espressomd.io package

Subpackages

• espressomd.io.writer package
  • Submodules
  • espressomd.io.writer.h5md module
  • espressomd.io.writer.vtf module
  • Module contents

Submodules

espressomd.io.mpiio module

class espressomd.io.mpiio.Mpiio

Bases: object

MPI-IO object.

Used to output particle data using MPI-IO to binary files.

Note

See the write() and read() methods for documentation.

read(prefix=None, positions=False, velocities=False, types=False, bonds=False)

MPI-IO read.

This function reads data dumped by :meth`write`. See the :meth`write` documentation for details.

Note

The files must be read on the same number of processes that wrote the data. The data must be read on a machine with the same architecture (otherwise, this might silently fail).

write(prefix=None, positions=False, velocities=False, types=False, bonds=False)

MPI-IO write.

Outputs binary data using MPI-IO to several files starting with prefix. Suffixes are:

• head: Information about fields that are dumped,

• pref: Information about processes: 1 int per process,

• id: Particle ids: 1 int per particle,

• pos: Position information (if dumped): 3 doubles per particle,

• vel: Velocity information (if dumped): 3 doubles per particle,

• typ: Type information (if dumped): 1 int per particle,

• bond: Bond information (if dumped): variable amount of data,

• boff: Bond offset information (if bonds are dumped): 1 int per particle.

Note

Do not read the files on a machine with a different architecture!

Parameters

• prefix (str) – Common prefix for the filenames.

• positions (bool, optional) – Indicates if positions should be dumped.

• velocities (bool, optional) – Indicates if velocities should be dumped.

• types (bool, optional) – Indicates if types should be dumped.

• bonds (bool, optional) – Indicates if bonds should be dumped.

Raises

ValueError – If no prefix was given or none of the output fields are chosen.

Module contents