City shrouded in fog

Fog computing: Get the most out of your IoT infrastructure

The Internet of Things (IoT) is rapidly moving from experimental to active production status for many companies.  With many building custom circuits and expanding networks, it's time to plan for unprecedented growth in data—which is all moving to the cloud.

This is where fog computing (a.k.a. “fogging”) comes in. A metaphor for a ground level cloud, fog computing moves the cloud closer to the devices collecting the data. This is done to address a big problem: lots of data.  Consider this scenario: a smart grid network is assessing, managing, and reporting on data. The data is useful to the owner of a house, the manager for a city block, the mayor of a town, the leaders of a country and state.

The traditional model for storing data and subsequent data analytics is to push the data up to the cloud. But for real-time or near real-time data assessment, the sheer volume of data overwhelms networks and cloud systems. The answer is to move the real-time analytics closer to the people who need the data. Fog computing pushes cloud services locally where the data is needed and then later syncs to a central cloud.

Application Migration to Cloud: Best Practices Guide

What should your first fog project be?

Fog computing should not be your first IoT project. Fogging is complex, and it is best to have an understanding of the IoT first, and to understand the significance of the data you are collecting.

But let's assume that you've gotten some IoT experience, and you've begun to amass more data than you know what to do with. When do you make the jump to a fog computing network? The tell-tale indicator is your data volume. How much are you collecting? The more data you collect, the more reliable your system must be and the more efficient your cloud must operate.

Imagine you own a logistics company, with 5,000 trucks that service regional and national accounts. Each truck is equipped with IOT sensors that provide location, vehicle diagnostics, traffic, and weather conditions. The data you collect helps you adjust and improve the efficiency of trips, ensures that drivers are stopping and resting as appropriate, records state and federal vehicle usage, and informs you when the product is picked up and delivered.

In this example, there are four interested parties that need to see the data that is being collected on the IoT-enhanced truck: the truck driver, the customer receiving the product, the company managing the logistics, and the state or federal authorities.

How fogging delivers value

There are three key places the data can reside:1. locally on the vehicle; 2. regionally for the logistics manager; and 3. in a central cloud for regulatory reporting. The goal is to move the data to the cloud, but the benefit of keeping data locally or close to specific customers provides the opportunity of real-time data.

The driver sees in real time when he or she is expected to rest, and the regional logistics manager can make route changes based on data from the trucks. This is not a crazy concept. Uber is providing the same network model for their drivers, routing, and payment tools.

Simply put, some systems must perform in real time and others don't have to. Fogging ensures that the appropriate data is available where and when it is required.

Who benefits most from fog computing?

Logistics is not the only example of where fog computing is useful. Here are other use cases where a fog implementation can be of tremendous benefit:

  • Healthcare: any device, such as a glucose monitor, can regularly send data. The data is useful for the patient, doctor, and potentially the insurance company. Both the patient and physician can benefit if a fog network is close and immediate to the data (the insurance company typically requires trending data patterns).
  • CPG: "Consumer packaged goods" companies ship massive amounts of product all the time. Adding sensors to packaging provides real-time data for when the product is shipped, received, and sold. The whole process can lead to better cash conversion cycles, or CCC. Real-time CCC helps a store manager keep optimum volumes of the product in stock.
  • Smart cities: Sensors are now appearing throughout cities. Sensors for parking, water systems, sewage, energy, and street lighting each provide opportunities for efficiency improvements. Of course, the volume of data acquired is massive. Again, a fog network will keep data close to the people who need to assess the data.

There's a common thread among various fog computing scenarios: Not all people need to have real-time data. But for those who are most impacted by data results, the fog architecture can offer them near real-time access while allowing historical data analytics to occur at whatever pace other teams require.

What are the big concerns for fog computing?

The concept of fog computing is very new. Cisco is the first company credited with the concept, having introduced it in 2014. As with many technologies, fog computing was created to address the specific problem of too many data reducing the efficiencies of cloud services.

But what are the risks? In my opinion, the risk is not designing for a fog network when it is clear that there is the strain on the existing system. Which means you need to be able to clearly explain what fog computing is. The core to fogging a network is based on mature cloud services. The same challenges of authentication, security, and data load management have long been addressed. 

Two groups to follow through the fog

The concept and continued support for fog computing began with Cisco, when they delivered a vision to accelerate value from billions of connected devices.

The software and networking industry, however, seized on the idea and many companies now support fog computing. This has led to the creation of the OpenFog Consortium. The goal is to drive the benefit of fogging out to leaders from a group who is not financial motivated.

Fog computing is a natural and needed extension of cloud computing. Facing the possibility that there will be billions and billions of IoT devices, it is only through leveraging emerging models such as fog computing that the data collected from all these IoT solutions will deliver greater return on investment.

Is your organization "fogging"? Share your experiences at the cutting edge.

Application Migration to Cloud: Best Practices Guide
Topics: IT OpsMobile