Gaussian

[en]

Prerequisities

• Science (CNCZ) account.
• Being member of the qm, qmorgchem or qmteaching and gaussian groups (ask Jan for access).
• Vpn (when working from home).
• SSH client (linux usually has OpenSSH, for instance Putty for Windows - guide).
• X11 client (linux usually has OpenSSH, for instance xming for Windows, guide).

Setting-up the environment

you must be within university network or connected through VPN.

Tedious, but explained

To work effectively with the cluster its wise to edit your ~/.profile file so you can easily execute Gaussian calculations and do few other tweaks. In following section I will explain both what to do and why to do it. Lest start with a ~/.profile. ~/.profile is a file which sets up working environment when connecting for an interactive session. The recommended version looks like this:

export PATH=$PATH:/vol/qmorgchem/scripts
module load Gaussian-16
umask 027

• export PATH=... - PATH is a variable where folders which contains executable files are linked - by this line you ensure that /vol/qmorgchem/scripts is added to PATH so you can execute scripts stored there - for instance a script for submitting a calculation.
• module - this line ensures that you modify your environment to be able to run Gaussian jobs properly.
• umask 027 gives permissions to newly created files - after the profile is active, this umask ensures that you have all the rights to work with your files, people in the same groups can view the files, but cannot edit them and people from other groups does not have permission to inspect the files. More about umask is on wiki, especially octal codes gets thing a bit more clear.

After you change the ~/.profile file accordingly and you log-out and log-in you should be able to get the calculations running.

To get things more smooth it is also recommended to create shortcuts in your home directory.

• Log-out and log-in if you've not done so after modifying your profile.
• Then execute following commands:

cnorris@lilo7:~$ mkdir -p /vol/qmorgchem/"$USER"

if you're a memeber of the qmorgchem group:

cnorris@lilo7:~$ mkdir -p /vol/qmorgchem-old/"$USER"
cnorris@lilo7:~$ chown $USER:qmorgchem /vol/qmorgchem/"$USER"
cnorris@lilo7:~$ chown $USER:qmorgchem /vol/qmorgchem-old/"$USER"

By this you created your own directory on the qmorgchem and optionally on the qmorgchem-old disks. By optional chown you changed the group to which this folder belongs.

USER is just another environment variable, to get what it does run following command:

cnorris@lilo7:~$ echo $USER

hopefully this clears the things a bit.

Now its time to create shortcuts - they're called symlinks in linux.

cnorris@lilo7:~$ ln -s /vol/qmorgchem/"$USER"/ ~/running

and optionally (if you're part of qmorgchem):

 cnorris@lilo7:~$ ln -s /vol/qmorgchem-org/"$USER"/ ~/old

ln is a command to link things together and -s parameter tells it that it should create a shortcut.

Fast & Dirty

Login to lilo.science.ru.nl or slurm20.science.ru.nl, then execute:

cnorris@lilo7:~$ /vol/qmorgchem/scripts/gaussify

ẂARNING this will change the access rights to your newly created files, if you do not like it, comment (add # at start) the line starting with umask in ~/.profile.

logout and login after executing the command so the changes can take place.

Checking calculations, submitting jobs, checking queues

submitting jobs, checking queues

There are currently two queues available on the cluster: orgfast for small and quick jobs and orglong for bigger and more-demanding jobs. We are currently using Slurm as our job queuing manager.

To submit a job, type:

cnorris@slurm20:~$ G16 $job_filename $queue

command where $job_filename is filename of your Gaussian input and $queue is either orgfast or orglong.

To submit a job and doing anything with the running jobs you MUST be on slurm20

To check current queues, type:

cnorris@slurm20:~$ squeue

Once calculation is running a .log file will start to appear in the folder where is the input file.

Checking calculations

once calculations are finished or you want to take a look, you can either use molden (its installed on the cluster) or you can connect to the disks.

Connecting to disks - windows

Open file explorer and into navigation bar type:

\\qmorgchem-srv.science.ru.nl\qmorgchem\

for qmorgchem-old, the link is:

\\qmorgchem-old-srv.science.ru.nl\qmorgchem-old\

Disclaimer

This guide is a draft, if you have any suggestions how to improve the readability, understandability or how to make it better do not be shy and write Jan.