Technical Report Number
The high costs associated with logic simulation of large VLSI based systems have led to the need for new computer architectures tailored to the simulation task. Such architecture have the potential for significant speed-ups over standard software based logic simulators. Several commercial simulation engines have bene produced to satisfy needs in this area. To properly explore the space of alternative simulation architectures, data is required on the simulation process itself. This paper presents a framework for such data gathering activity and uses the data in estimating the maximum speed-up attainable with a particular type of special-purpose parallel/pipelined simulation machine. First, possible sources of speed-up in the logic simulation task are examined. Then, the sort of data needed in the design of simulation engines is discussed. Next, the data is presented and the implication on machine design are discussed. This data includes information on subtask times found in standard discrete-event simulation algorithms, event intensities, queue length distributions, and simultaneous event distributions. Finally, a simple performance model of one type of simulation machine is developed, and the maximum speed-up attainable with this type of machine is predicted.
Wong, K. F. and Franklin, Mark A., "Logic Simulation: Statistics and Machine Design" Report Number: WUCS-86-17 (1986). All Computer Science and Engineering Research.