The Event Tile Data Modeling pattern captures business processes and transactions within AgileData's three-layered data architecture, modeling "who does what" at a specific point in time.
This is cool. I my Kimball-centric view, it is a change-tracked / type2scd'ed fact table with a couple of dims around it. Probably more efficient from a storage point of view, though.
The terminology is a bit foreign, but I'm coming around to it.
I wonder how to tackle this when sourcing from a system of capture that does not track granular changes. Perhaps something like the CDC on its database, but then we may not know who did that change necessarily.
This was a really complex one to try and explain in a simple way, and i'm not sure we have nailed it.
Whats your thoughts?
------
I have also added a new section to the Agile Data Engineering Pattern posts.
A Mind Map of the Podcast episode generated by Google NotebookLM (how did I miss that feature turning up!)
This is cool. I my Kimball-centric view, it is a change-tracked / type2scd'ed fact table with a couple of dims around it. Probably more efficient from a storage point of view, though.
The terminology is a bit foreign, but I'm coming around to it.
I wonder how to tackle this when sourcing from a system of capture that does not track granular changes. Perhaps something like the CDC on its database, but then we may not know who did that change necessarily.