ATSim: A Fast and Accurate Simulation Framework for 2.5D/3D Chiplet Thermal Design Optimization
By Qipan Wang 1,2, Tianxiang Zhu 1, Jiajia Cui 1, Yicheng Wei 1, Linxiao Shen 1, Zhe Cheng 1, Runsheng Wang 1,3,4, Ru Huang 1,3,4, Yibo Lin 1,3,4
1 School of Integrated Circuits, Peking University
2 Academy for Advanced Interdisciplinary Studies, Peking University
3 Institute of Electronic Design Automation, Peking University, Wuxi, China
4 Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China
Abstract
This paper reviews the thermal challenges in 2.5D/3D chiplet integration systems and introduces ATSim, a simulation framework with applications to chiplet thermal optimization. ATSim enables fast and accurate thermal simulation for both steady-state and transient conditions. It supports nonlinear, heterogeneous, and anisotropic materials. The framework features a multilevel grid generation scheme based on a novel hybrid tree structure. Compared to mainstream academic and commercial tools, ATSim achieves high accuracy and efficiency, making it a powerful tool for evaluating and improving thermal designs, including applications like thermal-aware placement.
To read the full article, click here
Related Chiplet
- Integrated voltage regulator (IVR) chiplet
- High-performance connectivity chiplets
- eFPGA Chiplet
- DPIQ Tx PICs
- IMDD Tx PICs
Related Technical Papers
- DeepOHeat-v1: Efficient Operator Learning for Fast and Trustworthy Thermal Simulation and Optimization in 3D-IC Design
- 3D-ICE 4.0: Accurate and efficient thermal modeling for 2.5D/3D heterogeneous chiplet systems
- ThermoDSE: A Thermal-Aware and Comprehensive Design Space Exploration for Chiplet-Based DNN Accelerators
- CLIPGen: A Chiplet Link IP Modeling and Generation Framework for 2.5D Architecture Exploration
Latest Technical Papers
- Self-Activated Direct Bonding with a Highly Polar Atomic-Layer-Deposited Film for 3D Integration
- FLINT: Efficiently Leveraging High Bandwidth Flash for Capacity-Scalable LLM Inference Acceleration
- Beacon: LLM Multi-Agent Driven Hardware Design Space Exploration for Heterogeneous Multi-Chiplet Deep Learning Accelerators
- CHIPSMORE: Compute-in-Interconnect and -Memory Chiplets for Multi-Mode Multi-Request LLM Inference Acceleration
- ECO-CHIP: Estimation of Carbon Footprint of Chiplet-based Architectures for Sustainable VLSI