ML-based Fast On-Chip Transient Thermal Simulation for Heterogeneous 2.5D/3D IC Designs
Akhilesh Kumar, Norman Chang, David Geb, Haiyang He, Stephen H. Pan, Jimin Wen, Saeed Asgari, Mehdi Abarham, Chris A. Ortiz
Abstract
Large 3DIC designs with multiple chips require several iterations of transient thermal analysis particularly for fine-grain on-chip dynamic thermal management. This requires a fast thermal analysis technology as opposed to traditional CFD/FEA based methods which have severe runtime/capacity limitations for large chips (e.g., 2cmx2cm) in 3DIC while generating fine grained (e.g., 10umx10um) transient thermal response. The fast transient thermal analysis is based on the idea of combining the global, intermediate, and local transient response curves generated from an ML-predictor. The local, intermediate, and global transient response curves are scaled based on the far-field and near-field transient decay surface components respectively, computed using the trained ML decay surface predictor, followed by linear superposition of the curves for each power value in the transient power profile to generate the effective transient response curve. The runtime for generating thermal results for a large chip is in the order of minutes, compared to several hours/days while using CFD/FEA based tools with good accuracy correlation. The fast transient thermal solver is implemented on distributed ML computing platform for parallel computation of transient thermal inference model.