The Weak Instance Model
Part of the
International Centre for Mechanical Sciences
book series (CISM, volume 347)
It is common during the design of a database to decompose, for the sake of normalization, a relation scheme R into two or more relation schemes R 1,..., R n that adequately represent R. However, the relation scheme R is the result of the modeling activity, and therefore it may represent the application requirements more closely than the normalized schemes. Also, the new database scheme may be more complex (at least because it includes more relation schemes), and queries would usually involve longer expressions, often based on the same sequences of joins. Therefore, in some cases it may be undesirable for users to have to know details of the decomposition: it would be useful to allow queries based on the relation scheme R (which does not really exist), thus liberating users from the need to know about the decomposition. Similarly, it may be reasonable to perform updates by inserting or deleting tuples (possibly defined on a proper subset of the attributes) without referring to the decomposition. The weak instance approach provides a framework that allows the user to refer to the original relation scheme and maps the external requests for queries or updates to the actual decomposed scheme. As often happens, the basic idea is quite natural, but some development is required to obtain a general and consistent framework.
A.V. Aho, C. Beeri, and J.D. Ullman. The theory of joins in relational databases. ACM Trans. on Database Syst.
, 4 (3): 297–314, 1979.CrossRefGoogle Scholar
P. Atzeni and E.P.F. Chan. Efficient and optimal query answering on independent schemes. Theoretical Computer Science
, 77 (3): 291–308, December 1990.CrossRefzbMATHMathSciNetGoogle Scholar
P. Atzeni and M.C. De Bernardis. A new interpretation for null values in the weak instance model. Journal of Comp. and System Sc.
, 41 (1): 25–43, August 1990.CrossRefzbMATHGoogle Scholar
P. Atzeni and V. DeAntonellis. Relational Database Theory: A Comprehensive Introduction
. Benjamin and Cummings Publ. Co., Menlo Park, California, 1993.Google Scholar
P. Atzeni and R. Torlone. Updating relational databases through weak instance interfaces. ACM Trans. on Database Syst.
, 17 (4): 718–746, December 1992.CrossRefMathSciNetGoogle Scholar
E.P.F. Chan. Optimal computation of total projections with unions of simple chase join expressions. In ACM SIGMOD International Conf. on Management of Data
, pages 149–163, 1984.Google Scholar
M.H. Graham and M. Yannakakis. Independent database schemas. Journal of Comp. and System Sc.
, 28 (1): 121–141, 1984.CrossRefzbMATHMathSciNetGoogle Scholar
H. Hernandez and E.P.F. Chan. Constant-time-maintainable BCNF database schemes. ACM Trans. on Database Syst.
, 16 (4): 571–599, 1991.CrossRefMathSciNetGoogle Scholar
P. Honeyman. Testing satisfaction of functional dependencies. Journal of the ACM
, 29 (3): 668–677, 1982.CrossRefzbMATHMathSciNetGoogle Scholar
M. Ito, M. Iwasaki, and T. Kasami. Some results on the representative instance in relational databases. SIAM Journal on Computing
, 14 (2): 334–354, 1985.CrossRefzbMATHMathSciNetGoogle Scholar
D. Maier, A.O. Mendelzon, and Y. Sagiv. Testing implications of data dependencies. ACM Trans. on Database Syst.
, 4 (4): 455–468, 1979.CrossRefGoogle Scholar
D. Maier, J.D. Ullman, and M. Vardi. On the foundations of the universal relation model. ACM Trans. on Database Syst.
, 9 (2): 283–308, 1984.CrossRefzbMATHMathSciNetGoogle Scholar
Y. Sagiv. A characterization of globally consistent databases and their correct access paths. ACM Trans. on Database Syst.
, 8 (2): 266–286, 1983.CrossRefGoogle Scholar
Y. Sagiv. Evaluation of queries in independent database schemes. Journal of the ACM
, 38 (1): 120–161, 1991.CrossRefzbMATHMathSciNetGoogle Scholar
J.D. Ullman. Principles of Database Systems
. Computer Science Press, Potomac, Maryland, second edition, 1982.Google Scholar
K. Wang and M.H. Graham. Constant-time maintainability: a generalization of independence. ACM Trans. on Database Syst.
, 17 (2): 201–246, 1992.CrossRefMathSciNetGoogle Scholar
© Springer-Verlag Wien 1994