A comprehensive querying database system based on the entity relationship model
MetadataShow full item record
This work proposes a comprehensive querying database system based on an enhanced entity relationship (EER) model. The DBMS is fully operational and performs all queries that are illustrated in the paper. This work is also applicable for the general ER model proposed by Chen. So far, the ER model is mainly used by database designers as a conceptual model during the database design phase. An ER schema is usually mapped into a representation in the logical model of the targeted database. As an analogy, the ER schema represents a program, written in a higher level language that is compiled into a lower level machine-executable equivalent. Semantics like the relationships among entities or the cardinality ratio constraints may no longer be available at the logical model level. Queries are then written against the logical model, which generates a discrepancy between the view of the database designer and the view of the database user. This work bridges this gap by introducing an EER-algebra and a high-level query language called ERSQL. The algebra is heavily based on the general ER-algebra proposed by Parent and Spaccapietra. To provide a semantic foundation for ERSQL we introduce a canonical translation algorithm that maps an ERSQL query into an EER-algebra expression. In recent years, in NoSQL data stores the functional primitives are greatly simplified for performance and scalability reasons. Our ERDBMS goes in the opposite direction: We use CISC (complex instruction set) operators but implement them efficiently in main-memory data storage.