Tagged: data modeling

Kimball versus Inmon: a peace offer? 5

Kimball versus Inmon: a peace offer?

Lately there were some interesting updates in the ever-existing ‘Kimball versus Inmon’ discussion. This this Bill Inmon wrote an article expressing his views on these things: http://www.b-eye-network.com/view/14115. A quote here sums it up for me: ‘If you want a tactical solution, then Kimball is the way to go. If you want a strategic solution, then Inmon is the way to go.’ All in all I think the Kimball approach is a good one if the goal is a stand...


Why a separate surrogate key table?

A question I often (and recently) get is why you should create a separate table to manage the surrogate keys. This is compared to star- or snowflaked models where the surrogate key (and distribution) is part of the table. For Data Vault the Hub entity is the place where the keys are managed. There is rationale for splitting this process into separate tables: A surrogate key table makes it easy to identify the records in...