White Papers

Dell - Internal Use - Confidential

LAMMPS

Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) is a classical molecular dynamics code and has potentials for

solid-state materials (metals, semiconductors) and soft matter (biomolecules, polymers) and coarse-grained or mesoscopic systems. It

can be used to model atoms or, more generically, as a parallel particle simulator at the atomic, meso, or continuum scale. It runs on

single processors or in parallel using message-passing techniques and a spatial-decomposition of the simulation domain. Many of its

models have versions that provide accelerated performance on CPUs, GPUs, and Intel Xeon Phis. The code is designed to be easy to

modify or extend with new functionality.

We performed a 3D Lennard-Jones melt simulation package which comes with LAMMPS. It simulates the movement of molecules

confined in a squared box. LAMMPS has built-in functions for placing molecules inside the box and moving them according to Newton’s

laws. This is arguably one of the simplest models capable of reproducing the complete thermodynamic behavior of classical fluids. The

details of the simulation set-up can be found in Appendix B.

Figure 16 shows LAMMPS performance on both configurations G and K. The testing dataset is Lennard-Jones liquid dataset, which

contains 512,000 atoms, and LAMMPS compiled with the kokkos package. V100 is 71% and 81% faster on Config G and Config K

respectively. Comparing V100-SXM2 (Config K) and V100-PCIe (Config G), the former is 5% faster due to NVLINK and higher CUDA

core frequency.

The C4130 server with NVIDIA

Tesla

V100™ GPUs demonstrates exceptional performance for HPC applications that require faster

computational speed and highest data throughput. Applications like AMBER and LAMMPS were boosted with C4130 configuration K,

owing to P2P access, higher bandwidth of NVLink and higher CUDA core clock speed. Overall, a PowerEdge C4130 with Tesla V100

GPUs performs 1.54x to 1.8x faster than a C4130 with P100 for AMBER and LAMMPS.

Figure 16 LAMMPS 3D Lennard-Jones melt simulation