On the Calibration of Abstract Performance Models for System-level Design Space Exploration
Andy D. Pimentel, Mark G. Thompson, Simon Polstra, Cagkan Erbas · 2006
High-level performance modeling and simulation has become a key ingredient of system-level design as they facilitate early architectural design space exploration. An important precondition for such high-level modeling and simulation methods is that they should yield trustworthy performance estimations. This requires validation (if possible) and calibration of the simulation models, which are two aspects that have not yet been widely addressed in the system-level community. This paper presents an initial attempt to provide support for calibrating various model components of a system-level performance model. We discuss these model calibration mechanisms in the context of our Sesame system-level simulation framework. An illustrative case study is also presented to indicate the merits of model calibration