Predixion dashboard. (Image courtesy of Predixion.)
IoT data is kind of like the attic of a house. It has a way of collecting objects and files that might be useful – if you can only remember where it is.
Data on the IoT is ever-growing, and the computational expense and latency of central cloud-based analytics means more of this data will go unused. As the IoT looks set to increase from 6.4 billion devices in 2016 to 20.8 billion devices by 2020, according to Gartner, it’s clear that this messy attic won’t clean itself.
“IoT promises significant advances for users, but the fact that only 10 percent of IoT data even makes it to the cloud proves that current analytics solutions are too reliant on central processing to be effective,” said Simon Arkell, CEO of Predixion Software.
This means the IoT industry needs to find a new way to process this data to ensure it is being used to its full potential.
Enter the cleaning crew: edge computing technologies.
Shifting Analytics to the Edge Gateway Will Make Better Use of IoT Data
Using Predixion RIOT, engineers will be able to make better use of edge computing. The software’s architecture is designed to process the data where it is most needed on the IoT device, the edge gateway or the cloud. To perform this function, RIOT offers three versions of their software:
- RIOT Nano, a self-contained analytics engine with a small footprint. This software runs on the IoT device to provide visual analytics or stream analytics to the gateway or cloud.
- RIOT One, a self-contained analytics platform for the network gateway. This software can connect to thousands of devices and offers visual analytics without the cloud, allowing for faster insights.
- RIOT Enterprise, a cloud-based offering of the software. It aggregates the visual analytics that are streamed from both devices and the gateway. This provides a centralized view of the system.
The idea is to speed up analysis and to process data and communications more efficiently. This should help to reduce storage and data transfer needs.
“To be able to deal with the dramatic growth in data volumes, analytics performed at the edge becomes mandatory,” said Arkell. “RIOT enables users in markets such as manufacturing, energy and healthcare to gain faster time to insight via the delivery of real-time visual edge analytics with full integration to the orchestration capabilities of the cloud.”
Some other capabilities of the Predixion RIOT software include:
- Quick deployment of analytics on the device, gateway and cloud
- Connected, partially-connected and disconnected decision making, ensuring the system can operate (at least partially) without network connectivity
- Visual analytics, real-time actions and predictive analytics
To learn more about data analytics, check out this online master’s degree on big data analytics for the IoT.