Information Model Creation Issues

When you insert an information model into a repository, you have the following choices and decisions to consider:

• How much of the tool information will you store in the repository database and how much will you store in other files?
  
  
• For each class you define in your information model, what Automation server will create instances of that class?
  
  
• Can you tune the database schema to improve performance?
  
  
• Should you create a new information model or extend an existing one?
  
  
• Should you assign object identifiers, or let the repository engine do this for you?
  
  
• How will your information model accommodate navigation from one object to another?

The following topics discuss each of these questions.

Topic	Description
Extending vs. Creating Information Models	Provides guidelines to help you decide whether to create or extend an information model
Choosing Which Information Belongs in the Repository	Provides guidelines to help you decide where a Microsoft® SQL Server™ 2000 Meta Data Services repository fits into your development environment
Choosing an Automation Server for a Class	Explains how to instantiate an Automation server in your application code
Tuning the Database Schema of an Information Model	Provides tips that can help you determine how the repository database is created
Accommodating Navigation Within an Information Model	Explains the core requirements for building a navigable information model
Assigning Object Identifiers	Explains the ways in which a repository identifier can be assigned to an object

See Also

Creating Type Information Programmatically