We can store as much duplicate data as we need in the NOSQL dbs. Data must be saved in the same pattern in which we want to access out of it. There might be cases where we don't the access pattern in which we might need to access the data e.g. Ad hoc queries.
My company works on the financial domain where ad hoc queries are used most often. Right now we use the SQL server so stored procedure are used to perform the ad hoc data calculations.
Here i mean the query which can't be predicted in advance by ad hoc query.
How we could model the this query behavior in NOSQL. Right now we are evaluating Cassandra but we are open to switch to other NOSQL databases
Any lead ?
