arXiv：cs.LG· Yixing Li, Jiahang Zhou, Zhiyu Zeng, Xin Ai·· 3 小时前AI 评分40
DAIST：面向 3D-IC 热建模的可组合 AI 加速迭代求解器
A Composable AI-Accelerated Iterative Solver for 3D-IC Thermal Modeling
AI 导读
研究提出 DAIST，一种将全局封装热仿真分解为块级子域问题、用神经算子替换子域求解器并通过界面温度与热流迭代耦合的可组合热求解器。在多芯粒系统和先进封装系统上，DAIST 相比传统 FEM 求解器最高实现 178 倍加速，平均温度误差分别为 0.068% 和 0.323%。块级神经算子可在结构不同的封装组合中跨拓扑复用，无需重新训练，迭代预算还可调节精度与运行时间的权衡。
正文
Abstract:Accurate thermal analysis of heterogeneous 2.5D/3D-IC packages is essential yet computationally prohibitive. A single full-package FEM simulation can take hours, while AI-based surrogates treat the entire stack as a monolithic prediction target and must be retrained whenever the die count or topology changes. To address this limitation, this work proposes Domain-Decomposed AI-Accelerated Iterative Solver for Thermal Analysis (DAIST), a composable thermal solver that decomposes the global package simulation into block-level subdomain problems, replaces subdomain solvers with neural operators, and couples them through iterative exchanges of interfacial temperature and heat flux. This local-to-global architecture eliminates the topology lock-in of monolithic models: block-level neural operators can be directly reused in unseen package assemblies without retraining. The iterative coupling strategy further provides a controllable accuracy-runtime tradeoff, where the iteration budget can be adjusted to trade accuracy for runtime. Evaluated on a multi-chiplet system and an advanced packaging system, DAIST achieves up to $178\times$ speedup over traditional FEM solvers with mean temperature errors of 0.068% and 0.323%, respectively, while demonstrating cross-topology reuse of block-level models across structurally distinct package assemblies.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.02461 [cs.LG] |
| (or arXiv:2610.02461v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02461 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yixing Li [view email]
[v1]
Thu, 1 Oct 2026 20:34:17 UTC (4,911 KB)
来源：arXiv：cs.LG · arxiv.org