|
|
Thesis: "Using Materialized
Views in Data Warehouses"
|
|
Abstract (full
text not available)
|
|
Decision support functions,
such as On-line Analytical Processing, issue queries that need to aggregate
large volumes of data present in a data warehouse. To improve the performance
of these queries, warehouses typically contain tables that contain data in
a summarized form. As new data is brought into the warehouse, these
summary tables are updated appropriately. Two factors limit the number
of summary tables that can be created: the amount of disk space required
by the tables and the processing time incurred to update the tables in response
to new data.
We examine how materialized views can be used to store and manage summary
data efficiently. This thesis discusses the design and implementation
of the Queen's Warehousing system which provides facilities to create materialized
views for certain aggregate combinations selectively and to maintain the views
using incremental maintenance techniques.
|
|