The dump netcdf keyword

From GPUMD
Jump to navigation Jump to search

Purpose

Dump the atom positions (coordinates) in the NetCDF format.

Grammar

  • This keyword's usage will have the following format:
dump_netcdf interval <optional_args>
  • The interval parameter is the output interval (number of steps) of the atom positions.
  • The optional_args provide additional functionality. Currently, the following optional argument is accepted:
    • precision: single or double. If precision is single, the output data are 32-bit floating point numbers. If precision is double, the output data are 64-bit floating point numbers. The default value is double.

Requirements and specifications

  • This keyword requires an external package to operate. The binary format NetCDF package instructions can be found here.
  • The single option is good for saving space, but does not follow the official AMBER 1.0 conventions.
  • All NetCDF output files can be read by VMD and OVITO for visualization.
  • The NetCDF files also stores cell lengths and angles which can be used in the visualization software to illustrate boundaries and show periodic copies of the structure.

Examples

Single precision example

To dump the positions every 1000 steps to a NetCDF file with 32-bit floating point values, one can add

   dump_netcdf 1000 precision single

before the run.

Double precision example

To dump the positions every 1000 steps to a NetCDF file with 64-bit floating point values, one can add

   dump_netcdf 1000 

before the run.

Caveats

  • This keyword is not propagating. That is, its effect will not be passed from one run to the next.
  • The outputs append to the same file for different runs in the same simulation. Re-running the simulation will create a new output file.
  • If the precision changes between different runs, the first defined precision will still be used (i.e. changes in precision are ignored during a simulation).

Output file