A PPA-Driven 3D-IC Partitioning Selection Framework with Surrogate Models

By Shang Wang ¹, Shuai Liu ¹, Owen Randall ¹, Matthew E. Taylor ^1,2
1 University of Alberta, Canada
² Alberta Machine Intelligence Institute (Amii), Canada

Abstract

3D-IC netlist partitioning is commonly optimized using proxy objectives, while final PPA is treated as a costly evaluation rather than an optimization signal. This proxy-driven paradigm makes it difficult to reliably translate additional PPA evaluations into better PPA outcomes. To bridge this gap, we present DOPP (D-Optimal PPA-driven partitioning selection), an approach that bridges the gap between proxies and true PPA metrics. Across eight 3D-IC designs, our framework improves PPA over Open3DBench (average relative improvements of 9.99% congestion, 7.87% routed wirelength, 7.75% WNS, 21.85% TNS, and 1.18% power). Compared with exhaustive evaluation over the full candidate set, DOPP achieves comparable best-found PPA while evaluating only a small fraction of candidates, sub stantially reducing evaluation cost. By parallelizing evaluations, our method delivers these gains while maintaining wall-clock runtime comparable to traditional baselines.

Index Terms — 3D IC Partitioning, Linear regression, D-optimal design, Power Performance Area (PPA)

To read the full article, click here