I am importing information from various source systems and putting them into local databases for reporting. For example, one of the sources is Google Analytics.

Here are some attributes of this import process:

  1. I am collecting data from the source system (e.g. Google Analytics).
  2. The data goes into my local database.
  3. The data flows in one direction, i.e. the only write access on the local database is the importer process.
  4. The local database is a view of the source system. i.e. it is not a full-fidelity copy of the upstream database.

Given these attributes, would my local database be called a data mart, data vault or data warehouse?

This word choice is important to me because our company will be publishing our tools as open source. If we describe what our tool does using the best industry-accepted terminology then it will be more valuable.


Having checked with the respective Wikipedia articles and from my experience (although it is limited in that area), I propose to go with data warehouse.

  • Data Warehouse This term is possibly best applicable to the application under description. Data warehouse is also closely linked to ETL processes which means if you are publishing ETL tools, people will understand them to work in cooperation with a data warehouse...
  • Data Vault This term seems to focus on the collection of historical data
  • Data Mart This term focuses on being a subset of an actual data warehouse.

I admit, that up to this question, I had never heard of "data vault". Finally, before publishing and chosing the "best" name, check with Marketing if they have any preferences on the name -- tools need not be always called after what they are doing (although I prefer that style).


Not the answer you're looking for? Browse other questions tagged or ask your own question.