Usually, it doesn’t.
ARDI only stores the minimum amount of information possible. Details about what your assets are, what properties they have, their relationships, location, and some key identifying pieces of information (such as a name, tag number etc.).
Other than that, it might store specific pieces of information you don’t have available anywhere else. If you don’t have a document management system, we might store some of your documentation (such as manuals and calibration sheets), or model numbers. These are stored in a MariaDB/MySQL database, although this can be moved and/or changed depending on the customers needs.
But we do not store any time-series data.
We consolidate the information you already have. If that’s in your on-site historian, your SCADA system, found on the enterprise network or pulled from the cloud, we bring it into one place so you can access it easily and consistently.
We’re not a data-lake or data-warehouse, we’re the tool that makes your scattered systems LOOK like one – then we go further to structure and deliver the data where it’s needed. No replication, no cloud required, no headaches with replication and back-filling or huge amounts of storage to maintain.
So you’re not a data warehouse?
No, ARDI isn’t a data warehouse – it’s closer to a Data DoorDash. Describe the dishes you’d like, and our drivers will go to the most reliable source and bring you back everything you wanted to consume, as fresh as possible. We’ll even do the carving and provide the cutlery.
What if I’m not recording anything?
For users that don’t have a historian, our preferred open-source time-series database for slow (ie. one sample per-channel per-second) data is Prometheus, and for faster (ie. sub-second) data is InfluxDB
We – or one of our partners – can also suggest commercial historians, depending on your needs.