Browse Prior Art Database

Method for Compact Storage and Retrieval of Relational Data Bases Using Sparse Matrix and Run Length Encoding

IP.com Disclosure Number: IPCOM000043063D
Original Publication Date: 1984-Jul-01
Included in the Prior Art Database: 2005-Feb-04
Document File: 2 page(s) / 69K

Publishing Venue

IBM

Related People

Ashany, R: AUTHOR [+3]

Abstract

A completely inverted file capable of general relational access and storage of generalized alphabetic and numerical data can be achieved with greatly reduced storage requirements by utilizing a Sparse Matrix Associative Relational Approach of Dynamic Data Structuring and Data Retrieval (SPARCOM) coupled with a specialized form of Run Length Encoding (RLE). The SPARCOM/RLE database technique supports any record processing front end whether it be simple sequential or a fully relational query language. SPARCOM relates entities to their attributes using a matrix organization coupled with a specialized use of run length encoding to compress out all null positions in the matrix. As shown in the figure, each field of the records represents columns of the matrix while the records make up the rows of the matrix.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 81% of the total text.

Page 1 of 2

Method for Compact Storage and Retrieval of Relational Data Bases Using Sparse Matrix and Run Length Encoding

A completely inverted file capable of general relational access and storage of generalized alphabetic and numerical data can be achieved with greatly reduced storage requirements by utilizing a Sparse Matrix Associative Relational Approach of Dynamic Data Structuring and Data Retrieval (SPARCOM) coupled with a specialized form of Run Length Encoding (RLE). The SPARCOM/RLE database technique supports any record processing front end whether it be simple sequential or a fully relational query language. SPARCOM relates entities to their attributes using a matrix organization coupled with a specialized use of run length encoding to compress out all null positions in the matrix. As shown in the figure, each field of the records represents columns of the matrix while the records make up the rows of the matrix. Alphabetic fields are represented using consecutive runs of A through Z, where the first run contains one bit at logical "1" for the first character in a word, the second run of A through Z contains one bit at logical "1" for the second character of the field, etc. Similarly, numeric fields are stored using consecutive runs of zero through nine. The relative position of logical "1" bits in the matrix is run length encoded by storing displacement values between the bits. This reduces the memory required by eliminating any storage requirement for the logical...