You need 2 scripts located in /opt/scripts/scratch-scripts/

  • To visualize your data in the /scratch of the nodes :

sh /opt/scripts/scratch-scripts/ and follow the instructions

  • To delete your data in the /scratch of the nodes:

sh /opt/scripts/scratch-scripts/ and follow the instructions


Your data contained in the temporary /scratch spaces of nodes are automatically deleted after 45 days on the nodes of the long partition.

On the nodes of other partitions, your data contained in the nodes’ temporary /scratch spaces are automatically deleted after 30 days.

However, remember to delete your data on the /scratch nodes when they are no longer in use.


  • Access

Access to the GPU partition is restricted.

A reasoned request must be made by a holder of one of the UMRs on the board here:

In case of acceptance, your access is valid for 3 months.

  • Usage

Once you have been added to the gpu_group, use the following options to use the gpu partition:

-p gpu -A gpu_group

For example, to open an interactive session on node 26:

srun -p gpu -A gpu_group

In order to distribute the resources among all, each user has the right to launch jobs on 32 cores at the same time.

If you need more cores for a limited period of time, you can make a reasoned request to which will be able to increase your number of cores temporarily.

Once connected to, type the command


Type the new password and press enter.

Enter the password a second time and then press enter.

The new password will be propagated to all cluster machines in a few minutes.

To quote us, please copy the following sentence into your publications:

“The authors acknowledge the ISO 9001 certified IRD i-Trop HPC (member of the South Green Platform)  at IRD montpellier for providing  HPC resources that have contributed to the research results reported within this paper. URL:”