PDNNet: PDN-Aware GNN-CNN Heterogeneous Network for Dynamic IR Drop Prediction
This addresses the computationally expensive simulation of IR drop in large-scale IC design, offering a machine learning solution with significant speed improvements.
The paper tackles dynamic IR drop prediction in integrated circuit design by proposing PDNNet, a dual-branch GNN-CNN heterogeneous network that incorporates PDN configuration and cell-PDN relations, achieving 545x speedup compared to commercial tools while outperforming state-of-the-art CNN-based methods.
IR drop on the power delivery network (PDN) is closely related to PDN's configuration and cell current consumption. As the integrated circuit (IC) design is growing larger, dynamic IR drop simulation becomes computationally unaffordable and machine learning based IR drop prediction has been explored as a promising solution. Although CNN-based methods have been adapted to IR drop prediction task in several works, the shortcomings of overlooking PDN configuration is non-negligible. In this paper, we consider not only how to properly represent cell-PDN relation, but also how to model IR drop following its physical nature in the feature aggregation procedure. Thus, we propose a novel graph structure, PDNGraph, to unify the representations of the PDN structure and the fine-grained cell-PDN relation. We further propose a dual-branch heterogeneous network, PDNNet, incorporating two parallel GNN-CNN branches to favorably capture the above features during the learning process. Several key designs are presented to make the dynamic IR drop prediction highly effective and interpretable. We are the first work to apply graph structure to deep-learning based dynamic IR drop prediction method. Experiments show that PDNNet outperforms the state-of-the-art CNN-based methods and achieves 545x speedup compared to the commercial tool, which demonstrates the superiority of our method.