Multidimensional databases (cubes) are another storage
option provided by the SAS Intelligence Platform. Cubes provide business
users with multiple views of their data through drill-down capabilities.
Cubes are derived from
source data such as SAS tables, SAS SPD Engine tables, and
SAS/ACCESS
database tables. To create cube definitions, and to build cubes based
on these definitions, you can use the
Cube Designer wizard, which is available from SAS Data Integration Studio and
SAS OLAP Cube Studio.
Cubes are managed by the
SAS OLAP Server, which is a multi-user, scalable server designed to
store and access large volumes of data while maintaining system performance.
The SAS OLAP Server
uses a SAS engine that organizes data into a streamlined file format.
This file format enables the engine to rapidly deliver data to client
applications. The engine also reads and writes partitioned tables,
which enables it to use multiple CPUs to perform parallel I/O functions.
The threaded model enables the SAS OLAP Server to create and query
aggregations in parallel for fastest performance.
SAS business intelligence
applications perform queries against the cubes by using the multidimensional
expression (MDX) query language. Cubes can be accessed by client applications
that are connected to the SAS OLAP Server with the following tools:
-
the SQL pass-through facility for
OLAP, which is designed to process MDX queries within the PROC SQL
environment
-
open access technologies such as
OLE DB for OLAP and ADO MD