Simulation of a novel multiprocessor system based on dataflow principles

Abstract

This thesis research is a study of a novel processor architecture for a massively parallel computer system. An existing simulation of the Fresh Breeze architecture has been extended to incorporate a multi-level memory hierarchy and dynamic load balancing. An efficient hardware-based garbage collection mechanism has been proposed. Various design trade-offs are evaluated. The simulation demonstrates that the architecture can support memory access with DRAM latency and still achieve high processor utilization.