A primary goal of high-level modeling is to efficiently explore a broad design space, converging on an optimal or near-optimal system architecture before moving to a more detailed design. This paper evaluates a high-level, layered software-on-hardware performance modeling environment called MESH that captures coarse-grained, interacting system elements. The validity of the high-level model is established by comparing the outcome of the high-level model with a corresponding low-level, cycle-accurate instruction set simulator. We model a network processor and show that both high and low level models converge on the same architecture when design modifications are classified as good or bad performance impacts.
Andrew S. Cassidy, JoAnn M. Paul, Donald E. Thomas