I'm working on developing a Logical Datamodel for a customer(Bank). As a first step I'm understanding the Business process of the customer, I feel the documentation provided for me is good.
I need your kind suggestions in developing a optimal (Snowflake schema) logical datamodel and the customer currently having around 100+ applications, from which I need to find the required source tables.
Please suggest me the possible best ways,
1 In developing a logical data model 2 In getting the source tables details
Since I'm doing this task for the first time getting scared to carry the required tasks in building the Data mart.
Please provide me your valuable suggestions.
Thanks In advance.
asked 19 Apr '12, 08:32
This question is far too large in scope to answer in this forum. I suggest you take a look at some data and systems analysis textbooks to give you an idea of what is required. You'll at least need to create a logical data model which you can iteratively modify as you discover additional requirements as you learn one application and go on to the next. Sybase PowerDesigner is one package that can help you do that (there are others), and it has the ability to generate physical designs and all the SQL you'll need to create all the tables, referential constraints, and so on.
One important item rarely mentioned in introductory texts is that you'll want to synthesize application requirements into both the logical and physical designs, particularly with things like the choice of primary keys and potential denormalization of the schema in order to reduce query complexity.
answered 19 Apr '12, 09:26