Redefining Middleware

Harry Forbes, lead analyst for the Distributed Control System (DCS) market at ARC Advisory Group makes an insightful observation: Skkynet redefines middleware.  In a recent paper, Middleware’s Changing Role – to Serve Industrial IoT, from the perspective of 30 years of hands-on experience in industrial automation and control, Forbes looks at Skkynet’s software and services and sees how they fit together to provide middleware for Industrial IoT.

In a way, you could say that Skkynet got its start in middleware.  The first large-scale implementation of the DataHub, installed in a chocolate making plant in Toronto, provided a fast and reliable connection between a Wonderware application running in Windows and supervisory control software running in QNX.  According to Wikipedia, the term “middleware” is “most commonly used for software that enables communication and management of data in distributed applications.”

Among other things, this is what the DataHub, the ETK, and SkkyHub do.  They enable communication and data management between applications.  What’s new is that they expand this middleware functionality across industrial networks and the Internet, in the Industrial IoT. As Forbes explains, “[T]he Industrial IoT requires applications and services that span across enterprises, as well as reaching assets deployed in field locations, many of them quite remote.”

It seems that we are, in a sense, redefining middleware.  Maybe that’s what we’ve been doing all along, but just never saw it that way.  “Middleware” was originally defined as software that intermediated between the operating system and software applications.  In the 1980s that definition was expanded to include linking between older and newer applications.  With the advent of networking, distributed systems and client/server architectures, middleware is now commonly referred to as simply the “glue” that holds things together.

The DataHub architecture combines several features that support the traditional understanding of middleware:

• It does protocol conversion, allowing software using one protocol to connect to software using a different protocol.
• It supports TCP communication, allowing connected software to connect over a network or the Internet.
• It has an API that allows custom connections from various software packages.

In addition, the DataHub offers features that may or may not be considered traditional middleware:

• It aggregates the data from all connected clients into a single, universal data set, to which any client can subscribe.
• Its built-in scripting language allows for data to be manipulated or transformed as it flows through the system.
• It provides an HMI that supports visualization of the data.

The combination of all of these features, and possibly others that I have not mentioned, makes DataHub technology an ideal middleware solution for the Industrial IoT.  The ETK provides a basic subset of this functionality at the embedded device level.  And SkkyHub offers a different subset, along with additional functionality on the cloud.  Taken together, “These three components can be assembled quickly and in different configurations to fit various Industrial IoT application requirements,” is how Harry Forbes described the Skkynet software and services, providing the security, scalability and performance that is critical for any industrial application.