arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.19330cs.LG
收藏

CTS-Bench: 面向时钟树综合中GNN的图粗化权衡基准测试

CTS-Bench: Benchmarking Graph Coarsening Trade-offs for GNNs in Clock Tree Synthesis

The University of Southern Mississippi(密苏里州南方大学);Intel Corporation(英特尔公司);Louisiana Tech University(路易斯安那理工大学)

Barsat Khadka, Kawsher Roxy, Md Rubel Ahmed

AI总结:

提出CTS-Bench基准套件,系统评估图粗化对GNN在时钟树综合中预测精度与计算效率的权衡,发现粗化虽降低内存和加速训练,但会移除关键结构信息导致零样本评估下R²为负。

摘要:

图神经网络(GNN)在电子设计自动化中的物理设计分析中越来越受到关注,特别是用于建模时钟树综合行为,如时钟偏斜和缓冲复杂性。然而,由于在原始门级网表上操作的内存和运行时间成本过高,实际部署仍然有限。图粗化通常用于提高可扩展性,但其对CTS关键学习目标的影响尚未得到充分表征。本文介绍了CTS-Bench,一个基准测试套件,用于系统评估基于GNN的CTS分析中图粗化、预测精度和计算效率之间的权衡。CTS-Bench包含跨越五个架构的4,860个收敛的物理设计解决方案,并提供来自布局后设计的配对原始门级和聚类图表示。以时钟偏斜预测作为代表性CTS任务,我们展示了明确的精度-效率权衡。虽然图粗化将GPU内存使用减少高达17.2倍,并将训练加速高达3倍,但它也移除了对建模时钟分布至关重要的结构信息,经常导致零样本评估下R²为负。我们的发现表明,即使全局物理指标保持不变,通用图聚类技术也可能从根本上损害CTS学习目标。CTS-Bench支持对CTS感知的图粗化策略进行原则性评估,支持在现实物理设计约束下对GNN架构和加速器进行基准测试,并为开发学习辅助的CTS分析和优化技术提供了基础。

英文摘要:

Graph Neural Networks (GNNs) are increasingly explored for physical design analysis in Electronic Design Automation, particularly for modeling Clock Tree Synthesis behavior such as clock skew and buffering complexity. However, practical deployment remains limited due to the prohibitive memory and runtime cost of operating on raw gate-level netlists. Graph coarsening is commonly used to improve scalability, yet its impact on CTS-critical learning objectives is not well characterized. This paper introduces CTS-Bench, a benchmark suite for systematically evaluating the trade-offs between graph coarsening, prediction accuracy, and computational efficiency in GNN-based CTS analysis. CTS-Bench consists of 4,860 converged physical design solutions spanning five architectures and provides paired raw gate-level and clustered graph representations derived from post-placement designs. Using clock skew prediction as a representative CTS task, we demonstrate a clear accuracy-efficiency trade-off. While graph coarsening reduces GPU memory usage by up to 17.2x and accelerates training by up to 3x, it also removes structural information essential for modeling clock distribution, frequently resulting in negative $R^2$ scores under zero-shot evaluation. Our findings indicate that generic graph clustering techniques can fundamentally compromise CTS learning objectives, even when global physical metrics remain unchanged. CTS-Bench enables principled evaluation of CTS-aware graph coarsening strategies, supports benchmarking of GNN architectures and accelerators under realistic physical design constraints, and provides a foundation for developing learning-assisted CTS analysis and optimization techniques.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑