dps mdrun

Run or resume a molecular-dynamics simulation from a DROPPS run-input file.

Synopsis and options

usage: dps mdrun [-h] -s FILE [-cpi [FILE]] -o PREFIX [-x FILE] [-e FILE] [-g FILE] [-cpo FILE]
                 [-c FILE] [--stress-output FILE] [--platform {auto,CUDA,OpenCL,CPU,Reference}]
                 [-gpu_id DEVICE_INDEX] [--precision {single,mixed,double}] [-nt THREADS]
                 [-nsteps NSTEPS] [--seed SEED] [-cpt CHECKPOINT_MINUTES] [-cpnum]
                 [-maxh MAX_HOURS] [-append | -noappend]

Run or resume a molecular-dynamics simulation from a DROPPS run-input file.

Input:
  -s FILE, --run-input FILE
                        Input DROPPS run file (.tpr) containing the system and simulation
                        settings. (required)
  -cpi [FILE], --checkpoint [FILE]
                        Resume from a native checkpoint (.chk) or portable State (.state.xml). If
                        FILE is omitted, use the resolved checkpoint path and fall back to its
                        paired State XML. (default: None)

Output:
  -o PREFIX, --output-prefix PREFIX, -deffnm PREFIX
                        Default output prefix for .log, .edr, production .xtc, optional
                        .warming.xtc, .chk, final .pdb, and optional .stress.xvg files. (required)
  -x FILE, --trajectory-output FILE
                        Override the compressed trajectory output path. (default: None)
  -e FILE, --energy-output FILE
                        Override the DROPPS EDR output path. (default: None)
  -g FILE, --log-output FILE
                        Override the fixed-schema progress-log output path. (default: None)
  -cpo FILE, --checkpoint-output FILE
                        Override the latest checkpoint output path. (default: None)
  -c FILE, --final-structure FILE
                        Override the final PDB structure output path. (default: None)
  --stress-output FILE  Override the standalone pressure-tensor output path. (default: None)

Parameters:
  -h, --help            Show this help message and exit.
  --platform {auto,CUDA,OpenCL,CPU,Reference}
                        OpenMM compute platform. (default: auto)
  -gpu_id DEVICE_INDEX, -gpu-id DEVICE_INDEX, --device-index DEVICE_INDEX
                        OpenMM CUDA/OpenCL device index or comma-separated indices. (default:
                        None)
  --precision {single,mixed,double}
                        CUDA/OpenCL arithmetic precision; GPU default is mixed.
  -nt THREADS, --threads THREADS
                        CPU platform thread count; 0 uses the OpenMM default. (default: 0)
  -nsteps NSTEPS, --nsteps NSTEPS
                        Override the TPR production target in steps for this run. (default: None)
  --seed SEED           Override the TPR random seed for a new run. (default: None)
  -cpt CHECKPOINT_MINUTES, --checkpoint-minutes CHECKPOINT_MINUTES
                        Wall-clock minutes between periodic checkpoints; 0 disables periodic
                        saves, but a final or interrupt checkpoint is still saved. (default: 5.0)
  -cpnum, --number-checkpoints
                        Keep step-numbered checkpoint copies in addition to the latest file.
                        (default: False)
  -maxh MAX_HOURS, --max-hours MAX_HOURS
                        Stop production safely after 0.99 times this many hours; -1 disables the
                        limit. (default: -1.0)
  -append, --append     Append restart outputs after rolling them back to the checkpoint. Default:
                        append mode.
  -noappend, --no-append
                        Write restart outputs to the next .partNNNN files. Default: append mode.

Examples:
  dps mdrun -s run.tpr -o run
  dps mdrun -s run.tpr -o run -cpi

Note

This page is generated from the DROPPS 1.0 argument parser. Run dps mdrun -h in the installed environment for the same reference at the command line.