https://en.wikipedia.org/wiki/Data_vault_modeling
My company has decided to use DV 2.0 for a coming big Cloud project, but I don't get the excitement, please enlighten me..
The proponents will say, you wont understand it (their table mapping) unless you get "Certified". They will claim it's all out in the open, while selling to Management that their teams need to get Certified on it first and that certification can only come from a certain place...
You probably want to read the threads from here:
http://kejser.org/the-data-vault-vs-kimball-round-2/
follow the discussion and this reference
https://timi.eu/blog/data-vaulting-from-a-bad-idea-to-ineffi...
So you understand what you are getting yourself into. If after reading all, you are still unclear on the value-add that is a warning about it.
Note you are still supposed to build then your Star Schemas and Data Marts from it. I am sure it was sold to the Management first...
This is just a framework around ETL processing that highlight the usage of table via name in the data warehouse layer.
When you implement it the big problem is that everybody may have a different definition of what the names mean (Ask WhatsApp a satellite is to 10 people and you will get 10 different answer).
Also as it adds a layer of abstraction, you will spend a lot of time searching where the data comes from.
Technically, you should be able to delete all the data as it's processed but practically you always ends up with a persistent layer.
I will never understand why they don't just replicate the source model.
https://datacadamia.com/data/warehouse/data_vault/data_vault