Dear cluster users,

in the SSH Client PuTTY a vulnerability was found (CVE-2024-31497  <https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2024-31497>), which allows an Attacker to gain access to the user private key.
This vulnerability only effects ECDSA keys, if RSA and ED25519 keys are not effected.

If you use PuTTY with ECDSA keys to conncet to the Cluster please follow the following steps in order remedy the issue:

* Update PuTTY to the version 0.81
* Create a new key pair
* In the/authorized_keys/  file Replace the public key of the key pair you have been using so far with the public key of the newly generated key pair

Please be aware that besides PuTTY the following programs are also efected by the vulnerability:

* Filzilla
* WinSCP
* TortoiseGit
* TortoiseSVN

Best regards

Your HPC support team

Mit freundlichen Grüßen Ihr HPC-Support-Team

Dear cluster users,

GPU sharding is now available on the OMNI cluster.
GPU sharding enables the parallel use of a GPU by multiple Slurm jobs, this allows a more efficient use of our GPU resources.
Please check if the use of GPU Sharding is feasible for your computations.

More information is available on our website:

GPUs

If you have questions or problems with using the cluster, open a ticket
by sending an e-mail to:

hpc-support@uni-siegen.de

Best regards
Your HPC support team

The GCC-Compiler version 12.2.0 is now available on the OMNI, which replaces the GCC-Compiler version 11.2.1. To use GCC-Compiler version 12.2.0, it is necessary to load the module gnu12/12.2.0.

$ module load gnu12/12.2.0

The GCC-Compiler version 11.2.1 will be removed on 01.05.2023. As the default compiler, version 9.3.0 is still loaded.

Dear Cluster Users,

Groupspaces are now on available on the OMNI.

Groupspaces are group folders for teams which are accessible from the login- and compute-nodes. Groupspaces will only be created on request of the team leader. More information regarding Groupspaces is available on our Website:

https://cluster.uni-siegen.de/omni/usage/file-systems/?lang=en#groupspaces

Dear Cluster users,

due to the current situation we turned off about 15 % of our compute-nodes as an energy-saving measure.

We apologize for any inconveniences.

Best Regards,

Your HPC-Team