I am new to dimensional data modeling. I have come across scenarios where I have to create one to many relationship between 2 dimensions and in one scenario I did create a one to many from the Fact table to a dimension table.
I understand these 2 scenarios are not ideal for dimensional modeling but I want to understand the disadvantages of having these in the dimensional model.
Thanks
It is hard to advise unless you share the actual problem, i.e. what business problem/questions are you trying to model for?
If you need a 1-* relationship between dimensions, that means probably that you should consider denormalizing them into a single dimension, because the functional dependence between that data is likely to exist already.
A 1-* between dimensions and facts, in which one dimension is related to multiple facts, is typical; but the other way around is uncommon.
Data modelling, namely the dimensional approach, is intimately related to the business problem. It's easier (and paramount) to start from there and then model the data to answers the business questions at hand.