With industrial Internet of Things (IIoT) applications pouring out petabytes of data, the opportunity for solution providers now encompasses figuring out new ways to sort, store, access and make sense of all that data. This is not typically the task of operations technologists (OT), nor is it the job of the IT dept.
Instead, the task falls to a whole new category of expertise that straddles OT and IT and comes at the IoT data problem from a more cohesive perspective, with tools to match.
“Look at the technology jobs listings, there are roles we never heard of before,” said Shelley Gretlein, vice president of corporate marketing at National Instruments (NI), which provides hardware for gathering data and testing electronic and electromechanical systems.
NI has been working with industrial, automotive, space and agricultural companies to help them acquire, tag and organize their data since they coined the term “Big Analog Data” before IoT was even “a thing.”
In its years of providing data acquisition tools and analysis capabilities, NI noticed something odd: it turned out only about 5% of the acquired data was actually being analyzed. At its annual NI Week convention it showed how it worked with Land Rover/Jaguar to increase the amount of data they were analyzing from 10% to 90%.
Exploring a Trio of Opportunities
According to Gretlein, the industry has a good handle on how to acquire and store IoT data, whether it is at the edge, on a local server or in the cloud. The problem is making sense of it, and the solution to that problem lies in remote systems management, software configuration management and data management (Figure 1).
Figure 1: Based on its expertise in data acquisition, NI sees new opportunities in helping solution providers solve a new category of problems that it classifies as remote systems management, software configuration management and data management. (Image source: National Instruments).
For anyone familiar with IIoT, a breakdown of these categories reveals some well-known concepts. Remote sytems management translates to effective asset monitoring through solid execution of provisioning, configuration, diagnostics and edge device administration.
Software configuration management includes security and its effective implementation means software revisions are kept up to date, bugs are fixed quickly, functions are modified effectively and security issues are addressed immediately. This “continuous delivery and improvement” model, said Gretlein, needs to operate in dynamic heterogeneous environments with a cohesive OT/IT solution.
The third leg of the platform-based solution that Gretlein hinted at and which is shown in Figure 1, is data management with analysis capabilities that stretch from the edge to the enterprise. This has strong overtones of what is now colloquially called Fog Computing, where the right analysis is done on the right data in the most appropriate location. That can be at the sensor, the gateway, the cloud or anywhere in between. Wherever there is appropriate CPU horsepower and memory.
Tools for Data Collection and Analysis
In its Trend Watch report, NI quoted numbers from IDC that show that at least 40% of IoT-created data will be stored, processed, analyzed and acted upon at the edge. (IDC FutureScape: Worldwide Internet of Things 2017 Predictions.) From NI’s point of view, this means, “An effective data management system must be able to incorporate data from multiple distributed sources and produce different levels of insight to get the right information in front of the right people so they can translate raw data to informed decisions.”
Though NI comes from a research, prototyping and development, and system test background, it believes its SimpleLink and Data Management Software Suite can form at least part of any IIoT solution (Figure 2).
Figure 2: NI’s SystemLink and Data Management Software Suite can monitor assets remotely and keep track of data. (Image source: National Instruments)
SystemLink performs management, deployment and systems health-monitoring, while the Data Management Software Suite handles data searching and indexing, data analysis and processing, and insights reporting and visualization.
In a real-world implementation, the sensor that is monitoring an asset would connect to the hardware (Windows PC, PXI (Windows), or CompactRIO (NI Linux Real-Time). From there the data can be relayed to a local server or to a gateway connection to the cloud to be managed by the Data Management Software Suite. The suite has three main sub-elements that are employed as needed to locate, index, analyze and report on the required data.
NI’s approach takes its test and measurement expertise and applies it to the IIoT. For IoT solution providers, it may be good to analyze what NI is doing and work with them to build upon what they have to extract the most out of the petabytes of data that upper management knows it requires, but isn’t quite sure why. Yet.