Candidate architecture
There are several tools that we need, to handle the issues presented.
1. We must extract existing data from multiple relational systems and load it into a clean environment where we will track data access and changes.
2. We must manage the definitions of key business nouns across multiple environments.
3. We must track metadata concerning various documents, spreadsheets, and other objects.
4. We must use unstructured and semi-structured data as a source for data mining and data analysis tasks by storing metadata concerning these objects in SQL Server.
5. We must search through and index semi-structured and un-structured data across the enterprise. This will include every server, every laptop, and all of the miscellaneous storage devices.
These needs cannot be met with any single product or technology in the Microsoft catalog. Rather, we will need to incorporate several technologies into our solution to gather, index, store, and present data to end users.
