Database Reference
In-Depth Information
Chapter 7
Physical Data Warehouse Design
The physical design of data warehouses is crucial to ensure adequate query
response time. There are typically three common techniques for improving
performance in data warehouse systems: materialized views, indexing, and
partitioning. A materialized view is a view that is physically stored in a
database, which enhances query performance by precalculating costly oper-
ations such as joins and aggregations. With respect to indexing, traditional
techniques used in OLTP systems are not appropriate for multidimensional
data. Thus, alternative indexing mechanisms are used in data warehouses,
typically bitmap and join indexes. Finally, partitioning or fragmentation
divides the contents of a relation into several files, typically based on a range
of values of an attribute.
In this chapter, we focus on a relational implementation of the data
warehouse and the associated data cubes. We first give in Sect. 7.1 an
introduction to the problems stated above. Then, in Sect. 7.2 , we study
the problem of computing and maintaining materialized views. In Sect. 7.3 ,
we study the data cube maintenance problem and discuss in detail the
classic algorithms in the field. Section 7.4 studies ecient ways of computing
the whole data cube, while Sect. 7.4.3 studies classic algorithms aimed at
materializing only a portion of the data cube. Section 7.5 studies data
warehouse indexing techniques in detail, while Sect. 7.6 discusses how indexes
are used to evaluate typical data warehouse queries. Section 7.7 overviews
data warehouse partitioning issues. Section 7.8 studies physical design
support in SQL Server and in Analysis Services, while Sect. 7.9 briefly
discusses query optimization in Analysis Services. Finally, Sect. 7.10 discusses
query optimization in Mondrian.
Search WWH ::




Custom Search