Analysis and performance of inverted data base structures
Communications of the ACM
An Introduction to Database Systems
An Introduction to Database Systems
Computer Database Organization, 2nd Ed
Computer Database Organization, 2nd Ed
Design of Man-Computer Dialogues
Design of Man-Computer Dialogues
Design of Database Structures
| Hi-index | 0.00 |
This paper reports research undertaken to design and implement a B+-tree-based keyed sequential-access method (KSAM). KSAM provides primary and secondary access, which can be based on direct or sequential processing. Primary access to a data file requires three levels of indexes: super, master, and primary indexes. Secondary access requires an additional index level: secondary indexes. The superindex and master indexes are transparent to the user and are used solely by the system. The primary index is organized as a B+-tree containing proper linkages to the respective data files. In the implementation of secondary indexes a file is used to store accession lists of the secondary indexes, and each secondary index is in turn organized as a B+-tree containing proper linkages to accession list files. Thus, linkage from the B+-tree of a secondary index to the respective data files is provided via the accession list file. Finally, another file is used to represent all the B+-trees associated with a data file. Thus, three files suffice for the implementation of a KSAM data file and its associated indexes. The implementation schema organizes each of the three files as a direct-access file. Thus the high popularity of direct-access files makes the implementation possible in almost any programming language.