This paper presents an analytical model for the early performance prediction of VLSI microprocessor cores. The model involves a wide set of architecture parameters (including data on memory hierarchy data, branch prediction and pipeline organization) and instruction level parameters as well as statistical information obtained from an ad-hoc version of the SimpleScalar toolset. The model validation against cycle accurate simulation shows a very good fitting.
Published in:
Industrial Technology, 2004. IEEE ICIT '04. 2004 IEEE International Conference on
(Volume:3
)
Date of Conference: 8-10 Dec. 2004