Fast and Accurate Library Generation Leveraging Deep Learning for OCV Modelling

Eunice Naswali1, Namhoon Kim1, Pravin Chandran2
1Intel Corporation, 2Intel Technologies India Pvt Ltd


Abstract

Statistical timing characterization for modeling On-Chip Variation (OCV) of library cells is critical in current technology nodes to avoid over-design and to improve design convergence and predictability. OCV characterization is, however, resource intensive as it involves running millions to billions of Monte-Carlo spice simulations to build the entire standard cell timing library. In this paper, we leverage a deep learning technique to model cell propagation delays as well as OCV sigma at target process-voltage-temperature (PVT) corners with a significantly reduced number of simulations of 60% while also achieving a high accuracy of 99.98%.