86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy

We present the GPU version of DeePMD-kit, which, upon training a deep neural network model using ab initio data, can drive extremely large-scale molecular dynamics (MD) simulation with ab initio accuracy. Our tests show that for a water system 12, 582, 912 of atoms, the GPU version can be 7 times faster than the CPU version under the same power consumption. The code can scale up to the entire Summit supercomputer. For a copper system of 113, 246, 208 atoms, the code can perform one nanosecond MD simulation per day, reaching a peak performance of 86 PFLOPS (43% of the peak). Such unprecedented ability to perform MD simulation with ab initio accuracy opens up the possibility of studying many important issues in materials and molecules, such as heterogeneous catalysis, electrochemical cells, irradiation damage, crack propagation, and biochemical reactions.本研究推出DeePMD-kit的图形处理器（GPU）版本，该版本可基于从头算（ab initio）数据训练深度神经网络模型，并可开展具备从头算精度的超大规模分子动力学（Molecular Dynamics, MD）模拟。我们的测试表明，针对包含12582912个原子的水体系，在相同功耗下，该GPU版本的运行速度可比中央处理器（CPU）版本快7倍。该代码可扩展至完整的顶峰（Summit）超级计算机集群，针对包含113246208个原子的铜体系，该代码每日可完成1纳秒的MD模拟，峰值性能可达86拍浮点运算每秒（PFLOPS），为其理论峰值性能的43%。这种具备从头算精度的超大规模MD模拟的前所未有的能力，为研究材料与分子领域的诸多重要课题提供了可能，例如多相催化、电化学电池、辐照损伤、裂纹扩展以及生化反应等。