In a data warehouse environment, materialized views can store aggregated data from a group
by rollup or a group by cube query. If the appropriate initialization parameter values are set, such
as QUERY_REWRITE_ENABLED, and the query itself allows for query rewrites (with the query
rewrite clause), then any query that appears to do the same kind of aggregation as the materialized
view will automatically use the materialized view instead of running the original query.
Regardless of the type of materialized view, it can be refreshed automatically when a committed
transaction occurs in the base table, or it can be refreshed on demand.
Chapter 1: Getting Started with the Oracle Architecture 21
Materialized views have many similarities to indexes, in that they are directly tied to a table
and take up space, they must be refreshed when the base tables are updated, their existence is
virtually transparent to the user, and they can aid in optimizing queries by using an alternate
access path to return the results of a query.
More details on how to use materialized views in a distributed environment can be found in
Chapter 17.
Object Views
Object-oriented (OO) application development environments are becoming increasingly prevalent,
and the Oracle 10g database fully supports the implementation of objects and methods natively in
the database. However, a migration from a purely relational database environment to a purely
OO database environment is not an easy transition to make; few organizations have the time and
resources to build a new system from the ground up.
Pages:
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88