Just the Facts

• ADO.NET is a data-access subsystem in the Microsoft .NET Framework and is made of two distinct but closely related sets of classes data providers and data containers.

• The functionalities supplied by a .NET data provider fall into a couple of categories: the capability of populating container classes, and the capability of setting up a connection and executing commands.

• The .NET Framework comes with data providers for SQL Server, Oracle, and all OLE DB and ODBC data sources.

• A data provider is faster and more appropriate than any other database technology for data-access tasks in .NET. Especially effective with SQL Server, a managed data provider hooks up at the wire level and removes any sort of abstraction layer.

• The data provider supplies an object to establish and manage the connection to a data source. This object implements connection pooling.

• In .NET applications, connection strings can be stored in a special section of the configuration file and encrypted if required.

• The data provider supplies an object to execute commands on an open connection and optionally within a transaction. The command object lists various execute methods to account for query and nonquery commands. Commands can execute either synchronously or asynchronously.

• Data returned by a query command are cached in a data reader object, which is a kind of optimized read-only, forward-only cursor.

• Local and distributed transactions can be managed through the TransactionScope class introduced with ADO.NET 2.0.