We need to have reporting / analytical database for our application. Our application will have reporting and analytical capability against these database. What i understand from CB analytics is that it will have a separate node for reporting replicating the same operational tables. My operational tables are not highly denormalized. And it is not suited to build reporting large data with filters, aggregations and grouping no way it could be used for slicing the data by dimensions against facts as we do in DW.
Can any one help with the approach or directions usually need to be taken in the context of nosql and CB analytics.
Can you be more concrete? I am not concerned about the physical layer.
Can you please relate to the industry practice in today’s world ?. This is not helping answering the question.
To me Document, collection etc are logical.
How would i design schema for an analytical database. I am not talking about big data.
I am talking about the analytical db schema for my operational tables.? I understand that in a relational world you may go with highly denormalized tables or you go with dimensions and facts which is more structured.
What are the approaches CB customer generally follow? any design patterns etc.