×

Indexing techniques for advanced database systems. (English) Zbl 0893.68057

The Kluwer International Series on Advances in Database Systems. 8. Dordrecht: Kluwer Academic Publishers. ix, 250 p. (1997).
Most of data stored in databases is indexed. With appearing of many new database applications such as CAD/CAM, spatial databases, multimedia systems etc., traditional well-known techniques as, e.g., B-trees, seemed to be not sufficient to support these applications. New data structures have been developed and new indexing techniques have been used in plenty both prototypes and advanced commercial systems. The book is addressing indexing problems in five chapters; the last chapter discusses indexing problems in emerging applications.
The authors are starting with object-oriented databases in Chapter 1. Since an object-oriented model distinguishes from the relational model in many aspects, new index techniques are necessary to support object-oriented queries. The chapter includes methods for aggregation and inheritance based on various types of trees (e.g. H-tree, \(\chi\)-tree etc.).
Chapter 2 concerns spatial databases used in popular applications such as e.g. GIS. For this type of applications, multi-dimensional structures, such as grid-files, kd-trees, and R-trees were proposed. Some methods based on dynamic hashing are presented here.
In Chapter 3 the authors introduce what constitutes an image database. They continue with taxonomy of existing content-based image indexes. Four indexes that facilitate speedy retrieval of images based on colour-spatial information are presented. Three hierarchical indexes that integrate colour and spatial information for image retrieval are then discussed. Each of them forms a single index structure.
Chapter 4 covers problems of indexing temporal databases. Temporal data can be indexed as line segments. The authors review several promising indexes for temporal data such as time-split B-tree, time index, R-tree, interval B-tree and others.
With explosive growth in use of digital documents, key issues in developing new index structures concern text databases in last years. In Chapter 5 the authors describe the ways in which text databases might be accessed, kinds of queries, index structures to support these queries, and query evaluation techniques. The rest of the book shows indexing in special applications of greatest importance today. It includes techniques for parallel and distributed databases, mobile computing, data warehousing, Web, and constraint databases.
All self-contained chapters are prepared by area experts in a similar level of details and on a high level. Both practitioner and researcher can find in the book interesting points. The text in not overloaded by formal definitions, all structures are explained on appropriate examples. Of great importance is a comparing the structures within each chapter. Despite of the fact that no algorithms for maintenance of indexes are introduced, the reader can gain good fundamentals in indexing and a global view. Last but not least, the book is worthwhile particularly for newcomers because it overviews wide range of advanced indexing techniques and give many references for a more deep study.
Reviewer: J.Pokorny (Praha)

MSC:

68P15 Database theory
68-02 Research exposition (monographs, survey articles) pertaining to computer science

Software:

QBIC; Chabot
PDFBibTeX XMLCite