Description
An analytics data warehouse stores essentially the same data generated by transactional databases — nothing more, nothing less. But whereas a transactional database may be designed to be as fast as…
Summary
- Properly Layering your Data into L1, L2 and L3 Layers An analytics data warehouse stores essentially the same data generated by transactional databases — nothing more, nothing less.
- The goal of the analytics data warehouse is to enable a data scientist (or interchangeably analyst) to rapidly mix data to answer complex questions and discover useful information that is not obvious.
- Another example is “How many orders did we have by business segment”?
- If it takes the analyst more than 2 minutes to quickly review the papers on the wall and write out the basic pseudo-code (or if you know the pseudo-code is wrong because the documentation is out of date) then you don’t have a good data model or proper documentation.