MOLECULAR DYNAMICS PERFORMANCE GUIDE - Digital Research Alliance of CANADA

BENCHMARK DETAILS

ID=308

  • Dataset: 9naw
  • Software: GROMACS.ROCm (container/2026.0-SYCL (Adaptive CPP)-rocm-7.2.0-avx512)
  • Resource: 2 tasks, 24 cores, 1 nodes, 2 GPUs, no NVLink
  • CPU: AMD Instinct Accelerator MI300A (Aqua Vanjaram), 2.0 GHz
  • GPU: AMD Instinct-MI300A-128GB, 24 cores/GPU
  • Simulation speed: 12.613 ns/day
  • Efficiency: 35.0 %
  • Site: Nibi
  • Date: March 12, 2026, 4:04 p.m.
  • Submission script:
    #!/bin/bash
    #SBATCH --gpus=mi300a:2
    #SBATCH --mem-per-cpu=4000
    #SBATCH --time=1:0:0
    #SBATCH --ntasks=4
    #SBATCH --cpus-per-task=24

    module load apptainer/1.3.5

    apptainer run --rocm gromacs_amd.sif gmx mdrun \
    -ntmpi 16 \
    -ntomp 6 \
    -npme 1 \
    -nb gpu \
    -pme gpu \
    -update gpu \
    -bonded gpu \
    -nstlist 300 \
    -noconfout \
    -s topol.tpr \
  • Notes:
  • Simulation input file:
    title                   = benchmark 
    ; Run parameters
    integrator = md
    nsteps = 100000
    dt = 0.002
    ; Output control
    nstxout = 0
    nstvout = 0
    nstfout = 0
    nstenergy = 1000
    nstlog = 500
    nstxout-compressed = 5000
    compressed-x-grps = System
    ; Bond parameters
    continuation = yes
    constraint_algorithm = Lincs
    constraints = h-bonds

    ; Neighborsearching
    cutoff-scheme = Verlet
    ns_type = grid
    nstlist = 10
    rcoulomb = 0.8
    rvdw = 0.8
    DispCorr = Ener ; anaytic VDW correction
    ; Electrostatics
    coulombtype = PME
    pme_order = 4
    fourier-nx = 324
    fourier-ny = 324
    fourier-nz = 324
    ; Temperature coupling is on
    tcoupl = V-rescale
    tc-grps = system
    tau_t = 0.1
    ref_t = 300
    ; Pressure coupling is on
    pcoupl = Parrinello-Rahman
    pcoupltype = isotropic
    tau_p = 2.0
    ref_p = 1.0
    compressibility = 4.5e-5
    ; Periodic boundary conditions
    pbc = xyz
    ; Velocity generation
    gen_vel = no