Graph neural network based cell library characterization method for fast design technology co-optimization

Design technology co-optimization (DTCO) plays a critical role in achieving optimal power, performance, and area (PPA) for advanced semiconductor process development. Cell library characterization is essential in DTCO flow, but traditional methods are time-consuming and costly. To overcome these challenges, we propose a graph neural network (GNN)-based machine learning model for rapid and accurate cell library characterization. Our model incorporates cell structures and demonstrates high prediction accuracy across various process–voltage–temperature (PVT) corners, technology parameters and aging effects. Validation with 512 unseen corners and over one million test data points shows accurate predictions of delay, power, and other cell metrics for 37 types of cells and a speed-up of 100X compared with SPICE simulations. Additionally, we investigate system-level metrics such as worst negative slack (WNS), leakage power, and dynamic power using predictions obtained from the GNN-based model on unseen corners. Our model achieves precise predictions, with absolute error ≤3.0 ps for WNS, percentage errors ≤0.60% for leakage power, and ≤0.99% for dynamic power, when compared to golden reference. With the developed model, we further proposed a fine-grained drive strength interpolation methodology to enhance PPA for small-to-medium-scale designs, resulting in an approximate 1%–3% improvement. •A highly generalized full-metric fast standard cell characterization method based on graph neural networks (GNN).•High prediction accuracy across diverse process–voltage–temperature (PVT) corners and technology parameters with 100X speed-up compared to traditional SPICE simulations.•A fine-grained drive strength interpolation method based on this model, achieving 1%–3% improvement in power, performance, and area (PPA) for small-to-medium-scale designs.