Mixed-Type Multiple-Valued Filing Scheme of Order One and Two
This paper will present a filing scheme for multiple-valued records. It is assumed that records are characterized by m attributes having n values each, and the query set consists of two kinds of queries, where one (called a first-order query) specifies a value of one attribute and another one (called a second-order query) specifies values of two attributes. The number of buckets to be organized and the redundancy of the file are less than those of an inverted file, provided the distribution of records is uniform. The number of pointer fields to be attached to the accession number stored in a bucket is considerably reduced in this scheme. Every first-order query can be answered by the access to one bucket. Every second-order query can be answered by the access to one bucket and the traversal of its contents without any set operations.
KeywordsPointer Field Access Path Attribute Number Finite Geometry File Organization
Unable to display preview. Download preview PDF.
- 4.R. C. Bose, C. T. Abraham, and S. P. Ghosh, File organization of records with multiple-valued attributes for multi-attributes queries, in: “Combinatorial Mathematics and Its Applications”, University of North Carolina Press, Chapel Hill, N.C., pp.277–297 (1969).Google Scholar
- 14.S. Yamamoto, S. Tazawa, K. Ushio, and H. Ikeda, Design of a generalized balanced multiple-valued file organization scheme of order two, Proc. of 8-th ACM-SIGMOD Int. Conf. on Management of Data (1978).Google Scholar