I'm working on my first data cube and have hit a potential roadblock. My problem might be a knowledge gap rather than a technological one, so I'm happy to consider all and any answers. Here's my situation.
I've been asked to see what BI can do for a specific client but I'm not having much luck moving forwards. Technologically speaking, I'm using a single relational database stored and maintained within SQL Server 2008 R2. I've access to BIDS, and
the end view of this is to create a DataCube that can be plugged into PerformancePoint 2007. I've gotten as far as setting up the data connections and a basic data view but from herein, I get stuck.
Because I've got the RDB, the view I've taken is to take the data with a view of applying a schema to it (star scheme would be sufficient I believe) rather than designing the schema first. Problem is, every table I can see I believe is a dimension
table, not a fact/measure table. As I'm lacking a grain statement, I'm not sure if there's a best way to move forwards. The data specifically is a case management database for general medical practitioners, logging things such as correspondence,
solicitors details and so on, not much of this is numeric and additive, although I need to get much closer to what's contained within it.
So, what can I
View Complete Post