The Microsoft® Data Warehouse Toolkit: With SQL Server™ 2005 and the Microsoft® Business Intelligence Toolset

13.4. A Practical Metadata Approach

This brings us to the question you've all been asking: What do we do about metadata to support data warehousing and business intelligence in the SQL Server 2005 environment? In the long term, we expect Microsoft to tackle the metadata management problem. Meanwhile, you have to figure out what you are going to do about metadata in the short to medium term. It's easy to get trapped in the metadata morass (as the authors can certainly attest). It's a major effort to figure out what metadata to capture, where to capture it, how to integrate it, how it should be used in the warehouse processes, and how to keep it synchronized and maintained. Vendors have been building metadata repositories and maintenance utilities for decades, and companies (some companies) have been trying to use these tools, or tools of their own creation, to tame the metadata beast for just as long. Even so, there are very few examples of large-scale, robust, successful metadata systems. It's a really hard problem.

Again you ask, "So what am I supposed to do?" Well, first, you need to appoint someone on the team to the role of metadata manager. If no one owns the problem, it will not be addressed. The metadata manager is responsible for creating and implementing the metadata strategy. The ideal candidate has to know everything. No joke. If one person has to do the whole thing, he or she will need to have SQL and DBA skills. The metadata manager needs to know how to program in ...