The new architecture is a sensor net talking to a fog talking to a cloud. This sounds complex, but it is not. This is where many hardware companies can come forward, and where hyperconverged can have a huge impact. The architecture seems simple, but is not. There are quite a few moving parts, all based on various container technologies. Are these the containers we are talking about as the next generation of applications, or something more? Can IoT survive without all these moving parts? Does this change how we define IoT?
When I think IoT, I think of millions of devices connecting up to a cloud service, and that is the way many folks see IoT as well. We often talk about three forms of IoT: consumer, business, and industrial. Each form of IoT is vastly different from the others. Within these categories, there are some interesting subcategories, such as infrastructure. Infrastructure is not the usual industrial IoT, but picture all the stoplights, cameras, streetlights, emergency vehicle sensors, and the like, and you start to see what I mean by infrastructure: the smart city.
The “fog” in those use cases is not each and every iDevice, but a series of rugged containers full of power, compute, and other equipment: the perfect containers to contain hyperconverged infrastructure (HCI). These containers need to be spread around to take the massive influx of data from the multitude of nearby sensors and feed them further up the chain for in-depth analytics. The fog is also intelligent; it has to be, as controlling all the sensors is as difficult as gathering all the data. Picture the fog as vast array of electrical substations throughout the city. This is the key that makes a smart city hum.
The application of the rugged container for the fog does not stop at the city. A small village or town would not need as many. We could place them on all ships as well. Even lightweight ones exist within airplanes. In planes, they are often part of the avionics room and suite of computing. When I look at the possibilities, I see the growth of a new sort of business: the fog repair-person. This is the person who can get into the substations, do the work, and get out. Telcos have been doing this for years. Scattered around every town and subdivision are innocuous-looking buildings where the trunks for a neighborhood come together. They contain a large array of switches and compute to support diagnostics and push data up the pike, back to the telco.
The fog is therefore nothing new, but what is new is that rugged containers for all IoT purposes are coming out. They are no longer something you need to put together yourself; you can now buy them and pop them where you need them. How rugged they should be depends on location: city, town, field, ocean, satellite, the Moon, even Mars? Now we’re starting to talk way out there, but not necessarily out of the question. The concept of the fog allows for a wide range of landscapes, distances, and networking to take effect.
Where would you need a pod for your fog to live? Can we start to rent space within fogs? We now have several new business models to consider here. Will businesses eventually connect their new products and sensors to Fog as a Service (FaaS) pointing to our other services within the cloud? How soon will it be before this happens? Could it mean extra income for a city or town? Want to roll out that new game or sensor? Rent some space in a FaaS, hook it up to your cloud, and away you go!
Infrastructure for fog will become as valuable as the bridges we have and need.