The enterprise is acquiring real-time data in its Time Series Databases (TSDB).
Well it's not several points of acquisistion but thousands and even hundreds of thousand.
The TSDB technologies cannot help on this very topic of functional vision.
Yet the challenge is here : how to use the same datas in different business use cases that ressemble, overlap or use the same data whatever.
Meaning is multidimensional in that several businesses in the enterprise will see the data using different point of views with their own semantics and structuration...
That's why some information systems happen to be built with several functional description repos one for each dimension. The risk of loosing global coherence is there for sure.
And yet, how to implement innovating complex business use cases that cross dimensions, say, that aim at coupling several businesses in the enterprise for improving overall performance?
Wide correlations are aimed at indicating if one point of acquisition is related to another point of acquisistion inside one functional dimension and even several functional dimensions at the same time.
That's why it is possible to correlate a default value at of point of acquisition with another default value at another point of acquisition.
These correlations can be used in real time or in time depth (using Data Science).
Your TSDB is huge and complex and yet its contents holds the promise to make your enterprise perform better!
Your TSDB needs functional meaning on top of it.
That functional contextualization on top of the data must be expressed in terms of asset hierarchy, in terms of influence zones, in terms of geography, in terms of IOTs, etc...
But it's not all, historization on the contextualization is also a requirement.
Building that functional contextualization will transform your huge and complex TSDB to make it smart.