It’s human nature to compartmentalize and organize our thoughts into subjects, topics, disciplines, etc. In the abstract we realize (hopefully) that the divisions among these are our own creation and can disappear — hence convergence.
Much of ARC’s research and thinking lately has focused on 3 topics: the Industrial IoT, IT/OT Convergence, and Open Process Automation (a move to break up large automation systems into many more highly standardized hardware/software components). The challenges I’ve been thinking about in these areas are related. For example:
Industrial IoT – As the number of monitored assets expands by orders of magnitude, the the operation and management of their monitoring must be done 100% remotely.
IT/OT Convergence – COTS IT is used all throughout industrial automation except for real-time and high-availability controllers, which remain highly proprietary. IT [cloud] methods for delivering real-time and high availability have not yet scaled down to the size of these controllers.
Open Process Automation – When you distribute process control software over 100x or 1000x more devices than today, how do you distribute the applications among so many devices and at the same time unbind those applications from a single set of vendor development tools?
All three of these challenges are related. The relationship is that each calls for easily managed yet widely distributed systems, including embedded systems which today are often outliers in terms of their hardware and (especially) their software development. In fact, these challenges fundamentally require improvements in software much more than hardware.
A Reason for Contained Optimism
Lately I’ve become much more optimistic that each of these challenges can be met, and met in the short-mid term. How is that possible? As Mr. McGuire said in the movie The Graduate “I want to say just one word to you. Just one word. Are you listening?”
Software containers are not a new technology, but they are a very powerful and useful idea. Until recently one needed very deep domain expertise to use containers effectively. No longer. In the past 3 years the ability to develop applications using containers has been simplified to the point that just about anybody can do it. For that reason, containers have become a very critical tool for enterprise cloud computing and enterprise applications. My bet is that as container tech rapidly develops and becomes even more standardized and commonplace, containers will drive a whole new wave of IT/OT convergence, Industrial IoT, and Open Process automation. The underlying reasons are too long for a single blog post, but that is my thesis. What do you think?