Thesis--University of Maryland. / Includes bibliographical references (p. 159-176).

Thesis (M.S.C.I.T.)--Regis University, Denver, Colo., 2006. / Title from PDF title page (viewed on Sept. 7, 2006). Includes bibliographical references.

Thesis (M.S.)--State University of New York at Binghamton, Computer Science Dept., 2005. / Includes bibliographical references.

This paper will examine two data base management systems; IMS, an example of a hierarchical data base management system and System R, a relational system. Each system will be described in general terms followed by a discussion of their relative performance. A claim is made that IMS, because of its structure and the procedural nature of its language can be more efficient than System R. Some examples are given to illustrate this claim. There is an increasing need today for readily available information; thus more and more information is being stored in disk files to facilitate rapid retrieval. A relational approach to data base management permits a high-level non-procedural interface for the user, which is important if more people require access to information in a flexible fashion and can not afford or wait for traditional application development. If however a relational data base management system is not as efficient as hierarchical or network systems then new retrieval methods must be found. Associative processors are introduced as a possible solution to this problem and three examples are discussed. RAP, a relational associative processor developed at the University of Toronto, has been used in a performance study to demonstrate the dramatic performance improvements offered by associative processors. This is also discussed. / Science, Faculty of / Computer Science, Department of / Graduate

The correctness (or integrity) of a database may be destroyed when the database is accessed concurrently by a number of independent transactions. Part of the job of a database management system is to control concurrency so that correctness is guaranteed. This thesis examines concurrency controls in general, and in particular those provided by the Educational Data Base System (EDBS). It is concluded that EDBS facilities are not sufficient to guarantee correctness. Several alternatives are proposed whereby correctness can be guaranteed. An approach is selected based on ease of implementation rather than on the degree of concurrency provided. Guaranteeing correctness and at the same time providing a high degree of concurrency turns out to be a very difficult problem both in theory and in practice. / Science, Faculty of / Computer Science, Department of / Graduate

No description available.

No description available.

With the advent of large and complex applications and the emergence of semi-structured information repositories such as the World Wide Web, new demands are being made on database systems. The TENTACLE database system is an experimental database system which provides facilities capable of meeting some of these demands. The distinguishing features of the system are that it: uses a graph-based data model (and storage subsystem) to provide a flexible means of representing poorly structured information, integrates a path expression-based query language with a general purpose language to query and manipulate the graph structures, thereby eliminating the impedance mismatch encountered in a two language system, and provides a programmable database kernel capable of executing the combined query and utility language, allowing the construction of domain specific applications inside the database without the assistance of wrappers or gateways. As a demonstration of the utility of the system, I have constructed a hypertext server inside the TENTACLE database without making use of external mediators or gateways. Since the hypertext server program is part of the database content, database facilities may be used to assist in the creation and maintenance of the hypertext server itself. In addition, the close integration of hypertext server and database simplifies tasks such as the management of associations between hypertext entities or the maintenance of different document views.

by Chu, Kam Wing. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 67-73). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgments --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Related Work --- p.6 / Chapter 3 --- Time-Series Searching with Scaling and Shifting --- p.12 / Chapter 3.1 --- Problem Statement --- p.15 / Chapter 3.2 --- Preliminary --- p.16 / Chapter 3.3 --- Geometrical View of the Problem --- p.18 / Chapter 3.3.1 --- Scale-Shift Transformation --- p.21 / Chapter 3.3.2 --- Determine Scaling Factor and Shifting Offset --- p.24 / Chapter 3.4 --- Algorithm --- p.25 / Chapter 3.4.1 --- MBR Penetration --- p.26 / Chapter 3.4.2 --- Long Sequence --- p.28 / Chapter 3.5 --- Implementation Details --- p.29 / Chapter 3.5.1 --- MBR Penetration Checking --- p.29 / Chapter 3.5.2 --- Dimension Reduction --- p.32 / Chapter 3.6 --- Experimental Results --- p.34 / Chapter 4 --- Metric Space Indexing for Multimedia Databases --- p.38 / Chapter 4.1 --- Preliminaries --- p.39 / Chapter 4.1.1 --- M-tree --- p.39 / Chapter 4.1.2 --- Range Queries --- p.41 / Chapter 4.1.3 --- Nearest Neighbor Queries --- p.44 / Chapter 4.2 --- Nearest Neighbor Search by dmin Only --- p.46 / Chapter 4.3 --- Analysis --- p.50 / Chapter 4.3.1 --- Critical Factor dmin --- p.52 / Chapter 4.4 --- Multiple Bounding Regions --- p.54 / Chapter 4.4.1 --- Computing Multiple Bounding Regions --- p.56 / Chapter 4.4.2 --- New Insert Strategy --- p.58 / Chapter 4.5 --- Experimental Results --- p.58 / Chapter 5 --- Conclusions --- p.64 / Chapter 5.1 --- Time-Series Searching with Scaling and Shifting --- p.64 / Chapter 5.2 --- Metric Space Indexing --- p.65 / Bibliography --- p.67

Thesis (Ph.D.)--OGI School of Science & Engineering at OHSU, 2003. / Includes bibliographical references (leaves 174-181).