Chapter X, OLAP with a Database Cluster, by Uwe R?¶hm, presents clustered data
warehouse architecture as an alternative to the approach discussed in Chapter IX.
The clustered architecture is based on a cluster of commercial off-the-shelf computers
as hardware infrastructure that run off-the-shelf database management systems
as transactional storage managers. In this architecture, the same data may be distributed
over several nodes by using replication mechanisms. As a consequence,
some replicas may be out of date. In order to handle queries on outdated data, the
author proposes an approach to replication management, called freshness-aware
scheduling, that introduces a new quality-of-service parameter. This parameter
allows the specification of an explicit freshness limit in queries having an impact
on replicas used in these queries. Based on the value of quality-of-service a query
may be routed to the most appropriate node. Multiple query routing strategies and
physical data design alternatives are also discussed in this chapter.
Section IV.focuses on new domains for applying the data warehouse and OLAP
technologies. These novel domains pose new challenges, among others in data
modeling techniques, assuring analytical processing efficiency as well as in data
storage/organization techniques.
Chapter XI, Towards Integrating Data Warehousing with Data Mining Techniques,
by Rokia Missaoui, Gana?«l Jatteau, Ameur Boujenoui, and Sami Naouali, addresses
xviii
problems and presents solutions for coupling data warehousing and data mining
technologies.
Pages:
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39