[This is prerelease documentation and is subject to change in future releases. Blank topics are included as placeholders.]

There are several stages in creating model-driven applications. In code name “Oslo” modeling technologies, you typically start by modeling one or more problem domains in Microsoft code name “M”. The tools in the code name “Oslo” SDK load these models into the code name “Oslo” repository, a SQL Server 2008 database designed to store and manage these models. The next step is to write various types of applications that operate with the models in the database. These applications include loaders that import data into the new models, management tools that help manage model data, and model-driven applications that read the models to direct their operation. The topics in this section provide examples of various data access strategies for creating, reading, updating, and deleting model data in the “Oslo” repository.

There are special “Oslo” repository patterns that must be understood when writing applications that interact with models. These patterns support security features, performance, and other services. For example, a common pattern involves the use of Folders and the creation of updatable security views. A Folder is a logical container for “Oslo” repository rows, also referred to as items. Updatable views use triggers to enable updates to views that are normally read-only. Users interact with the views instead of the underlying tables, allowing security checks that permit users to only interact with items belonging to Folders that the user has been granted permission to use. For more information about Folders, see "Oslo" Repository Folder Design Patterns. It helps to understand these types of “Oslo” repository patterns when writing data access code. Models that use Folders and updatable views require code that creates and queries Folders; the code must also operate on the views instead of the base tables.

In This Section

See Also