Some of the buzzwords gaining steam in the Internet of Things (IoT) and Industrial IoT (IIoT) world are “fog computing” or “edge computing.” My preference, says R. Scott Raynovich, founder and chief analyst of Futuriom, is for the term edge computing, because it accurately describes the IoT market,which is pushing computing power closer to the edge to power future applications for IoT such as artificial intelligence (AI) on devices.

IoT is yet another evolution in computing architectures, which are constantly shifting. First, we had mainframes, with centralised computing. With the advent of PCs, the market shifted to a client-server model, pushing computing power to the servers and clients.

Then came cloud and smartphones — balancing out the computing power between ultra-powerful computing in the cloud and powerful chips in smartphones, the combination of which delivers sophisticated applications such as Uber which can tell you exactly when your car will arrive.

Industrial IoT is an emerging market which will take advantage of the trend toward further connectivity and edge computing power, as Futuriom has outlined in a recent report, the Ultimate Industrial Internet of Things (subscription required – Readers should use discount code “EDGE” for 20% off – exclusive to and

By leveraging intelligent edge devices, low-power wide area networks (LPWANs), and IIoT cloud systems, IIoT will bring automation and AI to the industrial world, creating trillions of dollars in value by delivering more efficiency to the world industrial economy.

This trend is also going to drive more demand for computing closer to the billions of devices. This will require a new generation of chips and devices that can provide powerful processing power and low energy consumption — much lower than that of smartphones.

Next-gen chips power the edge

To deliver power at the edge, IIoT needs a combination of things: Efficient, power-efficient chips; cloud software that can manage devices; security in both the IoT endpoints and the cloud; and secure and reliable LPWAN networks. Eventually, the trend will accelerate with more powerful mobile cellular networks designed for IoT, including 5G and Narrowband IoT (NB-IoT).

Imagine a truck that’s using some sort of energy efficiency analytics program, which tells it how fast to drive to maintain to get the best fuel mileage. The truck may be using long-range communications such as GPS or LPWAN network to connect with the cloud. But ultimately, crunching the numbers about its speed, location, and trajectory can be best calculated with computing power on board. In that case, the edge has become the truck itself.

Cars and trucks can host powerful computing devices, so for transportation IoT applications, compute functions can happen in the vehicle itself. But what about a remote sensor on an oil rig? It’s likely the sensor is feeding a gateway device somewhere in the system, at the edge of the IoT network.

This may be where the processing will happen. For this reason, the edge is an ill-defined place – which is maybe why they invented the term ‘fog computing’, which indicates how the cloud is creeping outside itself.

Another example might be a factory that is also running energy efficiency analytics based on temperature sensors in a building — in order to save thousands of dollars a year. For IoT, gateway devices themselves may soon have enough computing power to run this software, rather than requiring the data to be shuttled back to the cloud.

This means that the next-generation of IIoT devices will be more powerful and be able to bring analytics to the edge — this all starts with the next generation of chips. IIoT chips are different from smartphone chips — they require high security and lower power consumption. An IoT gateway or device might be locked in a closet of even buried in the ground for months or years.

For some examples of how the chip market is evolving toward IoT, ARM was successful at developing chips with the characteristics for IoT, leading Softbank to acquire the company for US$32 billion to form the foundation of its IoT business. Qualcomm is working on an entire line of chips targeted at IoT, featuring built-in security.

NXP has a massive line of chips optimised for the connected vehicle, leading Qualcomm to bid on acquiring the company (the acquisition is expected to close this year). Many other chip players including SemTech, Silicon Labs, Samsung, and Texas Instruments — among others — have IoT-specific chipsets.

(Also see: and )

This new generation of chips is being used to build IoT gateways manufactured by companies including Advantech, Cisco, Dell, and HPE. Gateway devices are often the first collection point at the edge and they could set in vehicles, telecom access head-ends, telephone poles, and other places where the data from devices is aggregated.

The IIoT cloud

The development of the IIoT will have new requirements for cloud platforms. Specific IIoT software is used to connect, manage, and process data from the devices in the network — or send the data to deeper IoT applications hosted in the cloud.

These IIoT cloud platforms can be broken into three groups, which include platform-as-a-service (PaaS), such as Amazon AWS for IoT and Microsoft Azure, which allow hosted public cloud services for IoT; connectivity cloud platforms, such as Cisco Jasper, which are focused on connectivity; and data management cloud platforms such as GE Predix and Davra, which aren’t focused on the connectivity to devices themselves but are designed to host and analyse data.

The IIoT cloud platform is a robust, growing market, with many entrants, ranging from large players such as Amazon, Cisco Jasper and Microsoft, to start-ups such as Ayla Networks, Davra Networks, and Samsarra.

Connecting with the LPWAN

Finally, you need networks to connect all these new devices at the edge. This has led to the development of new IIoT-specific communications networks.

LPWAN is a generic term that refers to a number of technologies targeting low-power WAN connectivity, including NB-IoT, LTE-M, LoRaWAN, and other proprietary technologies including SigFox and Ingenu. The key differentiator is that some of these networks use unlicensed spectrum (LoRa and Ingenu, for example), while LTE-M, and NB-IoT include future versions of 5G targeted at IoT use licensed spectrum.

You can further break the market down into proprietary and non-proprietary. For example, although Ingenu and Sigfox run networks based on unlicensed spectrum, but they use their own technology protocols to deliver the WAN connectivity. Some supports of licensed standards, such as those developed by the 3GPP, are betting that LTE-M and NB-IoT are better ways to go.

Right now, it looks like LPWAN will be a dynamic market for a while, with a mix of standards and communications networks for a variety of use cases including transportation, manufacturing, and agriculture, for example.

Providers of IoT and IIoT platforms will need to carefully define their architecture in terms of location and the edge. IIoT devices are going to generate massive amounts of data, so intelligent algorithms will determine the most cost-effective place for processing.

Exclusive 20% discount offer to our readers

For a more detailed analysis of the Industrial IoT market, purchase Futuriom’s 50-page Ultimate Industrial Internet of Things (IIoT) Report. Futuriom is offering readers a special 20% discount — in the shopping cart enter “EDGE” as the coupon code, which discounts the price of a single-user license from $950 to $760. The IIoT report covers the range of communications and cloud technologies that are being applied to businesses around the world to provide connectivity, analysis, automation, and optimisation of a range of industrial applications.

About the author:

The author of this article is R. Scott Raynovich, the founder and chief analyst of Futuriom. For two decades he has been covering a wide range of technology as an editor, analyst, and publisher. Most recently, he was VP of Research at, which acquired his previous technology website, Rayno Report, in 2015. Prior to that, he was the editor-in-chief of Light Reading where he worked for nine years.

He was the founder of the Heavy Reading Insider research service. Raynovich has also served as investment editor at Red Herring, where he started the New York Bureau and helped build the original website. He has won several industry awards, including an editor & publisher award for Best Business Blog, and his analysis has been featured by prominent media outlets including NPR, CNBC, The Wall Street Journal, and the San Jose Mercury News.

Comment on this article below or via Twitter: @IoTNow OR @jcIoTnow