Browse Prior Art Database

# Determining Independence Dependence of State Functions in a Multipath Algorithm

IP.com Disclosure Number: IPCOM000081378D
Original Publication Date: 1974-May-01
Included in the Prior Art Database: 2005-Feb-27
Document File: 2 page(s) / 14K

IBM

## Related People

Dapron, FE: AUTHOR

## Abstract

In partitioning an algorithm, the best candidates for each partition are those state functions which are dependent on each other and independent of all state functions in other partitions.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 2

Determining Independence Dependence of State Functions in a Multipath Algorithm

In partitioning an algorithm, the best candidates for each partition are those state functions which are dependent on each other and independent of all state functions in other partitions.

This method was developed primarily for analyzing the sequence relationships in a set of decision table action rules and so will be presented in that notation, and with the resolving matrix being a precedence matrix. Other associative properties, however, can be cast in the same notation.

(Image Omitted)

The above decision rules, presumed ordered top-bottom, of a decision table have a high degree of action independence. Identified are the terminal actions in each path and these actions, identified by the symbol "X" have been moved to the bottom of the table. Their original relative sequence has been retained although, in a valid table, it would not be necessary.

The string A1 - X14 represents all state functions associated with the algorithm. The substring X8 - X14 represents those which are additionally terminal. Each column, 1 - 11, represents a path through the algorithm and the X's in each column indicate which state functions will be executed in that specific path.

Beside the action rule display is a skeleton logical matrix, in this case a precedence matrix, over the same sequence represented in the rule display with the primary precedence relationships displayed on it. It differs from a normal precedence matrix in that the symbol for a precedence relationship is "X" rather than "1", and in that an "X" on the principal diagonal is not an indication that the matrix is singular. As constructed here, it is impossible for any X's to appear to the right of the principal diagonal.

The matrix initially consists only of the indices and the diagonal. The diagonal is entered only as a visual aid in aligning entries. All spaces in the matrix are considered to contain binary 0's at the beginnin...