An adaptive approach to relational data base systems through evolutionary modeling
Date Issued
August 1977
Author(s)
Dougherty, James M.
Advisor(s)
James R. Pinkert
Additional Advisor(s)
Sara R. Jordan
J. Michael M.
Abstract
According to published descriptions, current relational data base management systems assume a static configuration of stored relations. Typically, that configuration is obtained by a decomposition of the original data according to functional dependencies inherent in it. In this thesis, the author shows that such an arbitrary, static configuration may be considerably sub-optimal with respect to several cost measures. Techniques modeling evolution are applied to the problem; improvements in cost measures are made by having the storage configuration adapt to the user queries. A sample implementation of these techniques is described, and empirical results are discussed.
Degree
Master of Science
Major
Computer Science
File(s)![Thumbnail Image]()
Name
Thesis77.D684.pdf
Size
7.93 MB
Format
Adobe PDF
Checksum (MD5)
44cbe6d9698fb25b52b078319f076521