Presentation at NVIDIA GTC 2023 (GPU Technology Conference)
Zijian Xu, Preferred Networks Engineer, will be presenting “Running 100,000-Atom Molecular Dynamics with Accurate NN Potential on a GPU using Automatic Recomputation” at NVIDIA GTC 2023 (GPU Technology Conference) held from March 21 to March 24, 2023.
Please see the details below. You may register for free to view the session or recorded video.
|Wednesday, Mar 22 12:00 PM – 12:25 PM JST
|Zijian Xu, Engineer, Preferred Networks
|Running 100,000-Atom Molecular Dynamics with Accurate NN Potential on a GPU using Automatic Recomputation
|Accurate and large scale atomistic simulation allows a new way of material discovery. Although neural network potential (NNP), a deep learning-based force field, dramatically accelerates computation time compared to conventional DFT methods, the amount of GPU memory often becomes a bottleneck to running large-scale simulations. Supporting a larger number of atoms in such simulations greatly increases the range of structures and phenomena that can be reproduced, including large molecules like proteins, heterogeneous systems like composites, additives, and lattice-mismatched interfaces. To overcome this issue, our method analyzes and reorganizes the computation graph of the neural network model to achieve exactly the same outputs with a smaller memory budget. The basic idea here is known as “recomputation,” which wipes out less frequently referenced large data from GPU memory and recomputes them from smaller values when necessary. Our experiments have shown that it is possible to run up to 5x larger simulations with only a 25% execution time overhead using the same amount of GPU memory.