Edge Computing

Primer

IoT Intelligence Starts

at the Edge
 Opto 22
43044 Business Park Drive • Temecula • CA 92590-3614
Phone: 800-321-6786 or 951-695-3000
Pre-sales Engineering is free.
Product Support is free.
www.opto22.com

Form 2184-160830
© 2016 Opto 22. All rights reserved. Dimensions and specifications are subject to change. Brand or product names
used herein are trademarks or registered trademarks of their respective companies or organizations.
 Edge Computing Primer
IoT Intelligence Starts at the Edge

Edge Computing Primer

On the train to work, Lee opened an PAC at the Edge
email on her smartphone sent from a
PAC (programmable automation That PAC operating the surface-mount
controller) operating a surface-mount tool at Lee’s factory operates at the edge
tool at her factory. The PAC attached a of the factory’s network.
quality control report to the email that Systems like these at the network edge
suggested changing the tool’s solder are increasingly able to leverage cloud-
temperature. based resources to perform edge
To generate that email suggestion, the computing—if computing resources
PAC had securely sent yesterday’s exist as needed along the path from a
production data to a cloud-based sensor to the cloud—and if these
analytics system to compare current and computing resources reduce the total
historical data for the machine. amount of data to be sent to the cloud
for storage, processing, and analysis.
Next, it accessed the machine
manufacturer’s website and obtained the As a result, businesses can more quickly
latest recommended settings. identify real opportunities for operational
efficiency improvement and meaningful
Finally, the PAC built a production revenue generation.
efficiency report with a suggested solder
temperature for today’s production run To foster such business benefits, data
that would increase yield by 7 percent from the physical world of machines and
over yesterday’s run. equipment must be available to the digital world of the
Internet and information technology systems, quickly,
Lee clicked a link in the email and connected to the PAC’s easily and continuously.
mobile interface over a secure, encrypted channel. Lee
logged in and navigated to the machine’s solder Successful IoT applications require operational
temperature setpoint, where she entered the technology (OT) professionals to make data from their
recommended value. systems, which monitor and control the physical world,
accessible to the data processing systems of information
All this took place before she got to the office. technology (IT) professionals.
Once the data is there, cognitive prognostics algorithms
running on IT systems can analyze it, refining raw
physical data into actionable information that can
WHITE PAPER
Form 2184-160830

predict outcomes in real time. The results can be used to

improve inventory management and predictive
maintenance and reduce asset downtime.
But before such benefits can be realized, three problems
need to be solved: connectivity, big data, and IoT
architecture.

PAGE
1
Edge Computing Primer

The Connectivity They also lack the physical

Problem connections and logical interfaces
to communicate on the Internet of
The Internet of Things runs on vast Things. They do not have a built-in
amounts of data, generated by the Ethernet jack or wireless interface.
physical world and then Few sensors and circuits can
transported and analyzed by the connect to the Internet, let alone
digital world. It’s an attempt to speak or understand the
achieve perpetual connectivity languages the Internet uses, like
and communication between JSON, RESTful APIs, and JavaScript.
people and things and even They don’t run an operating
between things and other things. system or have a built-in TCP/IP
But unfortunately most of these stack or web server.
things were never designed to And they have little or no built-in
serve this new purpose. They were computing power, so providing
designed and installed long before edge computing at this level to
the Internet was developed. filter volumes of data before
At the edge, things like sensors, forwarding to the cloud is
circuits, relays, and meters are impossible.
attached to industrial control Right now the Internet and the
systems used to operate things we want to connect to it aren’t communicating.
equipment and machines. These sensors translate what’s There’s a disconnect between the physical world of current
physically happening in the world (temperature, light, and voltage and the digital world of servers and clouds.
vibration, sound, motion, flow rate, and so on) into an
electrical signal like voltage or current that can be Integrating these disconnected things and systems is no
interpreted by other systems to monitor and control small task. And with the significant potential technical
physical equipment and machines. pitfalls and risks of integrating these disconnected systems,
we begin to wonder how long it will take to realize return
These sensors typically have little or no intelligence and are on our investments in IoT applications.
designed to merely observe and report. They were not
designed to communicate with the digital world of the IoT. One option is to simply wait for highly intelligent,
connected sensors to become available to the
The physical world’s language, the language of flow marketplace. But those sensors are years away from being
meters, temperature sensors, switches, and relays, is rarely cost effective.
digital bits and bytes. Their language is not the ones and
zeros that information technology and Internet devices Moreover, sensors installed today or even decades ago are
understand and use to communicate. still performing their tasks. They’re just not connected to
the IoT, so the data they generate is siloed and inaccessible
to IT systems for further analysis.

What is Edge Computing?

You’ve heard about cloud computing, which is using a network Edge computing basically brings cloud computing down to the
of remote servers, rather than a local server or personal edge of the network, in the physical world. Edge computing is
computer, to store and manage data and run computer using computing power at the edge to filter or process data and
programs. then send only the required data to the cloud.

The advantages of cloud computing are that people or The advantages of edge computing are many. First, it reduces
organizations can combine and share computer resources traffic on networks and the Internet by reducing the amount of
rather than having to build and maintain them. data sent. It also plays a valuable role in efficiency, security, and
compliance.

PAGE
2
 Edge Computing Primer

signals, PLCs were not designed to interface with the

Right now the Internet and the things we Internet of Things.
want to connect to it aren’t communicating.
PLC hardware, software, and programming languages were
There’s a disconnect between the physical
designed for repetitive, application-specific tasks like
world of current and voltage and the digital
process control and discrete automation. They typically use
world of servers and clouds.
proprietary protocols and languages for communication
and programming, and do not include information security
The Big Data Problem standards like encryption and authentication.
PLCs were originally designed as standalone systems. The
Across the globe a massive installed base of things exists protocols they use are seldom Internet compliant and are
today, generating useful data that the IoT wants to access designed for point-to-point communication instead of the
and consume. In oil and gas applications, a typical oilfield point-to-multipoint communication architecture found in
has up to 30,000 sensors installed. Factories and plants the IoT ecosystem.
across the world have billions of sensors.
If systems that communicate using Internet-compliant
Each sensor is capable of generating huge amounts of data protocols—such as PCs, web servers, and databases—
from the physical world. Some IoT applications could want to communicate with a PLC, a vendor-specific and
potentially generate terabytes of data per second. often proprietary software driver or hardware-based
These are volumes of data the digital world has never seen protocol gateway is required.
before. This is the Big Data problem. OPC (Open Platform Communication) software is one
Moving that much data onto existing network and Internet solution to this communication disconnect. But OPC was
infrastructures for cloud-based analytics and centralized originally designed around PC architecture using the
management will clog networks, vastly increasing network Microsoft Windows-only process exchange, COM/DCOM.
and Internet latency. For many industrial IoT applications, Most systems and devices connecting to the IoT are not
that is not acceptable, because real-time control and Windows-based devices.
monitoring are mandatory. For example, take your smartphone. It’s likely an Apple or
For the Internet of Things to reach critical mass, Android device, both of which run modified versions of the
intelligence must be pushed to the network edge, where Linux operating system, where COM/DCOM process
the physical world meets the digital world. exchange does not exist.
Computing systems at the network edge must have the OPC UA (Unified Architecture) has been released, but it’s
capability to collect, filter, and process data generated at merely a wrapper for existing OPC drivers built on
the source, before it’s transmitted up to the IoT.
And at the same time these edge computing systems must
be able to complete the local real-time process control and What is an Input/Output (I/O) System?
automation tasks of traditional industrial applications. Things in the physical world typically communicate in an
electrical signal like voltage or current, but the digital world
doesn’t understand these signals. I/O systems act as the
The IoT Architecture Problem translator.
I/O systems are wired directly to sensors and actuators.
Let’s take a look at how today’s IoT architecture works, so Their job is to convert analog data (like voltage or current)
we can see its complexity and perhaps find a path forward. from a sensor into digital data (ones and zeros) to send to a
computer, or to convert digital data into analog data to send
For a cloud-based server to capture data from an analog to an actuator, to take action in the physical world.
sensor today, the sensor’s data must be translated using a An input sends data from the physical thing to the computer;
series of disparate software and hardware tools. an output does the reverse. Inputs and outputs are like the
fingers on a hand, sensing the world, sending data about it
First, the sensor is physically wired to a device such as a PLC to the brain, and then taking commands from the brain back
(programmable logic controller). While modern PLCs do to the fingers to take an action.
provide basic analog-to-digital conversion of sensor

PAGE
3
Edge Computing Primer

Windows architecture. It requires design engineers to build Layers of complexity must be removed from the
an OPC UA client adapter into their products. And even communication process between digital systems and
then, modern network and Internet assets such as web physical assets. Modern IoT system architectures must be
servers, databases, smartphones, and tablets do not speak flattened, streamlined, optimized, and secured.
OPC UA. If we drive Internet connectivity and data processing
PLCs, OPC servers, proprietary drivers, and protocol power into edge devices, we can greatly accelerate our
gateways quickly become a convoluted IoT architecture. time to insight and action. Edge computing devices will
These layers of complexity not only require time, money, become the sensor on-ramp for the billions of data points
and specific domain expertise to install and maintain, but we intend to connect to the IoT.
also the data being sent from the physical world has been These edge computing systems will need the ability to
converted by so many different pieces of hardware and receive the input signals of the physical world and output
software that data integrity can be jeopardized. the meaningful data the IoT needs, in a form that digital
Imagine the difficulty in provisioning and troubleshooting Internet-enabled systems already understand.
these IoT systems. Edge computing systems must easily and securely access
And then consider that today’s automation architectures the cloud through the open, standards-based
often do not address information security. Sending data communication technologies the Internet is based on.
generated at the edge through so many layers of That means:
conversion not only increases network latency but also
• Internet technologies like TCP/IP, HTTP/S, MQTT, and
opens up complex information security concerns as the
RESTful APIs—the dialect of the Internet—must be
data is transported to the cloud.
built directly into the input/output level, or the point of
Multiply these issues across the billions of devices we physical to digital conversion.
expect to connect using the IoT, and you see the • Internet security technologies like SSL/TLS encryption
communication challenge the IoT faces. and authentication must be built in directly to edge
There has to be a better way. computing systems.
• Cloud-based systems must be able to make RESTful
Flattening the IoT Architecture API calls to access data, or subscribe to data points on
remote edge devices, without the layers of complexity
As we’ve seen, for the IoT to reach critical mass, Internet and conversions that exist in industrial applications
protocols and technologies need to be driven into systems today.
at the edge, where the physical world and the digital world
connect.

The Problem: Complex Current IoT Architecture

The Solution: Flattened IoT Architecture

PAGE
4
 Edge Computing Primer

The Power of Interoperability

In the long run, OT/IT convergence will
We did not always have one cohesive system for sending
demand a flattened architecture and
and transmitting information. Before the Internet and the
seamless communication between assets,
world wide web, many different Internet-like protocols and
using open, standards-based
architectures existed. Computer systems all ran different
communication protocols and
operating systems requiring different programming
programming languages.
languages.
Small pockets of interconnectivity existed, but for the most
part systems were disconnected from each other. It was Conclusion
very similar to the way industrial systems communicate
We’ve seen that edge computing is the sensor on-ramp to
today, with the need for converters, adapters, and
the IoT. Until the communication, security, and computing
gateways.
technologies of the Internet find their way into computing
The Internet was designed to allow input/output and at the edge, the IoT will fall short of its potential.
information systems to share data through a common
Internet technologies are available in some industrial
interface, removing layers of complexity and allowing for
systems today. And some vendors have already started
greater interoperability between systems designed and
bridging the gap between OT and IT by adding IoT
manufactured by different vendors.
technology like MQTT, RESTful APIs, and JavaScript directly
That’s why an Apple computer or Android phone today can into programmable automation controllers (PACs).
send an email to a Windows computer: they speak the
Our shortest path to a successful IoT is to leverage the
same Internet languages. Today’s Internet uses a common
existing interoperability technologies of the Internet in
set of protocols, tools, and routines designed to make the
industrial automation products and applications.
transportation, acquisition, and analysis of digital
information a seamless process, no matter what device
you’re using. How can Opto 22 help you?
Although sensors and other physical assets installed at the At Opto 22, our goal is to flatten the IoT architecture so you
edge may not have been designed with Internet can realize your IoT goals.
interoperability in mind, there’s still a massive opportunity
to collect meaningful data from the huge installed base of Opto 22’s engineering focus is on building hardware and
existing things. software tools to bring the benefits of the IoT—simply,
reliably, and securely—to the things that already exist in
But it will require a solution that understands both sides of your world.
the OT and IT convergence—something that can:
Our PACs offer an easy and cost-effective way to bridge the
• Locally translate the physical world of currents and
real world with the digital world, through a comprehensive
voltages (OT) into the secure, RESTful APIs and JSON
collection of input and output (I/O) modules designed to
(JavaScript Object Notation) frames the digital world
connect with virtually any electrical, electronic,
(IT) understands
mechanical, or environmental device.
• Process and filter mountains of data, sending only
the necessary data to the cloud for analysis This I/O system converts these raw signals to useful digital
data and shares it over the standard networks and
• Provide communications interfaces and processing
protocols understood by IT.
power to maintain the closed-loop, real-time
control requirements of industrial applications. For edge computing, our SNAP PAC controllers offer field-
• Deliver all of the above in a package suitable for proven control, data collection, and logic solving—plus a
challenging industrial environments where dust, RESTful API so you can access that data using the Internet
moisture, vibration, electro-mechanical frequencies, programming languages you’re familiar with: .NET, PHP,
and temperature vary widely. Perl, JavaScript, and more.

PAGE
5
Edge Computing Primer

About Opto 22
Opto 22 was started in 1974 by a co-inventor of the solid-
state relay (SSR), who discovered a way to make SSRs more
reliable.
For over 40 years, we’ve brought commercial, off-the-shelf
technologies to industrial systems all over the world,
designing our products on open standards. We pioneered
the use of PCs in controls back in the 1980s, Ethernet
networking at the I/O level in the 1990s, and machine-to-
machine connectivity in the 2000s. Today, we build secure
Opto 22 SNAP PAC and I/O convert Internet technologies into our PACs and I/O.
raw signals into useful digital data.
All Opto 22 products are manufactured
Securely access data through the
and supported in the U.S.A. Because the
PAC’s HTTPS server and RESTful API.
company builds and tests its own
products, most solid-state SSRs and I/O
modules are guaranteed for life.
When it comes time to visualize, notify, and mobilize your
information, our groov platform offers a simple way to For more information, visit opto22.com or contact
build mobile operator interfaces—mobile apps to securely Opto 22 Pre-Sales Engineering:
monitor and control virtually any automation system or 800-321-6786 or 951-695-3000
equipment. These interfaces can be used on any screen, Or email: [email protected]
from your smartphone to big-screen HDTV.
All Opto 22 products are backed by decades of expertise in
applications like process control, discrete manufacturing,
remote telemetry, data acquisition, and supervisory
control.
Opto 22 products are supported by our experienced
engineers at no charge and available worldwide. Opto 22 groov
for building and
viewing mobile
operator interfaces

PAGE Form 2184-160830 • Opto 22 • 43044 Business Park Drive • Temecula, CA 92590-3614 • www.opto22.com
6 SALES 800-321-6786 • 951-695-3000 • FAX 951-695-3095 • [email protected] • SUPPORT 800-835-6786 • 951-695-3080 • FAX 951-695-3017 • [email protected]
© 2016 Opto 22. All rights reserved. Dimensions and specifications are subject to change. Brand or product names used herein are trademarks or registered trademarks of their respective companies or organizations.