Data Access

Microsoft has developed the UDA architecture based on COM, which is a binary standard for objects that defines how an object should present itself to the system after it has been compiled from its target language into machine code. Defining a standard allows objects to be compatible regardless of their source languages. The UDA architecture allows applications to efficiently access data where it resides without replication, transformation, or conversion.

The strategy behind implementing distributed connectivity is to assure open, integrated, standards-based access to all types of data (SQL, non-SQL, and unstructured) across a wide variety of applications, from traditional client/server to the Web. Under UDA architecture, ActiveX Data Objects (ADO) is the high-level interface that most applications developers use. OLE DB providers are the data-access engines or services, as well as the business logic components, that these applications can use in a highly interoperable, component-based environment.

The two fundamental components of UDA architecture are the ODBC standard and the OLE DB standard. ODBC unifies access to relational data from heterogeneous systems. Any application that supports ODBC can access information stored in any database that houses relational data, including Oracle. If all of the data were stored in relational databases, integration could be solved by using ODBC only. However, much data is nonrelational, or unstructured (for example, audio and video clips, and e-mail messages). To simplify integration of unstructured data across the enterprise, Microsoft offers OLE DB.

OLE DB is a set of OLE interfaces that provides applications with uniform access to unstructured data regardless of type or location on the network. Developers can write applications that connect to any OLE DB provider, whether a file system, Oracle database, Microsoft Excel spreadsheet, or DB2 database, and can allow end users running Windows-based desktop applications to share and manipulate data stored there.

OLE DB is based on COM and provides:

• The ability to expose all types of data through a set of standard interfaces.
• Seamless integration between consumer applications and controls, data providers, and data sources.
• Plug-and-play usage of service providers, such as query processors and cursor engines. Query processors compatible with OLE DB can be smoothly integrated into the data access environment and can serve multiple consumer applications and data providers.