An important problem in the physical design of databases is the selection of secondary indices. In general, this problem cannot be solved in an optimal way due to the complexity of the selection process. Often use is made of heuristics such as the well-known ADD and DROP algorithms. In this paper it will be shown that frequently used cost functions can be classified as super-or submodular functions. For these functions several mathematical properties have been derived which reduce the complexity of the index selection problem. These properties will be used to develop a tool for physical database design and also give a mathematical foundation for the success of the before-mentioned ADD and DROP algorithms.
Intending to develop a tool which aims to support the physical design of relational databases can not be done Without considering the problem of index selection. Generally the problem is split into a primary and secondary index selection problem and the selection is done per table. Whereas much attention has been paid on the selection of secondary indices relatively less is known about the selection of a primary indez and the relation between them. These are exactly the topics of this paper.491 0-818&4212-2/93 $03.00 8 1993 IEEE Ip, M.YX., Saxton, L.V., Raghavan, V.V., On the Selection of an Optimal Set of Indexes, in IEEE TSE, Fbzen, S., Shasha, D., A Framework for Automating Database
The design of an optimol physical database entails art expo-
A n operation in object-oriented databases gives rise to the processing of a path. Several database operations may result into the same path. W e address the problem of optimal index configuration for a single path. As it will be shown an optimal index configuration for a path can be achieved by splitting the path into subpaths and b y indexing each subpath with the optimal index organization. We present an algorithm which is able to select an optimal index configuration for a given path. FOT the moment we consider a limited number of existing indexing techniques (simple index, inherited index, nested inherited index, multi-index, and multiinherited index) but the principles of the algorithm will remain the same adding more indexing techniques.
Abstract. We describe a tool for physical database design based on a combination of theoretical and pragmatic approaches. The tool takes as input a relational schema, the workload defined on the schema, and some additional database characteristics and produces as output a physical schema. For the time being, the tool is tuned towards Ingres. IntroductionThe design of databases takes place on several levels. One of these levels is the physical level. Typical subproblems on this level are, among others, selection of storage structures, secondary index selection, vertical fragmentation, materialization, etc. Solving these subproblems requires a sophisticated understanding of physical design options and query optimization strategies of the optimizer, and involve estimating query costs, which is a tedious and error-prone process when done manually. Moreover, several subproblems are NP-complete, such as the selection of an optimal set of secondary indices. Research in this area has been shifted to the problem of determining a good physical design instead of an optimal design [2,5]. A physical design is considered as good if a competent human database designer would produce the same or a worse design with the same available information. We present a tool, called TOPYDE, that takes as input, among others, a relational schema, the workload defined on the schema, and other database characteristics, such as page size, cardinality of a relation, etc., and produces for each relation a storage structure (including an ordering attribute or clustering index) and a set of secondary indices. This is called a physical schema. An overall physical schema is obtained by the union of the physical schema of each relation involved in the relational schema. For the time being, ordering attributes and indices concern single attributes, and a secondary index is stored as a Btree.Although TOPYDE does not cover the overall problem of physical design, it covers the most crucial parts. Moreover, TOPYDE can be easily extended with vertical fragmentation and materialization. We agree with Navathe et al. [7] that vertical partitioning precedes the selection of a physical schema. In [7], vertical fragmentation algorithms are presented that partition a relation into a set of fragments. Such algorithms can serve as a preprocessor for TOPYDE. In practice, materialization is often done as last; this means after the selection of a physical schema. So, TOPYDE can be extended by a postprocessor that aims to improve the physical schema selected by it.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.