Design Goals in This Version
OLE DB for OLAP is a set of Component Object Model (COM) objects and interfaces designed to facilitate communication between providers and consumers of multidimensional data. OLE DB for OLAP is an integral part of OLE DB version 2.0, and has many of the same design goals.
The following are specific features of OLE DB for OLAP:
-
OLE DB for OLAP interfaces enable representation, expression, transportation, and efficient navigation of multidimensional data within COM.
-
OLE DB for OLAP is optimized, to the extent that such optimizations do not represent tradeoffs, for all of the following configurations:
-
A single process space in which the consumer (or service provider) and the multidimensional data provider (MDP) run in the same process. An example of this configuration is an OLE DB for OLAP provider and the query engine of an RDBMS running in the same process.
-
A consumer and an MDP running on different machines or in separate processes. An example of this configuration is a pivot table control working with an MDP.
-
Service providers exposing OLE DB for OLAP interfaces and consuming other OLE DB interfaces.
-
OLE DB for OLAP leverages the OLE DB architecture. OLE DB for OLAP interfaces can interconnect so that the same interfaces are applicable at various levels of data providers and service providers.
-
OLE DB for OLAP reuses as much of OLE DB as possible. The number of OLE DB for OLAP–specific objects and interfaces are minimized.
-
OLE DB for OLAP is designed for interoperable access. That is, OLE DB for OLAP interfaces cover common consumer requirements for processing multidimensional data, but these interfaces are factored in a way that enables existing MDPs to expose OLE DB for OLAP without implementing changes at the server level.
-
OLE DB for OLAP is designed to enforce symmetry between dimensions and measures. In other words, the interface exposes measures as a dimension with functionality equivalent to all other dimensions.