execution functions. Symbols are used to represent components that may be hardware or software or a combination. These symbols are assembled and connected into diagrams representing systems. Each symbol contains a statement of the behavior of the component it represents. A simulation run calculates the behavior of the system by using the connections on the diagram to integrate and evaluate the statements that define component behavior. As soon as simulation of part of the system is possible, results can be used to refine the specification of other components. Advanced simulation systems allow the use of real hardware and real software in place of hardware and software components of simulation models. These capabilities are used for “Hardware-in-theLoop” (HiL) and “Software-in-the-Loop” (SiL) simulation studies. For embedded software development, SiL capability allows software to be tested in a simulated external environment. The simulated environment provides a platform for later stages of the software lifecycle, for example, integration of multiple software components, testing, fault simulation and calibration. Figure 4 shows an example screenshot. The value of deeper integration of simulation systems with model-based development systems is being explored in initiatives including OMG’s SysML4Modelica and ModelicaML. In this area, commercial products include the InterCAX range of solvers for modeling tools; and No Magic Inc’s combination of MagicDraw and Cameo. This combination of design models and simulation systems that can use hardware and/or software “in-the-loop” opens new possibilities for the levels of fidelity that can be achieved in virtual design.

Not Your Father’s Development Environment

For the engineers who influence and decide the toolsets that will be used for software development, it is a time of change and opportunity. But there are also difficult choices. This situation parallels the adoption of 3D geometry in tools for mechani-

Figure 4 This active suspension simulation example, courtesy of LMS International, provides a virtual environment to measure control software performance.

cal design. Mechanical engineers now routinely use 3D models to generate 2D drawings automatically (just as a software engineer can now use a model to generate software). A multi-view drawing with a parts list is no longer crafted line-by-line, dimension by dimension, on a 2D CAD system. Instead it is generated automatically from the 3D model. This is the type of change now faced by software engineers. Instead of writing the source code line-by-line, they will work on the system model, and generate the source code automatically. For mechanical designers with substantial skill and know-how built up from years of work with 2D systems, the transition to 3D has always been an alarming prospect. In the early days, the 3D model could not deliver everything, it was still necessary to do some manual drawing. Moreover, there was a range of apparently incompatible 3D tools to choose from. So the move to 3D was spread over decades. Indeed it continues even today. This same profile seems likely for the adoption of model-based software development. Today, in some cases, the whole software development job can be done

using models. But not every case can be handled by models alone. In many cases, hands-on work with source code is still needed. Use of the advanced simulation systems needs skill and experience. So there will be the same combination of resistance (because the models can’t yet do everything), and evangelical uptake (because the models can do some things much better). The change is being driven by software engineers for the same reason that mechanical engineers had an appetite for the transition to 3D. By working on the higher level representation, they increase reuse and spend more time on worthwhile innovation. It is the future, and largely the present in certain industry sectors, in particular automotive. If you make the decisions about development methods, the tough call is going to be when, not if, you promote these techniques from experimental to mainstream in your development organization. Cambashi Boston, MA. (508) 620-4746. [].



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.

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.

Company Page Website ACCES I/O Products, Advanced Micro Devices, Inc.............................................................................................52................................................................................................ American Portwell.............................................................................................................51............................................................................................................. End of Article Products

Amphion Forum 2013........................................................................................................21.................................................................................................. ARM, Ltd..........................................................................................................................31..................................................................................................................

Get Connected with companies and Get Connected Artila Electronics Co.,inLtd..................................................................................................27................................................................................................................ products featured this section. with companies mentioned in this article. Cogent Computer Systems, Inc..........................................................................................26.......................................................................................................... Design Automation Get Connected with companies mentioned in this article. Dolphin Interconnect Solutions. ...........................................................................................5.......................................................................................................... Get Connected with companies and products featured in this section. Extreme Engineering Solutions, Inc....................................................................................11.............................................................................................................. Innovative Integration.........................................................................................................25.................................................................................................. Intelligent Systems Conference & Pavilion...........................................................................36................................................................................................... Intelligent Systems Source.................................................................................................37................................................................................... KW-Software....................................................................................................................32..................................................................................................... Lauterbach........................................................................................................................20........................................................................................................ MSC Embedded, One Stop Systems, PC/104, PC/104 Express & ISM Showcase........................................................................17......................................................................................................................................... Pentek, Phoenix International..........................................................................................................4............................................................................................................

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.



RTC magazine  

April 2013

RTC magazine  

April 2013