This column has devoted a lot of ink to fog, edge, and cloud computing. I spent a considerable amount of time last week digging into the basics of fog and edge for those who still needed a better understanding of how these technologies are different.

For this column, we are going in deeper and I will be honing in on the market for fog and edge computing, including industry market opportunities, hurdles, and more. If you have been paying close attention you know by now that the fog or edge computing promises near realtime insights and facilitates localized actions. In some IoT business cases, data can be processed more efficiently at the edge of the network.

The “edge of the network” simply refers to the end point, machine, object, or person that’s generating the data. Pushing resources to the edge is a growing trend in the enterprise world.

Gartner is saying that currently around 10% of enterprise-generated data is created and processed outside of a traditional centralized data center or cloud. By 2022, Gartner predicts this percentage will reach 50%. That’s less than five years away. What does the market for fog computing look like in terms of industries with the most market opportunities?

At the end of last year, an organization called the OpenFog Consortium surveyed its member organizations to piece together a sort of “state of the market” for fog computing. The survey was very telling in some of its results. For instance, 40% of those surveyed say they have a higher budget for fog in 2018 than they have had in the past.

And, for those respondents that pursue research grants, 75% expect an increase in grants for fog-based research. Why are more companies turning to fog?

Three of the reasons we keep hearing about are latency, network bandwidth, and cost effectiveness. We’ve examined the benefits of fog in great length in this column already, so for this column, let’s move on to industries that are adopting fog computing.

The top three industries looking to adopt fog solutions, according to OpenFog’s survey, are manufacturing and factories, smart cities, and transportation. This is somewhat of a surprise. In manufacturing and factory settings, one of the best examples we’ve seen in the field is quality control. If you look at the lead feature on this month, we detail a quality-control solution in a factory that relies on fog computing.

Imagine this scenario: As products move along a conveyor belt, machine-learning algorithms and cameras work together to inspect each product and detect defects in near-realtime.

Within a few milliseconds—that means no room for latency—the system takes pictures and compares them against its algorithm to determine which products can go through and which ones can’t.

Transportation is another great example of why processing data at the edge of a network could be beneficial for IoT applications. In traffic situations, even microseconds count.

If we’re talking about end points like autonomous vehicles and street lights, road signs, and other elements of the transportation infrastructure, we might not want to rely on cloud computing, because the end-to-end latency could mean the difference between life and death.

Taking out the drama here, but when it comes to certain IoT applications, literally any delay is too much. The fog market opportunity in these verticals and others is set to grow rapidly, potentially reaching $18.2 billion or more by 2022.

We’re going to see vertical markets like energy and utilities contributing $3.8 billion to this market, transportation contributing about $3.3 billion, and healthcare contributing just over $2.7 billion.

You can’t talk about market opportunities without also mentioning market hurdles, and there are some hurdles for fog computing. To support and enable this growth, the industry will need to pursue capable hardware and security and management services, and we’ll need to see the emergence of fog service providers. For instance, 32% of respondents in OpenFog’s survey listed security as the top challenge in adopting fog solutions. And other concerns include interoperability and an unclear ROI (return on investment).

The market will also require proven business models, which will come in time. Regarding fog service providers, there are a few different types of players in the arena.

First, we have “fog as a service”—a product offering in which a client pays monthly, quarterly, or annually for a particular outcome and the required hardware and software.

Then, there’s fog application services, which are enabling applications designed as distributed computing and analytics functions usually within a customer environment.

And, finally, fog support services refers to ongoing maintenance and support services for customers that pursue an “as a product” purchase model. Think of it this way, “as a product” is in contrast to an “as a service” purchase model, and it requires end users to own the hardware and software in a fog environment and pay for annual service and support.

There are several organizations out there that are working toward advancing fog computing by addressing some of the hurdles, as noted earlier.

One great example is EdgeX Foundry, a vendor-neutral open source project that’s building an open framework for IoT edge computing. There is a lot going in fog, edge, and cloud, and there is no shortage of opinions and ideas. It’s certainly educational and, at the very least, worth checking out what they all have to say.

Want to tweet about this article? Use hashtags #M2M #IoT #blockchain #security #data #cloud #interoperability #fog #edge #AI #analytics #machinelearning #bigdata #cybersecurity #5G #EdgeXFoundry #OpenFog