In order to access the resources of the Center for Scientific Computation, one must both valid login and passwords, provided by the administration staff upon requesting an account via the available form.

Once you have an account you can access the Center for Scientific Computing through the command line using the ssh command. You have three possible access nodes in order to login.

If access from outside the UAM is attempted, one needs to use the VPN services provided by IT.

For the case of RES users external to the UAM, their IP must be provided to us in order to enable their access.

Using Linux is highly recommended. In Linux basic commands a brief introduction to the basic commands needed to work with said operative system can be found.

If you are using Windows, we recommend the usage of emulators such as PuTTY or MobaXterm.

If you use MacOS, you can use the Terminal app. If you need to make use of graphic programs, installation of XQuartz will be necessary.

Through the usage of the sbatch command, following the syntax

sbatch -A name-project -p name-queue script

where -A indicates the name of the project or ‘account’ and -p indicates the queue or ‘partition’.

More information about this can be found in The Slurm queueing system. Alternatively, at CCC user's guide an even more thorough explanation of this command and its most frequent options can be found.

Through the usage of the projects command one can know to which projects can time be charged, along with the queues accessible for those projects.

To know if you have enough time for job submissions, you can execute timeleft.

We can run squeue to check how your jobs are going, or the command infocolas and follow the menu's options.

To this end, we can run sinfo or infocolas, where a menu with different options is displayed. Wandering over said menu general information on the allocation status, and adequate your jobs to the circumstances.

The deactivation of the projects can be due to them having consumed all their assigned computation time. One can check if this is the case running timeleft and observing the number of seconds remaining for each project.

High performance computation projects are renewed the first day of every month, while user computation projects are renewed every three months.

Software is installed in /usr/local. To access it, it is advisable to check whether said software is provided with environment modules. These modules load the routes to the software, as well as any libraries necessary for its execution. You can check the environment modules available running module avail. Afterwards, you will need to load the corresponding module to your script.

For example, in order to load the the Gaussian module, one must include the line module load gaussian.

Due to the high demand of the simulation lab, we cannot indicate concrete dates with much anticipation, however they will always be announced in General information on courses.