To Nha Notes | Sept. 4, 2023, 10:58 p.m.
Inmon: a stable warehousing strategy where data consistency is the highest priority. All user-facing data marts are built on top of a robust and normalized data warehouse.
Kimball: a dynamic warehouse strategy where quick development of useful data structures is the highest priority. All user-facing data are built on top of a star schema which is housed in a dimensional data warehouse.
Data Vault: a fast and asynchronous warehousing strategy where speed of both development and run time is the highest priority. All user-facing tables are built directly from untransformed source data.
I would love to hear your experience with using any of the three.
Data Vault is an innovative data modelling methodology for large scale Data Warehouse platforms. Invented by Dan Linstedt, Data Vault is designed to deliver an Enterprise Data Warehouse while addressing the drawbacks of the normalized (3rd normal form), and Dimensional Modelling techniques. It combines the centralized raw data repository of the Inmon approach with the incremental build advantages of Kimball.
This article summarizes the drawbacks of the 3NF and Dimensional Design approach and lists the advantages and disadvantages of the Data Vault approach. Finally, it includes links to some useful background reading, and aims to answer the question: Should I use Data Vault or another data modelling method?
https://www.analytics.today/blog/when-should-i-use-data-vault
https://kentgraziano.files.wordpress.com/2012/02/introduction-to-data-vault-modeling.pdf