I am the following issue when running my LAMMPS input file:
ERROR: Invalid bond type in create_bonds command (../create_bonds.cpp:134)
Last command: create_bonds many Mo S 1 2.0 2.5
I suspect that is has something to do with the formatting of my data file as I did not specify bond types. When looking through the documentation for this I am getting confused about how to specify a general bond for example between H and O in water or Mo and S in MoS2. Is this done in the data file or input file? Do I need software to encode all bond information in the data file (will be tedious - 900 atoms system), I tried to get around this by using create_bonds command? My input file only runs without error when commenting out all bond and angle related information. I attached all necessary files below.
# Define initial structure
units real
atom_style full
# Read positions data file
read_data data.lammps
# Create bonds for the data file
# Note that coefficients are as follows for atoms (from data.lammps)
group H type 1
group O type 2
group Mo type 3
group S type 4
# Mo-S bonds
create_bonds many Mo S 1 2.0 2.5
# H-O bonds
create_bonds many H O 2 1.0 1.2
Your datafile contain no information concerning bonds. Therefore when reading it LAMMPS does not consider any bond information when creating the system. Consider reading about the optional keywords of the read_data command as well as the note of the create_bonds command.
I was able to start running the calculation after encoding the bond/angle info myself into the .data file via a python script and fixing/rearranging my input script. I was now wondering if there are methods to speed up computation. It seems that it is stuck after performing minimization as shown by the output log. Could you please help me solve this problem? All updated/necessary files are linked below. Please let me know if you need any information regarding computer architecture / memory allocation on my HPC.
Optimizing computation time and figuring out why a simulation is stuck are two different things. From the log file you provide, a final energy value of -5\cdot{}10^{17} kcal/mol after minimization seems bogus and there is likely something wrong either with your configuration or your model. You should double check your inputs, that is not something I have the time nor the will to do at the moment.
On the other hand, concerning computation speed, this will depend on your cluster/computing center installation (if any) and what you want to model (i.e. available versions of the potentials). There are general tips in the LAMMPS manual in a dedicated section or the balance command section, and several discussions here about parallelization, or KOKKOS and the GPU packages use. Additionally, if you can, you should reach any software/computer engineer that should tell you the what and the how of your installation and how to get the best out of what is at your disposal