't understand there is a problem, so I want to ask you, the use of traditional relational database to build several general warehouse, will use modeling paradigm modeling or partial paradigm, but if it is a basic data platform to build the data warehouse, also need to use modeling paradigm? Or can directly use the dimension modeling is ok?

CodePudding user response:

We create the traditional data warehouse are now rarely used paradigm, dimensional modeling is relatively popular in the way, as for the large data, this depends on your business requirements and considering the amount of data, it is not necessary to fully use dimensional modeling,

CodePudding user response:

refer to the original poster A362762533 response:

there is a problem don't understand, so want to ask you, the use of traditional relational database to build several general warehouse, will use modeling paradigm modeling or partial paradigm, but if it is a basic data platform to build the data warehouse, also need to use modeling paradigm? Or can directly use the dimension modeling is ok?

Big data platform to build the data warehouse, do not use paradigm modeling, modeling and are now latitude CIF mix build data warehouse platform

CodePudding user response:

Simple answer:
(1) the big data can not modeling modeling does not mean not specification), but the late maintenance cost is high, the low efficiency
2) if the modeling: a, if use the traditional DW for data mining, the use of dimensional modeling; B, if the use of modern technology of data mining, using denomalization can also,