1.7 ADO.NET

---

As its name suggests, ADO.NET is the next technology in the evolution of ADO and was released as part of the .NET Framework, the collection of technologies that supports .NET development.

ADO.NET supports typed programming. This useful feature uses a lot of end-user type words in the developer's code that are self-explanatory to anyone reading it and therefore make for easier support. It is also type-safe, since during code compilation, incorrect use of data types is picked up and reported back to the developer for remedial action, as opposed to a weakly typed environment that would report such an error at runtime only and be more difficult to track down.

While ADO may have been good enough for client/server-based development, there are some important differences between that type of development and Web-based development:

• The Web is disconnected . Managing applications that only connect to the database when they need data is different from the way in which client/server applications worked, when the connection was more or less open all of the time. Users of early versions of Microsoft Access and SQL Server realized this when upsizing Access to SQL Server. Very soon SQL Server ran out of connections since Access was consuming no end of connections per user for such things as keeping track of the table key. Thankfully, this connection limitation was removed from SQL Server a few years ago.

• Web applications need to scale . Keeping connections open to a database is expensive in terms of system resources and leads to applications collapsing with tens of users as opposed to the thousands of users of a Web application.

• New technological innovations such as XML are changing the way applications request and receive data across the Web . Connection technologies need to take this into account and support the use of XML.

To manage the disconnected nature of SQL Server data in a Web-based application, ADO.NET creates a data cache populated with the data the user wishes to see. Once the cache is populated, the database connection is disconnected until the user makes changes to the cached data, which is then sent back to the database server. The data is actually streamed in XML between the application and SQL Server, and if the data is required elsewhere, then again they are persisted (i.e., saved) in XML format.

---