Abstract
We argue that a parallel processing course should be offered to undergraduate computer science majors. A major component of such a course should be a series of programming laboratories where the student can investigate the strengths and weaknesses of different parallel architectures. The student should design and debug parallel algorithms on the different parallel models.We propose a cost effective solution to the teaching of the course which uses simulators and Transputer-based parallel accelerators in a personal computer or workstation environment.

This publication has 13 references indexed in Scilit: