1. [Publications](/index.php/publications) 2. DREAMPlace: Deep Learning Toolkit-Enabled GPU Acceleration for Modern VLSI Placement # DREAMPlace: Deep Learning Toolkit-Enabled GPU Acceleration for Modern VLSI Placement ![](/sites/default/files/styles/wide/public/publications/DREAMPlace_Capture.PNG?itok=y6dvTk_B) Placement for very-large-scale integrated (VLSI) circuits is one of the most important steps for design closure. We propose a novel GPU-accelerated placement framework DREAMPlace, by casting the analytical placement problem equivalently to training a neural network. Implemented on top of a widely-adopted deep learning toolkit PyTorch, with customized key kernels for wirelength and density computations, DREAMPlace can achieve around 40× speedup in global placement without quality degradation compared to the state-of-the-art multi-threaded placer RePlAce. We believe this work shall open up new directions for revisiting classical EDA problems with advancements in AI hardware and software. ## Authors Yibo Lin (Peking University) Shounak Dhar (UT-Austin) Wuxi Li (UT-Austin) Mark Haoxing Ren (NVIDIA) [Brucek Khailany](/index.php/person/brucek-khailany) David Z. Pan (UT Austin) ## Publication Date Monday, June 3, 2019 ## Published in [Design Automation Conference (DAC) 2019](http://yibolin.com/publications/papers/PLACE_DAC2019_Lin.pdf) ## Research Area [Algorithms and Numerical Methods](/index.php/research-area/algorithms) [Artificial Intelligence and Machine Learning ](/index.php/research-area/machine-learning-artificial-intelligence) [Circuits and VLSI Design](/index.php/research-area/circuits) ## Uploaded Files [54\_1\_Lin\_DREAMPLACE.pdf](https://research.nvidia.com/sites/default/files/pubs/2019-06_DREAMPlace%3A-Deep-Learning//54_1_Lin_DREAMPLACE.pdf "Open file in new window")3.72 MB ## Award DAC 2019 Best Paper Award ## Copyright This material is posted here with permission of the IEEE. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to .