Integrated Hardware/Software Codesign for Heterogeneous Computing Systems

Abstract

This paper describes a strategy that integrates the task mapping and task scheduling steps for heuristic search techniques, with multiple neighbourhood functions to reduce search time and enchance solution quality in developing heterogeneous computing systems. For case studies involving 40 randomly generated task graphs and the fast Fourier transform, experimental results show that our approach outperfroms previous approaches in terms of search time by up to 93 times, and solution quality by up to 22.6% for a system with a microprocessor, a floating-point digital signal processor, and an FPGA.