Database Reference
In-Depth Information
about dimensions and their attributes and versions
of attributes and dimension tables are maintained
in separate tables. Finally, for information about
retrieval from data warehouse through integrated
schema and mapping of attributes of integrated
schema to versions of data warehouse are stored
in other tables.
Two types of approach are used depending
on the type of users and required functionality:
1) for expert users (skilled and capable of writing
complex queries), the “SQL extensions” approach
is used, based on SQL with versioning-related
extensions; 2) for basic users, who which to use
basic SQL syntax only, the “transparent querying”
approach is available.
retrIevAl froM the
MultIverSIon dW
Sql extensions Approach
According to this approach, extensions to the
query language are proposed by adding clauses
to conventional SQL. Users can query the cur-
rent version of the data warehouse or a set of
data warehouse versions. A query that deals with
a single version of the data warehouse is called
mono-version query. In contrast, a query written
to retrieve data from multiple versions is called
multiversion query.
Retrieval and use of complete and consistent
information for decision-support is the original
objective of versions. Information required for
user is spread across versions and users may not
be able to retrieve this information (Golfarelli,
2006). This is due to the inability of conventional
SQL to be used for retrieval of data from multiple
versions simultaneously.
Figure 9. Generic metamodel for metadata about MVDW
Search WWH ::
Custom Search