Presentation

DescriptionFluid flow behaviors in nanoporous materials are distinct from those following the continuum physics. Numerical simulations can be a complement to laboratory experiments. This work presents a dissipative particle dynamics (DPD) package for GPU-accelerated mesoscale flow simulations in nanoporous materials. In an ideal benchmark that minimizes load imbalance, the package delivered nearly perfect strong- and weak-scaling (with up to 4 billion DPD particles) on up to 1,536 V100 GPUs on Oak Ridge National Laboratory's Summit supercomputer. More remarkably, in a benchmark to measure its usefulness with realistic nanopores in SBA-15 silica, the package exhibited more than 20x speedup over its LAMMPS-based CPU counterpart with the same number nodes (e.g., 384 V100 GPUs vs. 2,688 POWER9 cores). It is worth highlighting that the NVLink2 Host-to-Device interconnects kept the cost of CPU-GPU memory copy as low as only 10% of GPU activity time per rank: 4 times less than their PCIe counterparts.