Problem
Research questions and friction points this paper is trying to address.
3D-IC thermal modeling
domain decomposition
neural operator
iterative solver
composable AI
A Composable AI-Accelerated Iterative Solver for 3D-IC Thermal Modeling
🤖 AI Summary
This study addresses the high computational cost of 3D-IC thermal analysis and the need for AI model retraining upon topological changes by proposing the DAIST solver. Based on domain decomposition, this method partitions global simulations into subdomain problems, employing neural operators to replace conventional subdomain solvers while iteratively coupling interface data. A local-to-global architecture eliminates topological locking, enabling block-level model reuse across different packages and facilitating a controllable accuracy–efficiency trade-off. Experimental results demonstrate up to a 178× speedup over traditional finite element methods with an average temperature error of only 0.068%, effectively validating the proposed approach’s cross-topology generalization capability.
📝 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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Domain Decomposition
Neural Operator
Composable AI
Thermal Modeling
Iterative Solver
🔎 Similar Papers
No similar papers found.
