The
aim of the presented framework is to facilitate, manage, and optimize the design
and implementation of the ETL workflows in order to create an optimal workflow.
The framework supports all the phases of ETL design, from the initial design to a
deployment stage and utilization, under continuous evolution of a data warehouse.
The chapter contains also a comprehensive state of the art on commercially available
tools and research achievements in the field of ETL.
Section III.describes challenges and solutions to the problem of assuring the effi-
ciency of analytical processing. Fundamental research and technological solutions
to this problem include optimization techniques of star queries, indexing, partitioning,
and clustering.
Chapter VI, Advanced Ad Hoc Star Query Processing, by Nikos Karayannidis,
Aris Tsois, and Timos Sellis, focuses on efficient processing of OLAP queries.
OLAP applications rely heavily on the so called star queries that join fact tables
with multiple dimension tables. Reducing execution time of such joins is crucial
for a DW performance. To this end, a new approach to fact table organization has
been developed, called a hierarchical clustering. The hierarchical clustering allows
clustering of fact data according to paths in dimension hierarchies. This clustering
technique exploits path-based surrogate keys.
Pages:
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36