Run Jobs from the Command Line¶

Visualize the results in the GUI¶

GUI modules can also be started from the command line. For example, to open the geometry optimization trajectory in AMSmovie, run

$AMSBIN/amsmovie ams.results

Customize the run command¶

• To view the output on the screen, as well as save it to a file, instead run sh jobname.run | tee jobname.out

• You can also make the jobname.run file executable, if you prefer:

chmod +x jobname.run
./jobname.run > jobname.out

#!/bin/sh¶

This line indicates how the file should be executed. In this case, it means to use the sh shell interpreter. If you run the file using

• sh jobname.run > jobname.out, this line is optional, but using

• ./jobname.run > jobname.out, it is required.

“$AMSBIN/ams” << eor¶

This line executes the ams program. AMSBIN is an environment variable, and points to the directory containing the ams file. When you open a terminal following these instructions, the AMSBIN variable will be set to the correct value.

<< eor means that all the following lines, until there is a line exactly equal to eor, will be used as input for ams. The input to ams thus becomes

Task GeometryOptimization
System
    Atoms
        C  0.01247  0.02254  1.08262
        C -0.00894 -0.01624 -0.43421
        H -0.49334  0.93505  1.44716
        H  1.05522  0.04512  1.44808
        H -0.64695 -1.12346  2.54219
        H  0.50112 -0.91640 -0.80440
        H  0.49999  0.86726 -0.84481
        H -1.04310 -0.02739 -0.80544
        O -0.66442 -1.15471  1.56909
    End
End

Engine ADF
    Basis
        Type SZ
    End
EndEngine

Task and System block¶

When ams reads the Task GeometryOptimization line, it knows to perform a geometry optimization.

Similarly, the System block tells ams for which molecule to run the calculation. Each line contains an atomic element and the atomic coordinates in angstrom.

Technically, the Task and System block are input options to the AMS Driver. All allowed keywords are described in the AMS Driver manual.

Engine ADF block¶

The lines between Engine ADF and EndEngine are passed on to the ADF engine. In this case, the Basis block is used to specify a single-ζ (SZ) basis set.

All keywords within the Engine ADF block are described in the ADF manual.

Set the job name¶

ams will place all its output files in a results directory. By default, this directory is called ams.results. You can change it by replacing the "$AMSBIN/ams" << eor line with:

AMS_JOBNAME=my_jobname "$AMSBIN/ams" << eor

This will create the directory my_jobname.results instead.

Overwrite old results¶

If you want to overwrite the results if they already exist, add the --delete-old-results flag:

"$AMSBIN/ams" --delete-old-results << eor
    # ams input
eor

Run in serial/parallel¶

To set the number of processes used by ams, specify the -n flag. To run in serial (using only 1 core):

"$AMSBIN/ams" -n 1 << eor
    # ams input
eor

Alternatively, you can set the environment variable NSCM:

export NSCM=1
"$AMSBIN/ams" << eor
    # ams input
eor

Place the input to ams in its own file¶

Save the input to ams (all lines starting from Task to EndEngine) in its own file jobname.in.

Then the run script jobname.run can be simplified to become:

#!/bin/sh

"$AMSBIN/ams" < jobname.in > jobname.out

Because this script is now really only 1 line, you could also run this command directly in the terminal: "$AMSBIN/ams" < jobname.in > jobname.out

Specify a .xyz file for the geometry¶

When editing shell scripts manually, it is often more convenient to read the atomic coordinates from an external file. You can read the coordinates from .xyz files, for example, the output.xyz file from a previous calculation. You can also create or open the structure in AMSinput GUI and choose File → Export coordinates → xyz.

Then, in your shell script change the System block to become:

System
    GeometryFile my_xyz_file.xyz
End