Triantis, Konstantinos P.Egyhazy, Csaba J.2013-06-192013-06-191987http://hdl.handle.net/10919/20265This paper presents a framework for the study of the query decomposition translation for heterogeneous record -oriented database management systems. This framework is based on the applied database logic representation of relational, hierarchical and network databases. The input to the query decomposition translation is the query graph which is derived from the complex to basic, external to conceptual and logical optimization translations. Once the query graph is obtained the objective of the query decomposition translation is to break up a query expressed in terms of the actual or conceptual databases into its component parts or subqueries and find a strategy indicating the sequence of primitive or fundamental operations and their corresponding processing sites in the network necessary to answer the query. The query processing strategy is usually chosen so as to satisfy some performance criterion such as response time reduction. Contingent on after each primitive operation. The prequery decomposition translation, the query decomposition translation and the size estimation issues are presented through an example based on the current implementation of the Distributed Access View Integration Database (DAVID) currently being built at NASA's Goddard Space Flight Center (GSFC). The choice of a query processing strategy is the successful estimation of intermediate resultsapplication/pdfenIn CopyrightTechnical reportTR-87-31http://eprints.cs.vt.edu/archive/00000078/01/TR-87-31.pdf