Chapter 4. Using the cube model for summary tables optimization 127
The optimizer must recognize the existence and applicability of the summary
tables to these queries and automatically rewrite the queries to access the
pre-aggregated data.
Summary tables are a powerful performance feature. They are typically
considerably smaller than the cross product of the base tables on which they are
defined. Because of this and the fact that they contain pre-aggregated data,
queries requesting aggregates may experience dramatically improved
performance through their use.
DB2 provides support for summary tables through its Materialized Query Tables,
or MQTs. Its implementation of MQTs is more generalized than just summary
data. DB2 permits a materi ...