Embedded Intel Solutions Fall 2017

FROM THE EDITOR
Digital Signage Computing

Platforms Get Modular www.embeddedintel.com
Vice President & Publisher

Is Intel anticipating that digital sign displays wont need Clair Bright
Editorial
upgrades as much as computing performance? Editor-in-Chief
Lynnette Reese
lreese@extensionmedia.com
Managing Editor
By Lynnette Reese, Editor-in-Chief, Embedded Intel Solutions Anne Fisher
asher@extensionmedia.com
You may have noticed that many food menu boards, health club schedules, and signs in convention
Senior Editors
centers, hospitals, and even churches are now on large digital displays rather than permanent signs. Chris Ciufo
Its easier to change the content, but its about to get easier. Digital signage has a new tool in a modular- Caroline Hayes
minded computing platform. Traditionally, if software outgrows computing power, you get a whole David Bursky
Gabe Moretti
new system. But what if you could swap out the computing platform like a game cartridge?
Creative/Production
Production Trafc Cordinator
Earlier in 2017, Intel announced a family of Compute Cards, each of which has a processor, onboard Marjorie Sharp
storage, memory, wireless connectivity, and flexible I/O. Upgrading performance means a new Graphic Designers
Compute Card, not a whole new computing platform, which also means less for the landfill. An Intel Nicky Jacobson
Simone Bradley
Compute Card is a bit longer than a small stack of credit cards at 95mm x 55mm x 5mm, but its got Senior Web Developers
all of the elements of a full computer. Companies can make docks for all kinds of application-specific Slava Dotsenko
products, since Compute Cards bring the entire computing platform in as a single component. Mariam Moattari
Advertising / Reprint Sales
Meanwhile, Intel has created a 19V dock with an HDMI v1.4 port, Mini DisplayPort 1.2, a LAN RJ-45 Vice President, Sales
port with 10/100/1000 Mbps Ethernet controller, and three USB 3.0 ports. Some will wonder how this Embedded Electronics Media Group
Clair Bright
is much different from a USB stick. You can boot to an operating system from a USB stick, but USB sticks cbright@extensionmedia.com
dont have processors. Imagine going to a library with your personal computer card; your desktop and (415) 255-0390 ext. 15
software settings are there, you dont have to clean the browser cache when you leave, and your applica- Sales Manager
Elizabeth Thoma
tions are all there and run as expected. This is one of those why didnt I think of that products. ethoma@extensionmedia.com
(415) 244-5130
Digital signs with cumbersome existing physical installa- Marketing/Circulation
tions will find upgrading easier. Imagine updating the digital Jenna Johnson
signage systems in New York Times Square to accommodate jjohnson@extensionmedia.com
new software in various layers of the stack that enable revo- To Subscribe
www.embeddedintel.com
lutionary new programming features. The display would stay,
and the Compute Card would be changed out for one that can
handle the increased performance requirements. In an Intel
Figure 1: The Intel Compute Card and Dock.
press release dated May 30, 2017, Intel stated that there were Extension Media, LLC
Intels four Compute Cards provide various levels a large number of partners currently working on [dock] solu- Corporate Ofce
of performance. (Credit: Intel)
tions that include Dell, HP, and Lenovo. A whole President and Publisher
range of products can be offered, allowing end Vince Ridley
vridley@extensionmedia.com
users to start at the low end and buy up to higher (415) 255-0390 ext. 18
performance later on. Guaranteed effective Vice President & Publisher
upgrades can be shipped on a slim Compute Card, Clair Bright
cbright@extensionmedia.com
negating the need to negotiate perceived failures
Human Resources / Administration
due to inadequate platforms with customers who Darla Rovetti
might be better at managing retail than man-
aging an upgrade to their chic electronic displays. Special Thanks to Our Sponsors
Applications dont stop at digital signs and kiosks.

The Compute Card is proposed for use with Internet
of Things (IoT), tablet-based systems, interactive
Table 1: The Intel Compute Card presently features four levels whiteboards, intelligent vending machines, mobile
of performance. Intel is also releasing the Compute Card Device video production, smart TVs, robotics automation,
Design Kit, a set of guides and reference designs that are avail-
able via Intels Classied Design Information (CDI) portal for any security systems, point-of-sale systems, and all-in-
customer under NDA. one PCs that are easily upgraded.
Successful applications always get upgraded with more sophisticated software that pushes the perfor-
mance envelope. Can Intel patent the concept of a compute-platform-as-a-cartridge? Will motherboards,
in general, shrink to small modules worldwide? As of this writing, the Intel Compute Card has not yet Embedded Intel Solutions is sent free to engineers and embedded developers
been officially released, but should be out soon, as advance demonstration models have been sent out. in the U.S. and Canada who design with embedded Intel processors.
Embedded Intel Solutions is published by Extension Media LLC, 1786
18th Street, San Francisco, CA 94107. Copyright 2017 by Extension
Lynnette Reese is Editor-in-Chief, Embedded Intel Solutions and Embedded Systems Engineering, and has been working in Media LLC. All rights reserved. Printed in the U.S.
various roles as an electrical engineer for over two decades.
2 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

IN THIS ISSUE
FALL 2017
DEPARTMENTS
FROM THE EDITOR DIGITAL SIGNAGE/SMART DISPLAYS

2 Digital Signage Computing Platforms Get Modular 23 What Makes an Industrial Digital
By Lynnette Reese, Editor-in-Chief, Media Player Different?
Embedded Intel Solutions By Robert Suffoletta, Logic Supply
25 System-on-Chip: Not Yet

SPECIAL FEATURES Ready for Prime Time
By Mark Boidman, Ben Zinder, and Gilbert
Baltzer, Peter J. Solomon Company
FOCUS ON INTEL
ANDROID & LINUX
6 The Passenger Economy: Challenges Ahead
27 Docker Containers Ease Cloud
By Lynnette Reese, Editor-in-Chief,
Embedded Intel Solutions
and IoT Implementation
By Lynnette Reese, Editor-in-Chief,
IOT SECURITY Embedded Intel Solutions
9 Why Blockchain May Not Reach Beyond 30 Open Source Meets Standards: A
the IP Gateway in IoT Security Platform for Fast Implementation
By Guillaume Crinon, Avnet By Dan Demers, congatec
12 When Adding IoT Devices to the

EnterpriseKeep it Secure PRODUCT SHOWCASES
By Rusty Stapp, UbiquiOS Technology
14 We have to coordinate that root of 32 ADL Embedded Solutions

trust: Q&A with WinSystems
By Anne Fisher, Managing Editor
LAST WORD
NETWORKING & DATA CENTER TECHNOLOGIES
17 Delivering Carrier Grade OCP to Telco Data Centers 33 The Eyes of the IoT Rely on
By Todd Wynia, Artesyn Embedded Technologies Processor Innovation
By Caroline Hayes, Senior Editor,
21 Jumpstart Connected Strategies Embedded Intel Solutions
By Ed Trevis, Corvalent
On the Cover:
...its key to design and plan for security upfront. I still often
see that folks will look at the feature set first and then get
down the road before they start thinking about how they
are going to secure the system. See why right sizing is part
of securing the IoT in the WinSystems interview, page 14.
Image courtesy WinSystems

FOCUS ON INTEL
The Passenger Economy:

Challenges Ahead
As horseless cars become driverless cars, plan to see a tremendous
increase in productivity in every area that transportation touches today, and
in applications yet to be imagined.
By Lynnette Reese, Editor-in-Chief, Embedded Intel Solutions
Early Autonomous Vehicles (AVs) are here but pose grave

engineering challenges if we are to implement them in real life.
We are witnessing a moment of unprecedented transformation
in automotive history. The time when AVs are as common
as smartphones is a couple of decades or more away, but
the potential for new markets and significant increases in
productivity are seductive.
Autonomous Vehicles to Save $1.3 Trillion a Year

Morgan Stanley predicts that the U.S. alone stands to
save $1.3 trillion a year, presumably with a full-scale Table 1: Average Commute Time to Work, per Day by City, as of 2013
implementation of AVs. Nearly $500 billion annually would
be saved in accidents due to reductions in repairs, medical lower fuel consumption, less traffic congestion, and personal
bills, and lost time, with another $500 billion in overall productivity gains (assuming the average person drives an
productivity gains. Improvements include improved logistics, hour and a half each day). Time spent commuting, finding,
and paying for parking in crowded cities will
be significantly reduced as everything runs
more efficiently. Sharing vehicles means fewer
cars on the road and a lower cost of living for
individuals while maintaining autonomy. The
vision for a world with AVs includes less traffic,
better maps through detailed communications
from intelligent cars, and mobility for all. A
recently released report by Strategy Analytics
includes some tantalizing predictions, such
as lower public safety costs due to traffic
accidents that could total over $234 billion
from 2035-2045. Commuting time saved by
autonomous vehicles may amount to over 250
million hours per year for consumers in the
most congested cities1.
Once people cease to drive, what do they do? A

new market will open up, called the Passenger
Economy, a term coined by Intel CEO Brian
Krzanich. Different applications, markets,
Figure 1: Several sensors on the 2018 Audi A8 send data to the and businesses can shoot out of this inflection
autonomous driving system. (Source: Audi)
1. Intel Predicts Autonomous Driving Will Spur New Passenger Economy Worth $7 Trillion. Intel Newsroom. Intel Corporation, 1 June 2017.
Web. 26 July 2017.

FOCUS ON INTEL
in AVs and jobs are re-established in areas related less to the

actual driving and more to logistics management.
What can we expect? Winter points out the trucking industry

as one example where Mobility-as-a-Service (MaaS) will
dominate. Look at the trucking companies. Its well known
there is a shortage of truckers, people wanting to drive
long-haul trailers, and the need to move packages around is
escalatingbusiness [can take] advantage of Mobility-as-a-
Service in not having to have drivers. This is time and money
for them. The technical challenges to attaining AVs are not
Figure 2: Estimated global passenger economy service revenues
from 2025 2050 (Source: Strategy Analytics) small. AVs will shift the semiconductor industrys focus to
processing teraflops of data at blinding speeds to fulfill the
point in how we carry out our lives. Consumer services such as need for low latency, high bandwidth, and rapid throughput on
entertainment, advertising, and personal or financial services a scale that includes data creation and consumption the likes
might be carried out inside autonomous vehicles as people of which we havent yet seen as commonplace.
travel to work. Intels Katherine (Kathy) Winter, Vice President
and General Manager of the Automated Driving Division, The Technical Trinity: Car, Cloud, and
deliberates a passenger economy and the new markets it will Connectivity
incite. Its probably the smaller piece right now, but mostly The picture for AVs includes the car, a cloud, and connectivity.
because we cant imagine what it is yet, she states. Winter AVs are loaded with sensors to create enough information
suggests that this portion of the passenger economy would on which to make split second decisions necessary to drive.
include the new services that a person in an [autonomous] Myriad sensors are loaded onto AVs, including but not limited
vehicle could be consuming; something [like] entertainment, to radar, LIDAR, cameras, infrared sensors, and GPS. Many
education, advertising, things like that. of these sensors are placed all the way around the vehicle
and work together to build the cars vision of whats going on
As a population of drivers evolves into passengers, Intel, around it and where should it go next. Varying environmental
working with Strategy Analytics, finds that by 2050 the conditions make redundancy and overlapping data input
passenger economy is an estimated $7 trillion industry.1 necessary to fuse together a knowledge that is more reliable
Winter indicates that the semiconductor industry must than mere vision as to what is going on around, behind, and
squeeze yet more improvement in computing power to ahead of the car so that decisions can be anticipated. Changing
meet technical challenges that autonomous driving poses, environmental conditions include road conditions, variations
including leaving headroom for things we cannot imagine in street layouts, random or sudden incidents involving
yet. The largest factor holding the passenger economy back bicycles, pedestrians, or animals, and changing weather
will likely be social, not technical, as new laws and regulations conditions that can affect some sensors ability to gather
must be crafted. Consumer acceptance may be slower than accurate data more than others. Theres a reason why Phoenix,
technologists and early adopters would like, as trust builds Arizona is such a popular testing ground for AVs these days.
Winter points out that Phoenix is laid out like a grid and that
testing engineers are pretty much counting on it being sunny.
Youre not going to see snow, youre not going to see ice. But
we need vehicles that can drive autonomously every day of the
year, not just when its sunny, not just when its not raining or
snowing, or below 32 degrees.
AV prototypes must test to optimize the right mix of sensors

for different environmental conditions. Thus, accurate sensing
in every kind of condition is part of the challenge to making
the passenger economy a reality. Why do we need all these
sensors? They all complement each other, which is kind of the
bottom line. And as we go through more and more testing, and
theres more of those vehicles out there, we are learning about
the combinations, how much redundancy, things like that,
that you actually need in the vehicle.
Figure 3: Estimated global passenger economy service revenues by
region in the year 2050 (Source: Strategy Analytics)
www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 7

FOCUS ON INTEL
Proving Grounds for high bandwidth, super low latency, and wide-ranging wireless
High-Performance connectivity. But Intel has found that the use case for AVs has
Computing taken the lead for 5G, according to Winter. The autonomous
High performance embedded vehicle seems to be one of the forces really driving the need
computing (HPEC) is another area for [5G], and the lead use case from what we can see. You
that challenges the trinity of car/ need a way [to communicate], and weve talked about the end-
cloud/connectivity also known to-end, mission critical information, that needs to be really
as the three Cs. Extremely fast, fast, low latency [communication]. The other piece is the pure
low latency, high-throughput volume of all that data. Once 5G is out there, cars can use it to
devices that are extremely secure communicate to one another, to the cloud for critical updates
are on the agenda for acceptance both ways, and to and from surrounding infrastructure. It may
and success of this pending be that speed limit signs will disappear one day, replaced by
phenomenon. Winter comments innocuous transmitters that signal the speed limit to passing
Figure 4: Kathy S. Winter, that today, the estimated average cars. Stoplights may transmit status to vehicles directly. The
Vice President and General amount of data that a person anticipated 5G also adds another sensor to the car.
Manager of the Intel generates in a typical day is
Corporation Automated
Driving Division. (Source: around 650 MB today. By 2025, So How Do We Get There?
Intel Corporation) this figure might be nearer to 1.5 Labor costs may go down with AVs, but we will see an increase
GB per day. However, AVs produce in demand for cloud services, servers, predictive analytics,
an estimated 4,000 GB per day. and Internet of Things (IoT). Data collection, processing,
How do you get to that number at scale? Winter states that storage, retrieval, and analytics will be even more important
an average vehicle thats driven an hour and half a day yields as we analyze, predict, optimize, decide and anticipate using
that kind of data in the vehicle itself. This translates to a high-performance computing as just part of the picture. At a
challenge for HPEC. Today in 2017, theres probably half a full-scale implementation, the challenges of a driverless world
teraflop of data coming off those [self-driving] cars, on an seem tremendous. One step at a time, building on the work
average day. If you go out to 2025, you think about all the of others, is how science and technology have evolved for
sensors, all the data, everything coming off there, look at the centuries. What Intel has been pushing for, and what Winter
computing power needed to process, that, store, share, save, believes is critical to accelerating success, is in standards and
et cetera. collaboration across the industry. Winter believes that success
is in developing talent for the task and in sharing potentially
Other serious challenges include where and how to rapidly mission-critical safety dataacross the industry, so that
store and retrieve enormous amounts of data, and how to every vehicle manufacturer, every fleet operator, doesnt need
manage, process, save, store, and share that data from each to put in millions and millions of miles to understand the
and every car on a scale thats beyond anything we have seen safety aspects of driving autonomous vehicles. Sharing that
to date for an infrastructure that stretches coast-to-coast. kind of safety data will accelerate the pace of the industry.
Each state has differing traffic laws, landscape, and weather. Additionally, shared standards mean that no one has to learn
A large number of servers positioned across the nation will and develop uniquely, on their own. Putting some standards
handle storing data thats needed to perform deep-learning and industry-wide platforms in place will help accelerate the
processes, and create algorithms for use by a massive fleet entire industry. If you really think about this from a safety
of AVs that will learn collectively from what can only be perspectiveits worth doing for everybody.
described as each others learning experiences. And not all
data is equal. How do we decide what is safety-critical? If Winter says that one of the most common questions she hears
safety-critical data is obtained, how do we get it to a fleet of is What will it take to make this happen? Winter states, To
cars on a massive scale, and as soon as possible? Theres lots make this happen, the biggest thing we need to do is trust and
of learning and testing to do, and as yet, communications just let go. Let go of the wheel.
technology is missing a critical piece.
Lynnette Reese is Editor-in-Chief, Embedded
Connectivity Intel Solutions and Embedded Systems Engineer-
Communication is the last vital piece to work out in how the ing, and has been working in various roles as an
trinity of car, cloud, and connectivity brings AVs to full scale. electrical engineer for over two decades. She is
Wireless 5G communication has been on the radar for some interested in open source software and hardware,
time now as a well-needed means of upgrading cell phone the maker movement, and in increasing the num-
communications to much faster, higher bandwidth wireless ber of women working in STEM so she has a greater chance of
services. Intel has long recognized 5G as ideal for industrial talking about something other than football at the water cooler.
use cases, smart cities, and anything that requires super

Why Blockchain May Not Reach
SPECIAL FEATURE
Beyond the IP Gateway in IoT
Security
Is it time to walk away from the centralized architecture weve used for
IOT SECURITY
more than two decades?
By Guillaume Crinon, Avnet
Blockchain is a technology initially deployed at the heart of

crypto-currency systems such as Bitcoin or Ethereum where
the basic idea is to get rid of centralized banking systems ruling
monetary transactions. Instead, every transaction is signed by
its emitter (private key) and broadcast to everyone along with
the corresponding public key. Then, everyone is able to check
the validity of the transaction, going back in history to check
that the spender has the money, and registers it in the form of
a block mathematically chained to the previous block into a
shared and massively duplicated public ledger keeping track of all Figure 1: The Diffie-Hellman key contract established in 1975 is the
transactions to date. A public set of rules ensures that duplicated most important use of asymmetric cryptography.
ledgers are consistent with one another. Public cryptography tools
make sure that no one can corrupt past transactions in ledgers. A very efficient way to implement these concepts in the real
Distributed among participants mostly playing by the same world of low-power battery-operated sensors and gateways is
official public rules, trust is decentralized, no longer concentrated to rely on symmetric cryptography, which uses the same key
among a few banks. on both sides to encrypt/sign and decrypt/verify. For example,
mutual authentication and message integrity are commonly
To incentivize participants faced with the tedious job of checking achieved by appending each message sent with a 32-byte
others transactions and contributing to the ledger, crypto hash code (SHA256 function for instance) taking as inputs:
currencies have also engineered rewarding mechanisms yielding the message to be sent; a means to defeat replay-attacks by
a very controlled monetary creation, often referred to as mining. a third party; and a pre-shared 128/256-bit key. Message
confidentiality is usually implemented with an AES cypher
IoT Security Basics taking as inputs: the message to be sent and a pre-shared
Securing communication between two devices within a 128/256-bit key.
network relies on three ideas:
1. Mutual authentication: each device

should be able to prove its true identity
to the other.
2. Message integrity: each message sent

to the other side should be signed in a
way that prevents any interferer from
secretly changing its content.
3. Message confidentiality: option-

ally each message may be encrypted
so that only the authorized parties
understand the content. Figure 2: Exchanging public keys to compute the same AES/SHA
key on both sides

SPECIAL FEATURE
Whereas these implementations are widely deployed in Figures 3 and 4 show, the robot and the server willing to
most communication protocols such as Bluetooth, 802.15.4, open an end-to-end secure channel are often distant and
802.11, etc., only the IP-compliant ones can take advantage their communication has to go through many intermediaries
of an efficient symmetric key distribution scheme called and networking layers, such as a Local Area Network (LAN)
Transport Layer Security, or TLS, formerly SSL. The other gateway bridging to a Wide Area Network (WAN), and several
implementations rely on more or less manual pairing, factory telecom operators.
pre-setting of keys, or even distribution of keys in the clear!
Simply swapping public keys between robot and server
Efcient and Secure requires trusting that all the intermediaries will pass the
It turns out that the mechanisms at work within TLS can be public key to the next level. However, they could also keep it
IOT SECURITY
easily ported over non-IP communication protocols yielding for themselves and pass on their own, breaking the end-to-end
efficient and secure systems all the same, as demonstrated by tunnel into two tunnels and eavesdropping and stepping into
Avnet with the end-to-end security implementation of Visible the conversation unnoticed, thus successfully performing a
Things 2 between a Bluetooth sensor and an IBM cloud server. well-known man-in-the-middle attack.
In order to solve this famous problem, the

Internet has been relying on certificates
delivered by a Certification Authority (CA).
CAs are entities trusted by everyone whose
role is to bind a public key to its owners
identity into a document they sign in turn:
the certificate.
Anyone willing to share a public key will

therefore have it converted into a certificate
first. The other party receiving the
certificate will then trust that the public
Figure 3: End-to-end device-to-server security key belongs to the sender because the CA
guarantees the bond between the public key and the senders
In short, TLS relies on asymmetric cryptography and a identity.
transaction invented by Diffie and Hellman in 1975, allowing
TLS to compute the same shared secret between two entities, We now have a full solution working and relying on a
which are exchanging only public information. Asymmetric centralized trust architecture built around a few CAs able to
cryptography works with mathematically-correlated key pairs: bind in a well-protected database the unique identity of objects
One key is defined as private and should neither be shared nor and their public keys.
exposed, while its counterpart is public and can be distributed
to peers. When one is used to encrypt/sign, only the other one As long as everyone trusts these CAs and their capability to
can decrypt/verify. keep their databases highly secured, the architecture holds
perfectly well.
Applying the Diffie-Hellman key contract to any
communication session solves the
problem of securely distributing
and renewing symmetric keys in
the field, which is also how TLS over
HTTP (making it HTTPS) works:
Certicates and Trusted

Third Parties: A Centralized
Model
Swapping public keys between
devices, machines or servers
allows opening a secure channel
end-to-end between them and
implementing state-of-the-art
mutual authentication, integrity
and confidentiality. However, as Figure 4: Man-in-the-middle attack

SPECIAL FEATURE
Blockchain: The Promise of Decentralization Anything battery-powered and low-cost will certainly not have
But the word is out that Blockchain will disrupt this scheme the processing capability nor the energy to participate actively
very soon, allowing peer-to-peer secure communication in a public or private Blockchain infrastructure as previously
between sensors and devices not needing CAs anymore. described. So much for secure peer-to-peer communications
between battery-powered devices.
How trust could be decentralized with a Blockchain
architecture raises many questions: As a consequence, a Blockchain architecture for trust and
security is likely to stop at the network edge, no further than
1. Should all the sensors, devices, machines, servers able to the gateway, provided that the gateway has processing power,
communicate globally or within an application or a vertical network bandwidth and a sufficiently large amount of memory
IOT SECURITY
participate? to participate in the community job.
2. If the idea is to get rid of CAs, would it mean that every Conclusion: The Cost of Trust
device could sign certificates for others or that certificates A distributed trust architecture will not come at no cost. Some
themselves would not be needed anymore? mechanism and business-model will be needed to finance the
extra hardware, the extra cellular bandwidth, the extra energy
3. Distributing the trust would also imply distributing a trust and earlier wearing-out of those devices and machines which
database in the form of a ledger. What should it contain? will support the computing and storage of trust processing.
Bonds between IDs and public keys or more?
Looking at the industry, I am not sure we are ready yet to
4. Would every participating device be required to check the discard the centralized trust architecture the world has been
ID of others transacting messages and registering them relying on since the Internet went live more than 20 years ago.
into a shared ledger? More probably, trust entities will structure themselves with a
Blockchain architecture and offer compelling business models
5. How would the devices described in point 4 be incentivized? for their customers to collaborate in the overall maintenance
and redundancy of their trust business.
6. By which mechanism would they be capable of
synchronizing their copy of the public ledger with peers Further, Industrial IoT or Industrie 4.0 use cases will be
so as to maintain a consistent ledger overall? massively vertical and self-contained to start with, more like
Intranet of Things deployments. Therefore, its possible that
7. How would they efficiently store a valid copy of the major players that everyone trusts will offer centralized trust
duplicated ledger locally? architectures and business models 10 years from now.
With a current average of 250k monetary transactions every The future will tell
day, the Bitcoin ledger has grown to 130GB since inception
in 2009 and is expected to carry on growing at a rate greater Guillaume Crinon is the Innovation Technical
than 50GB per year. Marketing Manager for Avnet EMEA, where he
is responsible for Internet of Things (IoT) and se-
The process of mining new blocks, once affordable curity strategy across Europe. He has 22 years of
by a personal computer, has become so demanding in experience in the semiconductor business and has
computational resources and energy because of the monetary co-authored 11 international patents in wireless
creation regulation mechanism that only a small number of systems, IC architectures and design to date.
computer pools (less than 10 in the world, mostly in China)
now own 50% of the capacity, hence the trust of the Bitcoin
currency.
As a reminder, IoT hardware can be classified into two

categories:
1. Mains-powered: servers, gateways, high-end sensors and

actuators in addition to rechargeable devices to some extent
2. Battery-powered: low-cost sensors, actuators, tiny

gateways and rechargeable devices

SPECIAL FEATURE
When Adding IoT Devices to the

EnterpriseKeep it Secure
When will IoT device developers heed the call for full circle IoT-to-IT security?
IOT SECURITY
By Rusty Stapp, UbiquiOS Technology
The addition of Internet of Things (IoT) devices and to-end security. Among the security measures found in IoT
applications to enterprise IT networks is increasing the attack deployments are wireless protocols such as Bluetooth Smart,
surface available to hackers. Failure to secure that surface Thread, or Zigbee, which require gateways to access the public
presents a real danger to the security of the overall systems. Internet or even the corporate intranet. The use of firewalls
Traditional computing/IT systems have been working mightily and integration with public cloud services such as Microsoft
to include robust practices and standards to protect important Azure IoT Hub, AWS IoT and IBM Watson IoT has also helped
enterprise devices on their networks. This hasnt been the case secure such devices.
as much for IoT applications that increasingly must attach
to and communicate with the IT systems. That remains a Such underlying connectivity technologies as Wi-Fi and
primary source for widespread concern, and it has slowed but Bluetooth may well provide good security at the link layer,
not prevented the increasing adoption of such devices. The risk but again, that security exists only among the communicating
grows while the potential advantages of adoption are slowed. wireless devices and the endpoints they are connected to (e.g.,
sensor, router, gateway). Now, the challenge is to provide end-
to-end security beyond the IoT endpoint all the way to the
cloud server. Ideally, this should include a root of trust based
on secure hardware. Such end-to-end security needs to address
three main concerns: authentication, integrity, and privacy.
Clearly, there is a gap between what IoT developers create and
what IT architects need.
End-to-End Security from IoT to IT

To ensure the same level of end-to-end security from the IoT
level across the enterprise, the Transport Layer Security (TLS)
protocol is the first line of defense. With origins in the Secure
Figure 1: IoT deployments include multiple protocols, controlled Sockets Layer (SSL), TLS state-of-the-art algorithms secure a
access to the public Internet and integration with public Cloud ser-
vices such as Microsoft Azure IoT Hub, AWS IoT, and IBM Watson IoT.
connection between two nodes and ensure:
t Communicating parties identities can be authenticated

According to a study by the Ponemon Institute, 75 percent using public key cryptography. This authentication can be
of respondents say the use of IoT apps significantly increases made optional but is generally required for at least one of the
security risk, with nearly the same number being very parties (typically the server). Such authentication ensures
concerned about the use of insecure IoT apps. Despite that that the end-point connects only to the proper server, and
concern, 44 percent of respondents say their organization isnt that the server accepts connections only from bona fide end-
taking any steps to prevent attacks. For IoT device developers points.
to enjoy a ready market and realize maximum growth,
something needs to change. t The connection safeguards integrity because each message
transmitted includes a message integrity check using a
Merging IoT Security with Enterprise Security message authentication code to prevent undetected loss or
While IoT deployments often include architectural features alteration of the data during transmissioncaused either
to provide security, the challenge is to smoothly integrate by malicious action or unintentional corruption.
them into the enterprise environment to achieve end-

SPECIAL FEATURE
t The connection is private because
symmetric cryptography is used to
encrypt the data transmitted. The keys for
this symmetric encryption are generated
uniquely for each connection and are
based on a shared secret negotiated at the
start of the session.
To provide the highest levels of security,

TLS relies on asymmetric public key
IOT SECURITY
Figure 2: Enterprise IoT applications require true end-to-end security
cryptography in which each communicating entity is assigned between the IoT end-point and the Cloud server, with typical con-
cerns including authentication, integrity, and privacy.
a unique and strong private key, and a mathematically related
public key. The private key must be kept secure with its owner
and not disclosed to any third party. The public key, on the using measurement of dynamic supply currents or analysis
other hand, may be made generally available, as it is practically of electromagnetic emissions and cannot be considered
impossible to use it to determine the private key. reliably secure. For applications with the highest security
requirements, developers should choose a hardware secure
Public key cryptosystems structures enable two primary elementa device built on tamper-resistant hardware that
operations: implements mechanisms to:
t The public key may be used to verify that the holder of the t Store the private key in a manner that makes it irretrievable
corresponding private key signed a given message. by all practical means
t The public key may be used to encrypt a message such that t Perform the cryptographic operations necessary to sign
only the corresponding private key holder can decrypt it. and/or decrypt data using the private key
TLS builds on these functions to directly achieve the goals IoT products typically have limited user interface and
of authentication and ensure message integrity. Privacy frequently need to be installed and commissioned by non-
is indirectly achieved by using communication under the expert users. As a result, they must support provisioning
protection of public key encryption to establish a shared secret flow, which is very simple to follow, while still maintaining
that may then allow a more computationally efficient symmetric security of the overall system. With the device identity and
encryption scheme to protect subsequent communications. root of trust embedded within a hardware secure element
in the IoT end-point, the provisioning process can, without
Secure Hardware compromising security, enable the device with credentials and
With TLS securing the sender, the next step is to protect details for access to the wireless network and/or gateway via
the receiving IoT end-point. For TLS to secure edge-to-cloud which it will access the Internet.
communications, developers need to address the following:
Designing robust security into an IoT product whether a
t Secure storage of the private key in the IoT end-point, camera, thermostat or temperature sensor, is possible and
so an attacker wishing to impersonate the device cannot should be part of every designers criteria. Not only will
retrieve it. the use of industry standard practices provide a high-level
of protection and ease implementation in the field, it will
t Ensuring the authenticity and validity of the public key also provide the assurance needed for large enterprise level
certificate offered by a server to which the IoT end-point distributors to come onboard, resulting in a dramatic uptick
is attempting to connect. in market growth.
As noted, private keys must remain known only to the IoT end- Rusty Stapp is the CEO of UbiquiOS Technology,
point to ensure the overall security of the system. If compromised, which enables low-cost Wi-Fi, Bluetooth and LP-
an attacker could use the private key to impersonate the IoT end- WAN connectivity for the Internet of Things (IoT)
point, intercept communications, or gain increased information market. Stapp brings nearly 30 years of component
for a potential attack on the cloud infrastructure. and material customer engagement experience to
his role. He has led eorts in Europe, North America
While modern microcontrollers and architectures often and Asia for large companies like Texas Instruments and Kodak to
implement mechanisms to protect non-volatile memory VC funded startups. Stapp also worked at NextWindow, where he
and incorporate cryptographic co-processing engines, led engagements with HP, Microsoft and others. He holds a EET
such platforms can still be susceptible to discovery attacks from Texas A&M and a MBA from University of Texas at Dallas.

SPECIAL FEATURE
We have to coordinate that root of

trust: Q&A with WinSystems
Why right sizing is a part of securing the IoT
IOT SECURITY
By Anne Fisher, Managing Editor
well as enabling all of the broad features that Intel has brought to
the tablebecause every chip that Intel brings to the table has new
IoT features and improved security, Smith adds.
EECatalog spoke with Smith and George T. Hilliard, Director of

Technical Sales, WinSystems, about the Industrial IoT and security
recently. Edited excerpts of the interview follow.
EECatalog: Whats important for someone targeting Industrial

IoT applications to know about the elements supporting
Secure and Trusted Data in a product like WinSystems PX1-
C415 SBC? [Figure 1]
Editors Note: This spring saw publication of the Industrial Internet

Connectivity Framework by the Industrial Internet Consortium
(IIC), of which Intel is a founding and contributing member. Intels
involvement at that level in an organization striding purposefully
toward a trustworthy IIoT in which the worlds systems and devices
are securely connected and controlled.1 is a role well understood
by WinSystems, with roots firmly in industrial computing, and
now branched to the MIL-COTS, energy, transportation, and Figure 1: A PC/104 form factor SBC with PCIe/104 OneBank expan-
automation verticals as well. sion, the PX1-C415 single board computer from WinSystems includes
the latest generation Intel Apollo Lake-I E3900 SOC processor.
Understood too is embedded designers ongoing need for information

as threats to IIoT security persist. We have a strong relationship George T. Hilliard, WinSystems: From a
with Intel, and can help our customers understand the security systems-level perspective, its key to design
measures that exist and Intels security roadmap for the IoT on and plan for security upfront. I still often
all the device levels that are available, T.J. Smith, WinSystems see that folks will look at the feature set
Technology and Engineering Director, tells EECatalog. first and then get down the road before they
start thinking about how they are going to
On a regular basis, we help customers understand what is coming, secure the system. Then they have to do some
George T. Hilliard,
what the timeline is, and how to implement toward their needs as WinSystems
redesign. Thats sometimes a missing element
1. https://lwn.net/Articles/506761/

SPECIAL FEATURE
out there. A secure way to start is with the root of trust and the EECatalog: How do you assure various security elements
platform itself. And thats where we come in as an embedded complement one another in an SBC?
systems provider. We give customers a hardware platform
where they can create that root of trust using Intel or other CPU Hilliard, WinSystems: We give the choice to the customer. For
products and have that space as a starting point for the security. example, customers can add a Trusted Platform Module (TPM)
Addressing security in parallel with application development and take advantage of secure boot and other features, such as the
and design speaks to the customers bottom line. By using this integrated Intel security engine and ECC memory, which gives
approach, he can minimize risks and control costs rather than you a more reliable RAM for your device, and helps you protect it
going back at some point and having to redesign. from a forced reset vulnerability, for instance. As Jack [T.J.] notes,
they can right size that security solution for the application, and
IOT SECURITY
T.J. Smith, WinSystems: As soon as you that is where their planning and expertise are going to come in.
realize that your product has value, you
need to understand how you are going to Smith, WinSystems: Exactly. The platform is designed from the
protect that value. So, you have to do an start to enable, but not require, all levels of encryption and security.
assessment very early on in the product A strong level of security requires a good root of trust. And where
life cycle. that root of trust resides is an important decision between us, the
designer and manufacturer of the hardware, and the consumer,
T.J. Smith, Its also possible to oversize or overdevelop the integrator, and their development of their application. We
WinSystems
on security. One of the things we target have to coordinate that root of trust. Our hardware is capable of
with our customers is how to right size their security. You providing a hardware root of trust. We can also enable any of the
need to understand the value of what is being protected and variety of layers that are well integrated with the Intel platform.
put the appropriate measures in place to address that security The Apollo Lake platform provides a generous helping of security
now and into the future. What will need to be monitored? capabilities. The application author and system integrator have
What will need to be adapted 10 years down the road? an awful lot of options for developing security. We support them
through that decision process; we support them through that
EECatalog: How have some of the actions taken by Intel in the implementation process. And we definitely support them through
past couple of years, including licensing decisions, affected the the production and manufacturing ramp into volume.
solutions WinSystems is developing for its customers?
EECatalog: Please comment on the Industrial Internet
Smith, WinSystems: Longevity is very important. Intel Connectivity Framework (IICF).
has been strongly promoting its longevity and has been
making licensing decisions based on that. When you go into Smith, WinSystems: I like the direction the IICF is headed.
deployment, you have to understand not just security today, Its reached critical mass at this point and will be effective. The
but security tomorrow. How are you going to make sure that IICF defines a wonderful selection of reference architectures
two years down the road the millions of the products youve and frameworks to apply to basically any IIOT application.
deployed into the field dont suddenly become vulnerable with They are helping drive interoperability in the space so that
no cost-effective way to repair them? The advanced decisions developers can integrate layers even if they are from different
involve understanding what a proven framework can do and vendors. This is particularly helpful to WinSystems and our
how WinSystems and Intel and the different software and IIOT ecosystem partners as we help our customers get their
security layers play together to give you a secure yet flexible, products to market.
adaptable, solution for your IoT products without going
through the roof on costs. Hilliard, WinSystems: The [Industrial Internet Consortium]
IIC is involved in the standards, but they are not trying to set the
Hilliard, WinSystems: In the embedded space, which has standards so much as to get individual standards to talk to each
now morphed into what everybody is calling the IIoT, Intels other. With all these big installed bases for industrial control
acquisition of McAfee and VxWorks shows its commitment to there are so many different protocols on industrial Ethernet, for
embedded systems platforms. We are seeing these technologies examplethey are trying to find ways for all of these platforms
picked up into the consumer products, which is encouraging to talk to each other and then be able to transfer that data at
in that it can flow back into our area as well. And I think the different layers at a higher level. Its going to be beneficial long
fact that Intel has focused on including more of the hardware term, but I dont think were really seeing the structure around
security into its chipsets is also a positive sign and helps us to it yet, although publishing the Industrial Internet Security
set the platform for the root of trust. Framework2 is going to go a long way toward that.
2. Industrial Internet of Things Volume G4: Security Framework IIC:PUB:G4:V1.0:PB:20160926. Published in 2016

SPECIAL FEATURE
EECatalog: With regard to IIoT security as well as for other EECatalog: What tools and practices need to be developed
elements, the model referenced is often that of layerswill today to keep the things of the Industrial Internet of Things
this model continue to work? secure?
Smith, WinSystems: Yes, the layered model has been shown Smith, WinSystems: One of them is system management.
to be effective and beneficial for over 30 years in the networking When you have devices in the field you have to have a way to
arena. Applying this model to security and the IIOT provides manage them. Depending upon the requirements, this could
a similar abstraction from the complications and details that be aggregated to a single pane of glass cloud-based solution
each layer provides. This allows customers to quickly integrate or it might have to be an on-site hands-on management-
security features from trusted partners while maintaining based solution.
IOT SECURITY
focus on their own special requirements.

We know that the closer you have to get to a device, the more
For example, if we integrate a security layer from a trusted it costs. If you have to go up and service the device at the top
partner like Intel McAfee, the system security will be improved of the windmill, it costs more than if you can do it from your
without much extra complexity or burden to the customer. browser, so remote system management is very important.
WinSystems has a very similar role. We build very trustable Security detection is also very important. You need to have
hardware. We put on top of that very trustable BIOS; we some level of monitoring to understand when an attack or
put on top of that very trustable security layers, network breach has occurred. There is a tremendous amount of malware
layers, and Board Support Packages [BSPs] and Operating in the world, detection of that is going to be an ongoing
Systems all the way up into where the user runs their specific problem for decades to come.
application. One of the benefits the layers model gives us is
that users can almost be a la carte with what they do and Finally, once you detect an issue, you have to be adaptable
dont need. If they dont need a secure network layer from enough to correct the problem. You have to be able to correct
us, then that can be provided through their own existing problems and restore the system to a trusted state. That can be
capabilities. Depending on whether they need, for example, something as simple as changing passwords or as complicated
a unique identification key for each board or a shared key for as pushing a complete new software image, a more secure, not
each board or for their family of boards, we can abstract that corrupted software image into the device remotely.
in a different layer of security. Yet the layers above and below
dont have to change dramatically. Security needs to be included early, right-sized, and cost-
efficient because it will require attention for the life of your
The layered models give you the flexibility to adapt to the product. If you deploy security measures and think youre done
various needs of a customer without having to start over and and out the door, and youre forever secureWell, theres no
redevelop everything. There are substantial benefits there in such thing as forever secure.
terms of speed of execution, speed of production, re-using
tested code, and cost efficiency.

SPECIAL FEATURE
Delivering Carrier Grade OCP to
Telco Data Centers
Carriers as well as web service companies have some things in common
(and some differences) when it comes to capitalizing on a common set of
hardware solutions.

By Todd Wynia, Artesyn Embedded Technologies
Many communications service providers (CSPs) are looking t These servers typically have individual AC-DC power
to adopt COTS hardware and virtualize many of their supplies, increasing cabling and power costs and limiting
applications, deploying these diverse workloads on a common the ability for centralized management.
pool of hardware resources (Figure 1). The potential savings
of COTS computing and networking hardware is also creating t A short supply of real estate is available, curtailing attempts
great interest in the latest data center innovation: open to increase functionality.
compute technologies.
Facebook wanted a new rack-scale solution in which all servers
Pioneered and promoted by the Open Compute Project (OCP), would be identical no matter what company manufactures
these technologies focus on the most efficient and economical them. Servers needed to be powered, plugged into the rack,
ways of scaling COTS computing infrastructure. Founded by and cabled in the same manner. Determined to remove
Facebook, the OCPs original objective was to guide, from anything that didnt contribute to efficiency, Facebook even
the ground up, design of the most cost-efficient data center had manufacturers remove server faceplates and other
infrastructure. metalwork, choosing to handle the regulatory EMC shielding
at the facility level instead of at the server or rack level.
Today, the goal of the OCP is to spark a collaborative dialog and

effort among peers on OCP technology, collectively developing
the most efficient computing infrastructure possible. Project
focus includes addressing servers, storage, networking,
hardware management, Open Rack (a rack standard), data
center design, and certifications for solution providers.
Figure 1: Virtualization enables CSPs to run many applications

on more cost-effective standardized, multi-vendor hardware.
Deploying cloud technologies on this COTS hardware enables
greater agility in service delivery.
Facebook sought a new rack-scale architecture that would

use generic servers to make its new data centers as low cost
and efficient as possible. The company viewed traditional
rackmount servers as falling short in several areas.
Traditional Rackmount Server Drawbacks

t Each has a unique form factor, I/O, and management Figure 2: Front and back of a CG-OpenRack-19 system (CG-OPEN-
RACK-19 Image Courtesy of Pentair-Schroff).
system, locking a company into a single supplier.

SPECIAL FEATURE
Leveraging Open Compute and Open Rack Terms and Acronyms

Some applications must be hosted in central offices or t PODA logical and/or physical collection of racks within
similar environments at the edge of the network. And many a shared infrastructure management domain.
CSPs want to use the principles of the OCP in their network
infrastructure, but the specification doesnt lend itself to CSPs t POD ManagerThe software that manages logical
maintenance or other equipment practices. groupings of functionality across all infrastructure in a
pod.
So, a group of carriers and technology vendors have
collaborated to leverage Open Compute and Open Rack as t RMMRack Management Module. A physical system
a base model, but to adapt them for telecom central office element that is responsible for managing the rack, which
environments and carrier grade building practices. The normally assigns IDs for the instances of PSME in the
result is the CG-OpenRack-19 specification, which has been rack, and manages rack power and cooling.
designated OCP-ACCEPTED.
t PSMEPooled System Management Engine. System
CG-OpenRack-19 is a scalable carrier grade rack-level system management software that runs on the DMC and is
that integrates high-performance compute, storage, and responsible for the conguration of pooled storage
networking in a standard rack (Figure 2). CG-OpenRack-19 modules by the Pooled Node Controller (PNC), the
brings the OCP to carriers, tracking (but de-coupled from) network (SDN) the compute modules, and the switches.
changes driven by web companies and allowing compute,
storage, and acceleration to scale independently. The capital t MMCModule Management Controller. The controller
expense (CAPEX) is driven down by flattening the supply chain, that manages the blades in the module.
using OCP economies of scale and driving competition through
an open source specification, while OPEX benefits from lower t BMCBaseboard Management Controller. A specialized
power consumption and a reduced maintenance overhead. service processor that monitors the physical state of a
computer and provides services to monitor and controls
Functional Elements certain compute/ storage module operations.
There are six major system elements (Figures 3 and 4):
t MEManagement Engine. A physical hardware resource
1. System rack (19-inch) that gives access to hardware features at the baseboard
2. Power conversion and distribution via dual 12V bus bars level below the operating system.
3. White-box top-of-rack switches for optimized cable
handling t BIOSBasic Input/Output System. Firmware that
4. Two sizes of open bays for compute and storage elements initializes and tests compute/storage module hardware
(full- and half-width) components and loads a boot loader or an operation
5. A sled can be full- or half-width, each of which includes a system from a mass memory device.
single optical header on the back for connectivity
6. Pre-wired blind mate optical backplane
CG-OpenRack-19 Specication
Management Strategy
The CG-OpenRack-19 specification requires
each sled to have a dedicated baseboard
management controller (BMC) for various out-
of-band platform management services, which
is fully IPMI 2.0 and DCMI 1.5 compliant.
The specification goes on to stipulate certain
conditions that the BMC should meet,
but implementation details are left to the
developer.
Beyond the CG-OpenRack-19 specification

is another open, industry-standard platform
management specification and scheme
called Redfish, managed by the Distributed
Figure 3: The anatomy of a typical CG-OpenRack-19 compute sled.
Management Task Force (DMTF). See Figure 5.

SPECIAL FEATURE
Figure 5: Redfish pod logical hierarchy. See terms and acronyms.
field replaceable units (FRUs), carriers with many more physical

sites need to reduce costly on-site installation, maintenance,
and repair activities. These costs are being eliminated through
simpler hardware design. One example of such design is sled-
level field replaceable units with a minimal-touch environment
for service operatives.
Other differences found when comparing CG-OpenRack-19

and OCP include the following:
Physical:
t Suitable for current central office and new telco data center
environments
t 19-inch rack-mounting defined (versus 21-inch in OCP)

Figure 4: Functional elements of the CG-OpenRack-19 systems.
t Standard rack unit (RU) spacing
Redfish is a hierarchical pod/data center management tool,
where a pod is a pool of compute resources. It recognizes t 1000 to 1200 mm cabinet depth, supporting GR-3160
that the scale-out hardware usage model differs from that of spatial requirements
traditional enterprise platforms and requires a new approach
to management. t EMI shielding at the sled level
While Redfish is not a part of the CG-OpenRack-19 specification, t Terabit-capable blind mate optical cabled backplane with the
it is one of the most popular platform management approaches ability to individually hot swap sleds
and is a requirement for a solution to be Intel Rack Scale
Design (Intel RSD) compliant. t Consistent hardware user interface across different vendors
to shorten the learning curve
Comparing CG-OpenRack-19 to OCP
There are some key differences between the OCP specification t Option for central office seismic, acoustic, and safety
and carrier grade OCP as implemented in CG-OpenRack-19. For standards (NEBS)
example, while web companies are comfortable with rack-level

SPECIAL FEATURE
System Management:
t Ethernet based out-of-band (OOB) device management

network connecting all nodes and power shelf via a top-of-
rack (TOR) switch
t Ethernet based OOB application management network

connecting all nodes via a TOR switch
t Optional rack-level platform manager

Networking/Interconnect: Figure 6: CG19-GPU Sled from Artesyn Embedded Technologies
t One or more Ethernet TOR switches for I/O aggregation to Working with the OCP and other industry bodies, companies
nodes like Artesyn have developed solutions that will meet
CSP requirements with standardized architectures such
t Pre-cabled designfiber cables in rack, blind mate to node as CG-OpenRack-19 (Figure 6). Through a performance-
with flexible interconnect mapping optimized solution that maximizes data flows to virtualized
applications while maintaining high reliability, CSPs in the
t Durable, blind-mating coupling connectors provide for future should easily and confidently be able to implement
rapid insertion/extraction and prevent accidental damage open compute solutions.
or incorrect placement
These solutions will enable open compute successes not
One Infrastructure. Any Workload. only for their enterprise data center needs, but also for
CSPs are embarking on an exciting period of business NFV solutions and a range of new innovative services that
transformation. The ability to use high-volume COTS servers will help them better compete in the cloud provider and
to implement cloud technologies, such as virtualization and communications industries.
OpenStack, will help them reduce CAPEX and OPEX, unleash
new flexibility and elasticity in their operations, and radically Todd Wynia is Vice President of Communications
improve their time to market for new services. Products for Artesyn Embedded Technologies. He
has written a number of white papers on indus-
While OCP provides an excellent solution for the enterprise try standards and the telecom industry as well as
data center, CSPs require a higher grade of hardware platform serving on the board of CP-TA, VITA and partici-
designed to meet their more challenging needs for low- pating extensively in the PCI Industrial Computer
latency performance, bandwidth scalability, reliability and Manufacturers Group (PICMG). Wynia is a graduate of the Uni-
serviceability, and regulatory and safety compliance. versity of Wisconsin, where he earned his B.S. in economics with
a math emphasis.
Considering how critical and specialized their current
equipment is, the shift requires a carefully managed transition. Editors Note: More evidence of CG-OpenRack-19 benets is de-
tailed in this articles online version.

SPECIAL FEATURE
Jumpstart Connected Strategies
Remote monitoring reduces costs, improves uptime, and jumpstarts
connected strategies. Integrated platforms enable a flexible path,
protecting both edge and on-premise computing options.

By Ed Trevis, Corvalent
Application-ready platforms act as building blocks to the

Internet of Things (IoT), providing a secure development path
that speeds time-to-market and assures flexibility as needs sway
between centralized and de-centralized computing. Tapping into
these platforms to enable remote monitoring and management
capabilities is an ideal first step toward an IoT-based future.
To streamline the overwhelming nature

of creating an IoT system strategy, OEMs
and developers should address a primary
use case such as remote monitoring and
management as a first step.
The IoT can present a number of roadblocks for OEMs and

Figure 1: Based on Intel x86 processors, Corvalents CorEdge box
application developers. The impact of committing to network PC product family offers a powerful, exible, and reliable platform,
technologies, development platforms, cloud providers, and helping developers reduce time-to-market, protect application
more, means that a misstep could become a costly setback. security, and support reliable performance. Each application-ready
system can be customized to the individual needs of the project,
Such a misstep might not even reveal itself until an application including software pre-installation, hardware installation and con-
needs to scale or a security breach demonstrates a performance guration, and system branding. Systems can be shipped ready to
limitation. These are the kinds of factors that keep technology plug-in to manufacturing facilities, or directly to end users in non-
branded packaging. (Image credit Corvalent)
innovators awake at night, even while they are fully aware
of missed opportunities resulting from a lack of real-time
connectivity between systems and applications. The win comes from accessing and sharing that data in real-
time. If a system is running hot due to environmental issues or
To streamline the overwhelming nature of creating an IoT looming component failure, proactive measures can be taken
system strategy, OEMs and developers should address a only if operators are made aware of the rising issue. In another
primary use case such as remote monitoring and management example, a system may be performing well and even exceeding
as a first step. By capitalizing on integrated hardware/software its anticipated quota of scans or cycles; consider a high-
platforms optimized for flexible development, the path to performance MRI machine quickly reaching a milestone for
connectivity is both straightforward and secure. number of scans completed. Ideally it should warn operators in
advance that maintenance is soon required, before unexpected
Tapping into Remote Monitoring downtime costs hospitals thousands in unrealized revenue.
Accessing remote monitoring and management capabilities is a
good place to start, with the potential to deliver fast value from An application-ready integrated platform enables OEMs
an IoT investment. While deployed systems represent a wealth to configure these remote monitoring applications quickly,
of potential data, many are not yet equipped with sensors and adding sensors to aid in predictive maintenance operations
applications that enable data to be gathered and shared in a for the end-user. Integrated platforms may also ship with
timely fashion. Even if performance data is currently being sensors on board, enabling remote monitoring out of the box.
gathered, it may not be actionable quickly enough to support
predictive or preventive maintenance.

SPECIAL FEATURE
Solving Risk Factors to be an afterthoughtit is typically handled in the opposite

Achieving system longevity is an important part of the design fashion, with developers having to make choices about which
goalsaving millions of dollars over the life cycle of a product protocols to turn on.
based on reduced costs in hardware and software compatibility
testing, qualifying and validating new platforms, and agency Security faults arise when the development takes place, and
re-certification. With IoT technology investments, embedded security decisions follow. It is much more effective to incorporate
OEMs and developers face additional risk, as the pendulum a secure mindset as part of the development effort itself. This
constantly swings between centralized and de-centralized allows the system to be secure and interoperable at every layer.
computing. Solving this significant challenge involves working
with an application-ready platform that is inherently flexible. Developers in Mind
Developers can embrace any type of IoT environment, tuning An optimized platform is built with developers in mind,
system performance to the edge or on-premise without a using open architecture and a full range of communication
change in architecture. Ideally, software can be deployed in interfaces. Most interface protocols are available, as well
any manner that makes sense for the application at hand. as open source protocols and APIs for communicating with
devices or enterprise systems. Developers are empowered to
This advantage allows developers to reduce risk dramatically, focus on the business problem at hand, building their solution
removing difficulty by building their applications on what could as opposed to building a low-level infrastructure.
be called infrastructure middleware. Across the full spectrum
of embedded applicationsindustrial automation, medical, Your IoT Initiative
defense, energy, and moremost technology leaders compete Tackling a single mission such as accessing remote monitoring
best by focusing on their business of managing large devices or and management capabilities is a smart strategy, right-sizing
systems, not building core level platform middleware. the overwhelming proposition of becoming IoT-enabled. These
capabilities not only reduce maintenance costs, but also make
Addressing Pain Points and Simplifying Deployment OEMs more competitive for their end-user customers. Because
It is true that engineers love to build things, but their bosses deployed systems can be supported remotely, costly on-site
need to build the bottom line. Given that time-to-market is visits to either repair or simply diagnose systems are vastly
a critical factor in this effort, application-ready platforms reduced. Uptime is improved with more proactive options, and
address a range of options that improve the balance between overall support resources are significantly better managed
costs and development resources. based on predictive and preventive maintenance strategies.
End-users want to put software where they need it, and avoid Integrated hardware/software platforms provide the
being tethered to one cloud provider vs another. Rather than necessary toolsas well as the freedomto enable this value
committing an infrastructure build to a cloud service such as for developers. Faster time-to-market, reliable performance,
Amazon, Google, or Azure, an integrated container-based IoT and a security-first perspective are bundled in a flexible
platform can be deployed and moved as needed. Edge components system that accommodates both edge and on-premise
act as self-contained IoT systems, and can run completely computing. Investments are protected with longevity in
disconnected from the cloud or the on-premise server. mind, capitalizing on the remote management platform
as a flexible IoT framework on which to build additional
Improved Reliability with Security at Every Layer connected applications.
This flexibility and independent capability also protects uptime,
for example in scenarios where a rugged system may not have This initial step can open the door wide for whats next.
connectivity. The device continues to operate under the rules The ability to create IoT systems in a fast, scalable, secure
running at the software edge, an important value add for non- manner may be just what it takes to move your organization
stop applications such as manufacturing or hazardous security. into the future.
Security innovations capitalize on options built into Ed Trevis is President and CEO, Corvalent. Trevis
processors such as hardware-accelerated encryption; today, has been Corvalent President and Chief Execu-
a range of features is available below the operating system, tive Ocer since the companys inception in 1993,
creating comprehensive capabilities that enhance productivity leading the rm to double-digit growth, rapid
and secure management. In an integrated system, secure gain in market share, and numerous business
hardware dovetails with secure software. The perspective is awards and recognitions. He is an active CEO
security first, with protocols architected into the platform at member of Vistage International, and maintains
every opportunity. Everything is locked down by default, and a leadership philosophy promoting employee edu-
developers must specifically open ways to access the system. cation and encouraging personal and professional growth. Connect
This is in contrast to ground-up solutions, where security tends with Ed via LinkedIn or at ed.trevis@corvalent.com

SPECIAL FEATURE
What Makes an Industrial Digital
Media Player Different?
With digital displays popping up well outside their climate-controlled
comfort zone, a new kind of media player is increasingly in demand.
DIGITAL SIGNAGE/SMART DISPLAYS

By Robert Suffoletta, Logic Supply
The digital signage industry is booming. Analysts predict an But what makes an industrial digital media player different,
8.94 percent compound annual growth rate1 between now and what factors play into an educated hardware choice for
and 2020, and that by 2023 the market will be worth $32.84 signage integrators and ISVs?
billion2. The expansion of signage implementation throughout
industry has been a major contributor to this exponential A Different Look and Feel
growth, with displays being used by businesses of every shape In the past, theres been an attitude toward disposability in
and size for a huge range of content delivery needs. Of course, much of the digital media player space, with entry level device
Digital Out of Home (DOOH) advertising is still a significant builders suggesting that when a media player fails, the user
part of the overall signage equation, but increasingly digital simply throw it away and replace it. But that assumes a certain
displays are being utilized for applications far beyond level of nonchalance toward the information being displayed.
customer acquisition, and in locations that would challenge, Industrial digital signage players arent throw away devices
or even destroy, a typical media player. because the content theyre displaying isnt disposable. In fact,
its often mission-critical.
Perhaps the most striking difference between consumer-

grade digital media solutions and industrial media players
is the way theyre constructed (Figure 1). The vast majority
of commercially available media players are built using
some combination of plastics and polycarbonate. While
these materials are relatively inexpensive, they dont offer
much in the way of durability, particularly in challenging
environments. Industrial media players, which may be subject
to extreme temperatures, moisture, vibration or even impact
forces, most commonly utilize all metal enclosures and
internal components designed specifically for industrial use.
Figure 1: Unlike traditional media players, fanless industrial media

players, like this ML400 system from Logic Supply, employ custom
machined heatsinks and fanless enclosures.
Static and interactive digital displays are popping up

everywhere from medical facilities and manufacturing floors,
to transportation hubs and outdoor events (Figure 2). These
varied applications for digital content delivery bring with
them a host of logistical and environmental complications
Figure 2: An industrial digital signage installation located above the
that have required signage professionals to reevaluate the JetBlue terminal at JFK airport in Queens, NY. Industrial digital media
hardware they utilize to convey their message. players inside power two displays each. (Courtesy Logic Supply)
1. http://www.tmcnet.com/usubmit/2014/03/12/7720094.htm
2. http://www.marketsandmarkets.com/PressReleases/digital-signage.asp

SPECIAL FEATURE
Industrial media players may also look very different than their interior menu boards are increasingly turning to fanless
their commercial counterparts. With enclosure extrusions solutions to prevent dust, grease, and moisture from reaching
designed to dissipate heat, and form factors tailored to sensitive hardware components and causing failures.
installation behind low profile displays or within electrical
cabinets, industrial signage devices dont always conform Constructed for the Long Term
to the nondescript black box archetype. The connectivity One important aspect of industrial signage that frequently
options available on industrial media players also tend to take goes overlooked is the concept of life cycle. While the
into account the variety of output devices they may need to reliability of the hardware contributes to its life span (how
interface with. In addition to HDMI, USB, and DisplayPort (of long its expected to operate without a failure), life cycle
various flavors), its not uncommon to find VGA or even DVI refers to the manufacturers commitment to produce and
connections to accommodate legacy displays that may be part support a given device.
of the existing infrastructure at a given installation location.
In the world of consumer technology, frequent hardware
Purpose-built Reliability turnover due to obsolescence makes life cycle less of a concern,
A hardware crash at a retail facility utilizing a digital signage but for industrial applications that may depend on a system
solution might be inconvenient or embarrassing for the to operate for three to five years or more once installed,
proprietor, but imagine the potential ramifications to loss the ability to order additional devices or get support for
of signal at a high-paced manufacturing plant or, worse any necessary updates is paramount. When dealing with
still, a busy medical facility. Reliability is paramount in any international or safety certifications, even slight changes to a
industrial signage application, making attention to detail device configuration can cause huge logistical headaches, not
and careful engineering of industrial media players vital to to mention the significant costs of re-certification. The ability
their longevity. For signage integrators installing hardware at to order a locked-down configuration for the foreseeable
client sites, a single failure that results in the need to roll a future of a project is a huge advantage industrial media players
support truck can cost the company hundreds of dollars, not offer over their consumer-grade counterparts. In addition to
to mention the potential lost revenue for its customer. life cycle management, industrial media player manufacturers
frequently offer more inclusive and longer warranty support,
With outdoor signage deployments becoming more providing additional peace of mind to signage integrators.
commonplace, systems designed for industrial use are
employing components rated for extreme operating The Bottom Line
temperatures, from -25 C (-13 F), all the way up to +70 C The continued evolution of digital content delivery is changing
(+155 F) or more. In addition to outdoor use, these wide the way signage professionals evaluate the hardware platforms
operating temperature ranges provide integrators the they utilize. Even the most user-friendly, fully featured
flexibility to install systems in cars, busses, trains, and ships software suite is still only as viable as the hardware its running
where theyre commonly used for everything from passenger on. The widespread use of digital media players in increasingly
information delivery to infotainment and wayfinding. challenging environments puts pressure on integrators to
ensure the hardware they select will survive the rigors of
In addition to environmental resistances, many industrial installation, no matter where that might be. Ultimately, the
signage players limit, or even completely forgo, moving parts. quality of your industrial digital signage hardware should
Solid state storage offers faster read and write speeds to always match the gravity of your message.
allow for smooth content delivery while also eliminating the
noise and data corruption that can result from spinning hard Robert Suoletta is a Visual Communications
drives. The most reliable breed of industrial media players Specialist at Logic Supply (www.logicsupply.com),
also leverage fanless, solid state cooling solutions. Removing an Intel IoT Solutions Alliance member. A 20-year
a cooling fan from the equation can result in a digital media veteran of the computer industry with more than
player with zero moving parts, greatly improving overall 10 years experience in providing hardware solu-
system reliability, especially for installations where the tions to the digital signage market, he is dedicated
hardware is subject to vibration. to matching clients building innovative signage solutions with the
most capable and reliable hardware for their unique installation.
Above and beyond fanless cooling, some industrial media
player hardware manufacturers take the extra step to create
systems that are also fully sealed against contaminant ingress,
with no vents or extraneous openings in the enclosure. Quick
serve restaurants who employ digital signage displays for

SPECIAL FEATURE
System-on-Chip: Not Yet
Ready for Prime Time
Factors involved in weighing SoC, Integrated OPS, and external hardware
choices for digital signage

By Mark Boidman, Ben Zinder, and Gilbert Baltzer, Peter J. Solomon Company
System-on-Chip (SoC) is an integrated digital signage solution often faced a substantial learning curve when required to use
where media player hardware is embedded within a display, proprietary operating systems (e.g., Samsung Tizen OS, LG
eliminating the need for third-party, external media players. WebOS). Finally, adoption was hindered by skepticism about
These external media players are the commercial equivalent of long-term commitment and focus of manufacturers to what
consumer products such as Roku, Chromecast, and Amazon Fire. will always be for them a relatively niche product.
Initial System-on-Chip Products Had Limited Subsequent generations have addressed some of the issues
Capabilities encountered by early adopters. Improved hardware offerings
First-generation SoC products from Samsung and Sony have greatly improved the functionality and flexibility of
launched in 2013, followed by second generation displays and SoC displays. New entrants into the market (Panasonic,
the entrance of LG into the market in 2014. First- and second- Philips, Sharp, Toshiba, etc.) have adopted Android as a
generation products had minimal processing and graphics standard, and Samsung and LG have made their operating
power, which restricted capabilities to playback of images systems more developer friendly. However, Android remains
and basic video. Content creators and software developers a consumer-grade operating system and the proprietary OSs
(Photo: Samsung)

SPECIAL FEATURE
are derivatives of consumer TV platforms. They are primarily Another Option for All-in-One
based on the world of streaming, and theres not a lot of focus A different type of integrated solution is gaining momentum
on software development. The core hardware is good, but the in the industry. Displays and players that have adopted Intels
software ecosystem is changing all the time. That makes life Open Pluggable Specification (OPS) are offering the Pro AV
difficult for end-users and developers. market a solution thats both powerful and integrated. The
combination of an OPS display and a commercial-grade OPS
What the industry needs are very stable APIs because so many media player delivers an all-in-one signage solution that is
software developers are developing on these platforms. When powerful and feature-rich while still being simple to install,
the software changes with every new version of the consumer eliminating the need for wiring and technical expertise. This
TV models, development is quite difficult. OPS display/player combination can offer all of the benefits of
a commercial-grade media player, a purpose-built OS, stable

System-on-Chip Now Offers Some Advantages software with open APIs and an all-in-one solution without
The primary selling point for SoC is that the upfront cost any of the downfalls.
is approximately 25-40 percent lower than comparable
solutions. Users also typically experience a reduction in To Date, SoC Has Struggled to Gain
hardware support expense due to fewer points of failure. Widespread Adoption
Power consumption is usually lower, as an integrated design Despite most display manufacturers now bundling SoC at
improves energy efficiency. In addition, with no need for minimal incremental cost, it is estimated that fewer than
configuration, the installation process is straightforward and 10 percent of displays with SoC have their capabilities used
requires less technical expertise. by operators, who still prefer external hardware solutions.
Nevertheless, SoC products have improved significantly since
Most display manufacturers require screens to have constant market entrance and offer a compelling alternative in some
Internet access both to stream data and to download software circumstances.
updates. While this connectivity is convenient, it also creates
a potential security concern for corporations with large Ultimately, the decision to purchase external hardware, an
installations consisting of hundreds of displays. integrated OPS solution, or a SoC display must be made on
a case-by-case basis. For many basic digital signage needs,
External Hardware Remains Essential in SoC may very well be an adequate solution. Customers who
Certain Situations place a premium on performance, reliability, and scalable
Despite recent improvements, SoC devices continue to face software will likely continue to rely upon external hardware or
some limitations in their ability to support the breadth of integrated OPS solutions for the foreseeable future.
features needed for many types of digital signage projects.
Consequently, external media player solutions remain the Mark Boidman is a Partner and Managing Director
solution of choice for businesses aiming to ensure seamless at Peter J. Solomon Company.
transitions between media content, complex video layering
and HTML rendering, or synchronization between multiple
displays. The primary advantage of dedicated external media Ben Zinder is an Associate Director at Peter J. Solomon
players is the breadth of the solutions software offering. Company.
External hardware also accommodates applications that

vary from touch screens to complicated display solutions Gilbert Baltzer is an Associate at Peter J. Solomon
and/or video walls with a range of display sizes mounted at Company.
various angles. An additional consideration for many buyers
is external hardwares upgradeability, the ease of which helps
to future-proof the display, not locking buyers into a specific
product ecosystem.

Docker Containers Ease Cloud and
SPECIAL FEATURE
IoT Implementation
Open-source Docker leverages Linux resource isolation features to deliver
lightweight, efficient, and repeatable software delivery on complex
infrastructure.
ANDROID & LINUX

By Lynnette Reese, Editor-in-Chief, Embedded Intel Solutions, Embedded Systems Engineering
Three items are key in IoT: a thing, a cloud, and the internet Create Precisely Defined Containers
for communication. Within this model, the cloud is not only Docker is a concise tool that containerizes a process
someone elses computer, but in reality, the cloud is a complex or application and isolates it from other applications. A
compilation of interconnected servers with variable hardware containerized app runs anywhere. Docker simplifies getting
infrastructure, software stacks, and other effects that can be an environment up and running on your machine and
combined in millions of ways if you count all possible settings. provides a container system for code and a very consistent
IT staff loves Docker since it removes the agony factor to a large way to get code running on a specific set up. Linux containers
extent when setting up or bolting on software components have been around for years, but many use containers as
into any infrastructure. Docker eliminates the need to become a kind of tiny server. Docker began as a tool for creating
a pseudo-expert in various packages and hardware settings containers for convenient, reliable, repeatable software
and isnt limited to clouds. Docker, an open source project, delivery and is meant to be as easy as possible to use. Docker
makes it easier to get software and hardware working together gets around the fact that software stacks today are run on
in an efficient, repeatable pattern, so developers dont have to different frameworks, switching between toolchains of
deal with the complexities of servers and storage1. Docker is a different languages, and running on increasingly complex
tool that allows us to innovate without reinventing the wheel and diverse hardware infrastructure. Docker allows people
and like Linux is the crux of why Docker is so successful. to share containers, which has become possible as a critical
mass of people that want to share and reuse containers pull
them from Docker Hub.
The result is a new means
of delivering software that
is lighter than a virtual
machine (VM) yet sidesteps
the complexity of setting up
the run environment for the
recipient. Docker.com states,
When an app is dockerized,
that complexity is pushed
into containers that are
easily built, shared, and run.
Setting up to work on a new
codebase no longer has to
mean hours spent installing
software and figuring out
Figure 1: Left: Virtual machine (VM). VMs run a resource intensive OS
setup procedures. Code that
and establish a configuration entanglement. Right: Containers can ships with Docker files is simpler; dependencies are pulled as
share a kernel. Only the executable and package dependencies are in neatly packaged Docker images and anyone with Docker and
a container image. Processes run as native and can be managed indi-
vidually with no configuration entanglements. (Source: docker.com)
an editor installed can build and debug the app in minutes2.
1. Levy, Ari. One of Techs Most Ambitious Open Source Projects Puts a Software Veteran at the Helm. CNBC. CNBC, 02 May 2017. Web. 31 May 2017
2. What Is Docker. Docker. N.p, 15 Apr. 2017. Web. 31 May 2017.

SPECIAL FEATURE
to replicate fully portable environments

between infrastructures so that dotCloud
could help customers. Sharing and assisting
others have become more complicated due
to multiple hardware platforms, appliances,
and other variables such as different
versions of software. Something that
works on your machine may not work on
anothers. A bundle of software typically
has a file extension that indicates a specific
ANDROID & LINUX
environment for unpacking and executing

it. However, if you ship that software
package, youre not sure what is on the other
end. You dont know the version or settings
of the recipients environment; all you know
for certain is that the software package
worked on your machine. Increasingly more
complex software stacks make it likely that
the recipients hardware, operating system,
and other environment variables do not
Figure 2: The entire container market (including containers, virtual-
ization, Private PaaS and others) is expected to hit $2.688 billion by
have everything that your software package
2020. (Source: 451 Research) used on your end. To solve this problem, you might have to
ship everything surrounding the application, too, down to the
Put Simply, How Does Docker Work? build, system configuration, and the version of the application
A good analogy for Docker is the shipping industry. In the and libraries.
olden days, individual items were packed on wooden ships and
individually unloaded by stevedores at the port of destination. One option is to re-create your entire system by re-creating the
As shipping got more complicated, at some point modular, environment with a virtual machine so that you have the exact
standardized shipping containers were implemented. Product environment for success for the recipient. But virtual systems
is placed in a container, sealed, put on the ship, and sent are very large, on the order of 10 MB or more, and if youre
off. A global infrastructure has adapted to this container testing a stack on 12 different virtual machines representing
including ships, portside loading and unloading equipment, 12 specific environments then youre going to need more than
train flatcars, tractor-trailer trucks, and even airplanes; all one personal computer to deploy 12 virtual machines. Virtual
are meant to accept this same container
regardless of location, company, or mode
of transportation. With Docker, the
software developer can hand off his or
her container to others to handle; others
who are experts in the infrastructure,
surrounding software, and tools. The
developer need be concerned only with
the inside of the box. Docker provides a
consistent way to get code up and running
on the developers machine yet can also
move code forward to staging, production,
or various testing scenarios.
Docker is valued at more than $1 billion

and is expected to grow in revenue to
$2.7 billion by 2020, according to 4512
Research3. Docker was born at dotCloud, Figure 3: https://docs.resin.io/understanding/understanding-
a platform-as-a-service company in 2013 out of the necessity devices/2.0.0/
3. Buckley, Kaitlin. 451 Research: Application Containers Will Be a $2.7bn Market by 2020. PRWeb. PRWeb, 10 Jan. 2017. Web. 31 May 2017.
4. https://docs.resin.io/introduction/ N.p. Web. Accessed 31 May 2017.

SPECIAL FEATURE
machines are terrific, but they do carry a lot of baggage and embedded development boards. According to the Resin.
overhead that can weigh down performance. io site, The ResinOS is an operating system optimised for
running Docker containers on embedded devices, with an
Docker is an open collaboration with all parties involved in emphasis on reliability over long periods of operationthe
creating a specific image for a specific stack and hardware core insight behind ResinOS is that Linux Containers offer,
infrastructure. Docker takes advantage of Linux evolving such for the first time, a practical path to using virtualisation on
that executing processes in a sandbox (a resource isolation embedded devices. 5
feature) can be achieved concisely because youre working
with low-level primitives in the Linux kernel. As far as Docker Docker is another stepping stone in navigating the next
knows, it is the only process talking to the Linux kernel and is wave of increasingly complex technology. Docker is open
ANDROID & LINUX

cleanly isolated. This means that you can deploy applications source and owes a great deal to the open source community
without worrying about conflicting dependencies with other that participates in creating Docker containers, however,
applications. Docker can run a specific library version, while Docker does have an Enterprise version with over 400 paying
another application runs another version of the same library customers. The Docker site includes case studies of Docker
without conflict. use at Uber, MetLife, Expedia, General Electric, Cornell, Ebay,
ADP, PayPal, and several others. As for the embedded world,
Docker for the Embedded Developer lightweight containers can make test and development faster
You can install Docker by downloading it (as a static binary) and can add to security as VMs have done for computers by
and installing it. Its initially kicked off as a daemon. After isolating applications.
the daemon is running, the client is run, and then commands
can be typed that get passed to the Docker daemon. Docker Lynnette Reese is Editor-in-Chief, Embedded
definitely supports the 64-bit x86. The Intel Joule development Intel Solutions and Embedded Systems Engi-
kit, well-suited to IoT applications, provides a Docker neering, and has been working in various roles
container and uses Docker Toolkit to facilitate it. Another as an electrical engineer for over two decades.
company apparently incorporates Docker into its operating She is interested in open source software
system. Resin.io, a company that makes it simple to deploy, and hardware, the maker movement, and in
update, and maintain code running on remote devices4uses increasing the number of women working in
Docker and claims support for the full Raspberry Pi family, STEM so she has a greater chance of talking about something
Beagle Bone Black, Intel Edison, Intel NUC, Odroid C1+ and other than football at the water cooler.
XU4, BeagleBone Green Wireless (beta), and several other
5. https://docs.resin.io/understanding/understanding-devices/2.0.0/ N.p. Web. Accessed 31 May 2017.

SPECIAL FEATURE
Open Source Meets Standards: A

Platform for Fast Implementation
Mini Linux platform utilizes COM Express modules.
ANDROID & LINUX
By Dan Demers, congatec
Like it or not, embedded development is getting more platforms arent appropriate for embedding in finished
demanding. Devices like the iPad and Surface have raised devices. Professional embedded system engineers lag behind,
the bar for the quality of the hardware, software, and user lacking comparable ecosystems.
experience. Accelerating investment in hardware start-ups
means there are a larger number of talented and well- One response to the gap between embedded engineers needs
funded competitors fighting over the same markets. With and ecosystem readiness could be to leverage a standards-
heightened expectations and competition, developing based Linux platform. Such a solution capitalizes on the vast
better embedded devices on shorter timelines is essential, ecosystem of open source software available to embedded
highlighting the importance of hardware standards and a devices One aspect that an environment like this excels at
comprehensive ecosystem! is in bringing together the best parts of WebKit for HTML
rendering and the V8 JavaScript engine. Out-of-the-box
they support audio, video, and 3D rendering with WebGL
as well as raster and vector artwork and UIs that reflow
to different screen sizes. Beautiful applications with rich
user interactions can be created with these technologies.
Moreover, these web technologies benefit from broad
support and a large development community. They surpass
other environments in the number of developers, support
resources, best practices and community activity. As it is
rare to find a task that hasnt been solved before, software
engineers can make efficient re-use of existing knowledge
and concentrate on innovating their products rather than
redeveloping code that others have already written.
Shrinking the Workload

It makes no sense for OEMs to use such environments if they
are not supported by embedded hardware platforms. It can
Figure 1: An example of a low-power consumption (5-7W) embedded take hours, days, or even months to debug insufficient driver
computing platform applicable to wired and mobile use. Depending
upon the processor choice (1 and 8GB RAM), it equates to consumer implementations or incompatible code. One company which is
laptop or small server. (Source: Embedded Now) helping application engineers avoid this potential workload is
Embedded Now. Its single board computer and ecosystem of
Today, most software engineers live in a world of open peripherals target HTML 5, CSS, and JavaScript applications
source, cloud computing, and rapid agile development. in Linux-based embedded systems. Embedded Nows systems
They can prototype applications in days by creating novel come preloaded with either Lubuntu, Ubuntu 14.04 LTS
combinations of powerful open source software, or launch a Desktop, or Ubuntu Server. Support for developers building
few thousand servers in a matter of minutes. So far, these applications with Electron, Chromium, or the Chromium
efficiencies have been confined to the world of software Embedded Framework is also available. Features make the
development. They havent translated over to professional companys Linux platforms suitable for embedded applications
embedded development. Hobbyists benefit today from include, for example, those, preventing corruption of storage
platforms like Raspberry Pi and Arduino, which have broad volumes, and solutions for rendering wireless networking
ecosystems providing real productivity gains, but these more robust in commercial and industrial settings.

SPECIAL FEATURE
With regard to peripherals, Embedded Now advises on
Wi-Fi or Bluetooth implementations and helps customers
develop specialized peripherals for tasks like analog to
digital conversion and motion control. As the companys
platforms are all standards-based, application engineers
on the customer side typically find that needed specific
peripherals work immediately out-of-the-box. Thats given
that the platforms use only standard interfaces (like USB
ports) and widely supported Linux distributions with
extremely comprehensive driver support.
ANDROID & LINUX

Figure 3: A tailored OS with all components provided by the plat-
form provider is one of the essential connections that conducts the
communication between the applications and the hardware as the
major abstraction layer. It relies on comprehensive driver support
from the embedded board vendor.
to +100 C. All components and engineering are certified for

industrial use. The components, like the wide range locking
power connector, have all been chosen to withstand the most
demanding environments.
Customers can pick the appropriate Computer-on-Module with

Figure 2: The Embedded Now Piconium embedded computing plat- COM Express Type 10 pinout for scaling the performance from
form, shown here with a different cooling solution than that pictured
in Figure 1, provides multiple interfaces including 6x USB 2.0 and 1x 3rd Generation single-core and multicore Intel Atom, Celeron
USB 3.0 for application specific extensions including Bluetooth and processors up to modules featuring the latest 5th Generation of
Wi-Fi as well as 1 HDMI port. (Source: Embedded Now) Intel Atom, Celeron and Pentium processors (codenamed Apollo
Lake). Both processor technologies support the extended tem-
Describing the speed with which a standards-based Linux perature range and offer a long-time availability of at least seven
platform can get an embedded product to market, Eric Mar- years, to match embedded application needs. With a low-power
thinsen, Chief Software Architect and Founder of Embedded envelope of less than 12W, fanless designs can easily be achieved
Now, notes, Our customers usually go from having nothing without sophisticated cooling solutions, reducing design-in
to shipping production-ready devices in between three and efforts. The broad prevalence of the Intel Atom technology makes
twelve months. Anecdotally, we are hearing that people are strong Linux support out-of-the-box possible. The memory is
shaving around twelve to eighteen months off of their engi- configurable for all versions from 1GB to 8GB.
neering schedule.
congatec module solutions
Modular with Linux Support from the Get Go Embedded Now chose the COM Express standard because it
To expertly serve customers with application ready Linux offers the broadest scalability for future developments up to
platforms requires concentrating on core competencies and server grade platforms, and picked congatec as its module vendor.
finding the right ecosystem partners. For its commercial off- We are very happy to have chosen congatec, says Marthinsen.
the-shelf (COTS) single-board computer, called the Piconium, Prior to founding Embedded Now, my business partner and I
Embedded Now has chosen a modular approach on the basis were customers of congatec. We selected them due to the quality
of the COM Express Computer-on-Module standard. This of engineering that goes into their modules and the high level
ensures the reliability customers demand. The FCC and CE of support they provide. We consider them to be an important
compliant platform, which can be utilized from mobile client strategic partner and a key component of our success.
devices up to micro server applications for edge computing, is
based on a COM Express Type 10 module and measures only Dan Demers is the Director, Sales & Marketing
130 x 55 mm. Its small size means it can fit into whatever nook congatec America. He holds a B.B.S degree in
or cranny applications call for. Despite its small footprint it International Business from Grand Valley State
offers a comprehensive set of interfaces, including 2x USB 3.0, University, Grand Rapids, Michigan and an
5x USB 2.0, Ethernet, and HDMI as well as Wi-Fi and Blue- M.B.A. from Ashford University, Clinton, Iowa.
tooth via USB extensions, and LVDS displays and cables with Mr. Demers has over 19 years of experience in em-
touch panel support as accessories. bedded computing having worked with Fortune 500 companies in
the Industrial, Medical, and Communications markets.
The Piconium can be used in nearly all applications as it can
operate even in the extended temperature range from -40 C

Product Showcase
Intel E3800 Series Edge-Connect Architecture
Applications
Unmanned drone/robotics (air, surface, underwater)
for leak detection, security, agriculture, science
research, and energy
Mobile computing for industrial IoT, payload/mission
computers, intelligent controllers or datalogging in a
variety of rugged environments.
Portable healthcare instrumentation and equipment.
Man-Wearable Computing...especially in rugged use
scenarios.
At just 75mm x 75mm the ADLE3800SEC is ideal for rugged,

or extended temperature use in a variety of industries Mini Embedded PC: ADLEPC-1500
including: military, rugged industrial, unmanned, energy, Dimensions: 40mm x 87mm x 87mm
transportation, medical or security and surveillance. I/O: 2x 10/100/1000 LAN; 1x USB2.0; 1x
Edge-Connect Architecture provides easy expansion and USB3.0
helps reduce cabling, integration time and system size all 1x DisplayPort
while increasing quality and overall MTBF. Options: 1xM.2 KeyB socket for PCIex1 or
SATA SSD modules
ADLE3800SEC: E3845 Quad, E3827 DC Microsoft Azure Certified for loT
Front-Side I/O
2x 10/100/1000 LAN
1x USB 2.0 ADL Embedded Solutions
1x USB 3.0 855-727-4200
1x DisplayPort sales@adl-usa.com
Microsoft Azure Certified for loT
Embedded Intel Solutions

Designing with delivers in-depth product, technology
Intel Embedded and design information to engineers
and embedded developers who design with
Processors? Intel Embedded processors
CA
I Santa Clara,
April 26-27 vCon.c om
www. IoT-De
NOW !
REG ISTE R
Spring 2017
Industrial IoT
Winter 2016
Strength VRs Next

Brakes
Securing
Medical IoT
Wave
Putting the ware
on Ransom
The Case
for
s: Edge Analy IoT
Autonomou s tics
ve
Intel GO Pa y Natural La
the Wa ng
Processing uage
into Its Ow Comes
n
edd edin tel.com www.em
bed
www.emb dedintel.com
Gold Spon
ors sors
Gold Spons
Subscribe Today at
Visit www.embeddedintel.com
www.embeddedintel.com Free!

LAST WORD
The Eyes of the IoT Rely on

Processor Innovation
Manufacturers are increasingly using connected nodes to monitor production
on the factory floor, as well as for surveillance, traffic, and retail applications.
By Caroline Hayes, Senior Editor, Embedded Intel Solutions
Seeking processing power that can boost IoT board and blade been available on desktop Personal Computers (PCs) and larger
performance, manufacturers are eagerly integrating upgrades devices to this small form factor.
in microarchitecture.
The CL has a PCIe x16 EPIX Camera Link card, and the CXP
Earlier this year ADL Embedded announced the ADLVIS-1700 configuration has PCIe x16 Euresys two-port CoaXPress
CoaXPress/CameraLink vision system. The prototype debuted framegrabber. There are four removable Serial Advanced
at Embedded World, in Nuremberg, Germany. At 5.0 x 7.0 x Technology Attachment (SATA) drives (two SATA II and two
7.6-inch, it is claimed to be the smallest CoaXPress system in SATA III) in each, and a single cable to provide power while
the industry. It consists of a CoaXPress board inside the vision sending data up to 300 feet. Transfer speed is 1200MB per second
box (see Figure 1). The company has used the small PCIe/104 for image capture and storage. Target applications include
form factor, and believes it is the first to do so. 3D machine vision for factory automation, hyperspectral
and multi-spectral imaging, traffic surveillance, security
Compact and Connected monitoring and control, military and defense Intelligence,
Both the ADLVIS-1700-CL Camera Link configuration and the Security and Reconnaissance (ISR), unmanned drone vision,
ADLVIS-1700-CXP CoaXPress configuration are based on an and high frame-rate motion analysis and recording.
Intel architecture (formerly known as Skylake) central processor
unit (CPU), the QM87HD Intel Core 6th Generation i7-4700EQ Industrial Controls
with 8GB DRAM. At Embedded World, the company explained The company also uses dual and quad core Intel E3800-Atom
that Skylake combines security with hardware level TPM 2.0 processor options in the ADLEPC-1500 embedded PC, which
encryption technology and power virtualization capabilities. The is designed for unmanned, industrial controls, and robotics.
vision system also brings the speed and resolution that has only
The compact PC measures 1.3 x 3.4 x 3.2 inches and has a wide
voltage range of 20 to 30 VDC, a 24V nominal input (optional
7.0 to 36V) a temperature range of -20 to +50C, and an
extended range of -40 to +70 C, for operation in a variety of
harsh environments.
The PC is based on the ADLE3800SEC Single Board Computer

(SBC), shown in Figure 2, with either the dual core Intel Atom
E3800-3827 quad core Intel Atom E3800-3845 low-power
processors to support DirectX 11, Open GL 4.0 and full
high Definition (HD) video playback. There is also on-board
DisplayPort, USB 3.0, USB 2.0, M.2 KeyB 2242 SATA, and two
Local Area Network (LAN) ports.
Another company, ADLINK Technology, Inc., also uses the

former Skylake, 6th generation Intel processor for its cPCI-
6630 CompactPCI processor blade, which targets Industrial
IoT applications in factories, industrial automation, and
Figure 1: The ADLVIS-1700 Vision System comprises a CoaXPress security systems (Figure 3).
and Camera Link technologies.

LAST WORD
Seventh Generation Processors

The next generation of Intel Core processors, formerly known
as Kaby Lake, are exploited by congatec, in its industrial,
Thin Mini-ITX motherboard, the conga-IC175.
Figure 2: The ADLE3800SEC SBC has dual and quad core Intel Atom
E3800 options to support graphics technologies.
This also breaks ground, as it is believed to be the first

CompactPCI blade with the 6th generation Core i7 processor
in a 6U form factor. The blade integrates the quad-core 2.8GHz
Intel Core i7-6820EQ processor and Mobile Intel HM170 Figure 4: The conga-IC175 targets IoT applications, and High
chipset, with up to 32-GB of dual-channel Double Data Dynamic Range provides dramatic images.
Rate Fourth Generation (DDR4), Small Outline, Dual Inline
Memory Module (SO-DIMM) 2133 non-Error Correcting
Code (ECC) memory. The blade supports a 64-bit/66MHz This low-profile motherboard (Figure 4) features the 7th
CompactPCI bus. generation Intel Core U processors for IoT connected devices.
It targets space-constrained, low-power IoT designs, with a
Subscriber Identity Module (SIM) card socket for 3G, 4G, or
NarrowBand connectivity.
With the selection of Intel Optane memory via an M.2

connector fast system boots, application starts, video
recording and processing, and software updates are enabled,
easing the throughput of IoT applications.
The Intel Core processors were chosen for use in embedded

and IoT applications. Martin Danzer, Director, Product
Management, congatec, said: The new, low-power versions
of the high-end 64-bit Intel Core processors offer a highly
Figure 3: The ADLINK cPCI-6630 is small and uses 6th generation
Intel processing technology to support graphics and connectivity
configurable Thermal Design Power (TDP) of 15W, with the
for Industry 4.0. flexibility to scale dissipation from 7.5W cTDP to 25W cTDP,
for an exceptional balance of power and performance.
For imaging, the front panel I/O includes a Digital Visual
Interface (DVI)-I and DVI-D port. DVI-I is for digital and The industrial grade, Thin Mini-ITX motherboard ships with
analog signals, and DVI-D is for digital only signals. The front four dual-core variants of 7th generation Intel Core U SoC
panel also has three Gigabit Ethernet (GbE) powers, three USB processors. It also follows the trend for multiple displays,
3.0, one USB 2.0 and an RJ-45 COM port. For storage, there is connecting to the DirectX 12-capable Intel HD Graphics 620
a 2.5-inch SATA drive and a seven-pin SATA connector as well with up to three independent displays in 4k resolution at
as onboard CFast and CompactFlash option. 60Hz via two DisplayPort (DP++) and embedded DisplayPort
(eDP) or dual-channel Low Voltage Differential Signaling
The Intel Core i7-6820EQ processor supports the High (LVDS). It also has HEVC support, while the addition of
Efficiency Video Coding (HEVC) video compression standard High Dynamic Range (HDR), makes video more vibrant
and the blade can support three independent displays when for a lifelike image quality, giving texture and depth for
mated with ADLINK Technology cPCI-R6002 or cPCI-R6100 surveillance and monitoring of areas or production lines.
rear transition modules.
Caroline Hayes has been a journalist covering
The blade requires 5V input, to provide value-per-Watt the electronics sector for more than 20 years.
computing power required by industrial applications, says the She has worked on several European titles,
company, with the computing performance of the Intel Core reporting on a variety of industries, including
6th generation processor for communications and monitoring. communications, broadcast and automotive.

Embedded Intel Solutions Fall 2017

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Embedded Intel Solutions Fall 2017

Enviado por

Direitos autorais:

Formatos disponíveis

FROM THE EDITOR

Digital Signage Computing

Vice President & Publisher

Applications dont stop at digital signs and kiosks.

2 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

FROM THE EDITOR DIGITAL SIGNAGE/SMART DISPLAYS

25 System-on-Chip: Not Yet

12 When Adding IoT Devices to the

14 We have to coordinate that root of 32 ADL Embedded Solutions

4 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

The Passenger Economy:

By Lynnette Reese, Editor-in-Chief, Embedded Intel Solutions

Early Autonomous Vehicles (AVs) are here but pose grave

Autonomous Vehicles to Save $1.3 Trillion a Year

Once people cease to drive, what do they do? A

6 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

in AVs and jobs are re-established in areas related less to the

What can we expect? Winter points out the trucking industry

AV prototypes must test to optimize the right mix of sensors

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 7

8 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

Blockchain is a technology initially deployed at the heart of

1. Mutual authentication: each device

2. Message integrity: each message sent

3. Message confidentiality: option-

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 9

In order to solve this famous problem, the

Anyone willing to share a public key will

Certicates and Trusted

10 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

As a reminder, IoT hardware can be classified into two

1. Mains-powered: servers, gateways, high-end sensors and

2. Battery-powered: low-cost sensors, actuators, tiny

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 11

When Adding IoT Devices to the

By Rusty Stapp, UbiquiOS Technology

End-to-End Security from IoT to IT

t Communicating parties identities can be authenticated

12 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

To provide the highest levels of security,

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 13

We have to coordinate that root of

By Anne Fisher, Managing Editor

EECatalog spoke with Smith and George T. Hilliard, Director of

EECatalog: Whats important for someone targeting Industrial

Editors Note: This spring saw publication of the Industrial Internet

Understood too is embedded designers ongoing need for information

14 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 15

focus on their own special requirements.

16 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

NETWORKING & DATA CENTER TECHNOLOGIES

Today, the goal of the OCP is to spark a collaborative dialog and

Figure 1: Virtualization enables CSPs to run many applications

Facebook sought a new rack-scale architecture that would

Traditional Rackmount Server Drawbacks

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 17

Leveraging Open Compute and Open Rack Terms and Acronyms

Beyond the CG-OpenRack-19 specification

18 | Embedded Intel Solutions Fall 2017 | www.embeddedintel.com

field replaceable units (FRUs), carriers with many more physical

Other differences found when comparing CG-OpenRack-19

t 19-inch rack-mounting defined (versus 21-inch in OCP)

www.embeddedintel.com | Embedded Intel Solutions Fall 2017 | 19