Distributed Computer Architecture for Fast Qualitative Simulation

Qualitative simulation is a new and challenging simulation paradigm. A drawback of current implementations is poor execution speed. In this research project a special-purpose computer architecture for the widely-used qualitative simulator QSim is developed to increase the performance. Two approaches are considered to improve the performance. Complex functions are parallelized and mapped onto a multiprocessor system. Less complex functions are implemented in hardware. These functions are executed on specialized coprocessors. The prototype implementation of this computer architecture is based on digital signal processors TMS320C40 and field programmable gate arrays (Xilinx). The experimental evaluation proves that a significant speedup can be achieved with this specialized computer architecture.