Design Considerations for a Prolog Database Engine

作者: K.F. Wong , M.H. Williams

DOI: 10.1016/B978-1-4832-1313-2.50016-0

关键词: Process (computing)WorkstationResolution (logic)Set (abstract data type)PrologSelection (relational algebra)Programming languageDatabaseDatabase engineComputer scienceUnification

摘要: One major hindrance to using Prolog for implementing large databases is the slow speed of unification. The size physical memory required store a database renders main impractical and sets clauses need be stored on secondary disk memory. Full unification performed directly resident would unacceptably slow. For this reason, special purpose machine being designed which will provide fast filtering system. system, known as CLARE (CLAuse Retrieval Engine), discriminantly retrieves all potentially unifiable queries provides small resolution set full system based two-stage process aspects design are discussed. Three configurations suitable implementation studied. studies enable selection most cost-effective configuration applicable workstation-based

参考文章(5)
Jayasooriah, R. M. Colomb, A clause indexing system for PROLOG based on superimposed coding Australian Computer Journal archive. ,vol. 18, pp. 18- 25 ,(2000)
Kotagiri Ramamohanarao, John Shepherd, A Superimposed Codeword Indexing Scheme for Very Large Prolog Databases international conference on logic programming. pp. 569- 576 ,(1986) , 10.1007/3-540-16492-8_105
W. F. Clocksin, Design and simulation of a sequential Prolog machine New Generation Computing. ,vol. 3, pp. 101- 120 ,(1985) , 10.1007/BF03037117
K. F. Wong, M. H. Williams, Limiting a resolution set in a large Prolog database using special hardware Proceedings of the 1988 ACM SIGSMALL/PC symposium on ACTES. pp. 133- 145 ,(1988) , 10.1145/62453.62484
Nam Sung Woo, A hardware unification unit: design and analysis international symposium on computer architecture. ,vol. 13, pp. 198- 205 ,(1985) , 10.1145/327070.327203