Thursday, March 2, 2017

The Trouble with Data Warehouse Analytics

My February post @All Analytics

Warehouses are essentially databases biased for some data applications (and against others) and are rooted in poor database foundation knowledge and logical-physical confusion. Even when warehouses consist of relations, warehouse developers often do not understand their precise interpretation. The design is based on various unwarranted or false assumptions about what the data means. The above modeling methods do not allow documenting the transformation itself -- the relational algebra operations that comprise the transformation. But more often than not warehouses do not consist of relations, which are minimally required to be in first normal form (1NF), and are, therefore, not just denormalized, but non-relational. Consequently, all bets are off; sound derivations of correct analytical results are not guaranteed.

Read it all.

Do you like this post? Please link back to this article by copying one of the codes below.

URL: HTML link code: BB (forum) link code:

1 comment:

  1. Today people have problem understandig science (based on mathematics - logic, sets) and some srbitrary concepts that are based on someone's authority or 'industry standard' devoid of any logical framewor. (Hello Object oriented and data warehousing ;-) )

    Data Warehousing 'science' suffers from lack of logic. It contradicts itself: it is supposed to be used for ad hoc querying, yet it requires aggregation and assumption about intended use. It is built for 'speed' yet cube queries take hours to execute - no kidding - this is from a Microsoft sponsored course. Data cleaning is the most laughable requirement - isn't data already clean if it comes from a relational database? What is the purpose of keeping data somewhere if it is not clean? I it was not clean at the moment of entry into database/file/spreadsheet, how are we going to clean i? Connecting data from heterogeneous sources. Like SQL databases, Access files, Excel tables and Word documents? Thank you, but no thank you. Analytic for business? Good luck with present skill level of executives and decision makers. Since executives are not capable doing simplest things in spreadsheets, we must invent mambo-jumbo arbitrary concepts, and make it look serious and scientific, to keep our jobs? Again, thank you but no thank you.