Cloud, fog or edge computing: which one do you need for your IIoT applications?

Nowadays it is common to find many articles on Internet about cloud, fog or edge computing, in the context of industrial IT infrastructure. Using Google Trends, it is possible to observe that the search interest about "Cloud computing" started to increase around 2009, followed by "Fog computing" around 2014, and "Edge computing" around 2017.

An example of architecture, in which it is possible to observe the 3 types of infrastructure, is showed in next figure. As you can see, in general, the closest you are performing computations near the field, the fastest you may be able to respond to those events. On the other hand, the computational power available in the cloud is obviously huge, compared to the other two layers below.

In a paper published in 1973, unfortunately rarely cited, W.R Langenbahn discussed critical considerations in the application of supervisory control systems, that in fact are still valid today [1].

For example, just in the Introduction section, Langenbahn states: "the definition of a supervisory control system is, indeed, as viable as the industry itself. Each dawning day brings with it improvements and additions, which must be included in any comprehensive definition". Amazingly, almost 50 years later, this remains true.

Later in the paper, he states: "In defining the signals to be input to, and output from, a remote terminal unit, the economics of the overall system must be considered. This determination can only be made after the application engineer has answered the following questions. First, the engineer must address himself or herself to the functions that the system is given to perform. Having defined these functions, the applications engineer must decide what input and output signals are critical to the execution of these functions".

Here again, replace the term "remote terminal unit", which sounds a little out of date in this era of IoT, with "edge device" for example, and the advise remains perfectly valid.

The?MYWAI? EDGE?platform is as an open, flexible, modular AIoT platform, optimized for delivering equipment intelligence at the edge layer, by seamlessly integrating IoT, AI and DLT, to enable OEMs to transition from a traditional product-based to a service-based business?model.?

If you have any query about this article, you are most welcome to send an email to: g.sanchez@myw.ai

REFERENCE

[1] W. R. Langenbahn, “Critical Considerations in the Application of Supervisory Control Systems,” IEEE Trans. on Ind. Applicat., vol. IA-9, no. 5, pp. 545–551, Sep. 1973, doi: 10.1109/TIA.1973.349928.