Graph neural networks accelerate TSV network optimization for 3D integration

Researchers have developed a new framework that uses graph neural networks (GNNs) to optimize through-substrate via (TSV) networks for 2.5D/3D heterogeneous integration. This approach combines physics-informed analytical modeling with GNN surrogates to significantly speed up the computationally intensive process of simulating TSV arrays. The framework can explore millions of TSV configurations in minutes, reducing evaluation time by over six orders of magnitude compared to traditional methods, and has been validated with industry-standard simulation tools. AI

IMPACT Accelerates design cycles for advanced semiconductor packaging by enabling rapid electro-thermal co-design.

RANK_REASON The item is an academic paper detailing a new computational framework for optimizing TSV networks using graph neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Ansys Mechanical
• graph neural network
• HFSS
• Mohamed Gharib
• TSV Networks
• TSV-PhGNN

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →