I'm starting a bi project using VS2010 and C#/.Net 4 to analyse and monitor an important app. I need to keep every event or action inside the app in a historical data base so i can tell what happened excactly for a given factor (user, time, invoice,
I've found in the internet that star schema is the best approach to gain in effeciency and performance. With no experience in the subject, i would appreciate some tips and the tools offered by Microsoft to resolve this.
I've created a star schema and implement it in a relational data base on SQL Server 2008.
I've used Business Intelligence Developpement Studio to use the tables as an OLAP Cube by specifing facts and dimensions.
Can somebody correct or confirm my approach? Here my point of view and questions :
Is what i've mentionned over here correct? nothing missed ?
How to populate the cube? I will create a componant handling the mapping between logs and relational database (my star schema) to feed it in real time.
Do i need to process the cube after every insertion ? If yes, i will use a batch to feed/process the olap cube.
Using C#, and Microsoft;AnalysisServices and ADOMD.NET are enough to build programatically multidimensional queries and get the results to show them in a GUI ?
Please, forgive my stupid questions, but since i'm short in time i need someone experienced with BI and SQL Server to advice me.
Microsoft is conducting an online survey to understand your opinion of the Technet Web site. If you choose to participate, the online survey will be presented to you when you leave the Technet Web site.