- Readme: 我爱搜集网关于【LAMMPS文档中文翻译系列】的说明
atom_modify keyword values ...
- keyword = map or first or sort
- map: value = arrayorhash
- first: value = group-ID
- sort: value = Nfreq binsize
atom_modify map hash atom_modify map array sort 10000 2.0 atom_modify first colloid
The first keyword allows a group to be specified whose atoms will be maintained as the first atoms in each processor's list of owned atoms. This in only useful when the specified group is a small fraction of all the atoms, and there are other operations LAMMPS is performing that will be sped-up significantly by being able to loop over the smaller set of atoms. Otherwise the reordering required by this option will be a net slow-down. The neigh_modify include and communicate group commands are two examples of commands that require this setting to work efficiently. Severalfixes, most notably time integration fixes like fix nve, also take advantage of this setting if the group they operate on is the group specified by this command. Note that specifying "all" as the group-ID effectively turns off the first option.
It is OK to use the first keyword with a group that has not yet been defined, e.g. to use the atom_modify first command at the beginning of your input script. LAMMPS does not use the group until a simullation is run.
The sort keyword turns on a spatial sorting or reordering of atoms within each processor's sub-domain every Nfreq timesteps. If Nfreq is set to 0, then sorting is turned off. Sorting can improve cache performance and thus speed-up a LAMMPS simulation, as discussed in a paper by (Meloni). Its efficacy depends on the problem size (atoms/processor), how quickly the system becomes disordered, and various other factors. As a general rule, sorting is typically more effective at speeding up simulations of liquids as opposed to solids. In tests we have done, the speed-up can range from zero to 3-4x.
Reordering is peformed every Nfreq timesteps during a dynamics run or iterations during a minimization. More precisely, reordering occurs at the first reneighboring that occurs after the target timestep. The reordering is performed locally by each processor, using bins of the specified binsize. If binsize is set to 0.0, then a binsize equal to half the neighbor cutoff distance (force cutoff plus skin distance) is used, which is a reasonable value. After the atoms have been binned, they are reordered so that atoms in the same bin are adjacent to each other in the processor's 1d list of atoms.
The goal of this procedure is for atoms to put atoms close to each other in the processor's one-dimensional list of atoms that are also near to each other spatially. This can improve cache performance when pairwise intereractions and neighbor lists are computed. Note that if bins are too small, there will be few atoms/bin. Likewise if bins are too large, there will be many atoms/bin. In both cases, the goal of cache locality will be undermined.
IMPORTANT NOTE: Running a simulation with sorting on versus off should not change the simulation results in a statistical sense. However, a different ordering will induce round-off differences, which will lead to diverging trajectories over time when comparing two simluations. Various commands, particularly those which use random numbers (e.g. velocity create, and fix langevin), may generate (statistically identical) results which depend on the order in which atoms are processed. The order of atoms in a dump file will also typically change if sorting is enabled.
- 分子问题默认map = array。