The continued scaling of semiconductor technologies leads to diverse challenges such as power and temperature, which also forces reliability as another design metric of prime concern. There exists strong need to link reliability with physical metrics in a high-level architecture design environment, where estimation of reliability impacts can be performed in the early design stage. In this paper, we propose a joint modeling and simulation framework for power, thermal and timing variation, which is integrated into a commercial high-level processor design environment. A custom timing variation model is provided for estimation of dynamic timing variation, which is demonstrated using one nanoscale thermal effect known as Inverted Temperature Dependence. The complete modeling flow is automated for customized processor model with arbitrary architectural hierarchy, which assists designer to perform architectural and application-level design space exploration with power, thermal and reliability impacts.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.