Document Type
Thesis
Date of Award
5-31-1988
Degree Name
Master of Science in Electrical Engineering - (M.S.)
Department
Electrical Engineering
First Advisor
John D. Carpinelli
Second Advisor
Anthony D. Robbi
Abstract
Data flow is a new approach to parallel computing. The concept is to enable the execution of an instruction whenever its operands become available. The only constraints in a data flow machine are the data flow dependencies between instructions.
In this thesis, a simulation package called the Data Flow Dependency Generator is developed. This package is designed to express the idea of data flow computations, the data flow dependencies and the use of multiprocessors.
Given a number of processors, a number of time units for each operation and a program, the Data Flow Dependency Generator finds the data dependencies between every two formulas in the program, schedules the processor(s) and calculates the results.
Recommended Citation
Chan, Chin-Piao, "Data flow dependency generator" (1988). Theses. 3013.
https://digitalcommons.njit.edu/theses/3013
