Purdue University

Can you remove me from the Gautschi mailing list?

Your subscription in the Gautschi mailing list is tied to your account on Gautschi. If you are no longer using your account on Gautschi, your account can be deleted from the My Accounts page. Hover over the resource you wish to remove yourself from and click the red 'X' button. Your account and mailing list subscription will be removed overnight. Be sure to make a copy of any data you wish to keep first.

How is Gautschi different than other Community Clusters?

Gautschi differs from the previous Community Clusters in many significant aspects:

• Gautschi has a unique sales protocol. While previous clusters were available for sale based on compute resources, Gautschi is sold based on compute resource hours.
• Gautschi has eight NVIDIA H100 GPUs, each with 80 GB of GPU memory.

Do I need to do anything to my firewall to access Gautschi?

No firewall changes are needed to access Gautschi. However, to access data through Network Drives (i.e., CIFS, "Z: Drive"), you must be on a Purdue campus network or connected through VPN.

Does Gautschi have the same home directory as other clusters?

The Gautschi home directory and its contents are exclusive to Gautschi cluster front-end hosts and compute nodes. This home directory is not available on other RCAC machines but Gautschi. There is no automatic copying or synchronization between home directories.

At your discretion you can manually copy all or parts of your main research computing home to Gautschi using one of the suggested methods.

If you plan to use hsi or htar commands to access Fortress tape archive from Gautschi, please see also the keytab generation question for a temporary workaround to a potential caveat, while a permanent mitigation is being developed.

How should I launch common GUI applications on Gautschi?

When our users access Gautschi via ThinLinc, they can follow these steps to launch an interactive jobs.

• In the upper left corner, the user can click on Applications, then Cluster Software, where multiple software options are listed with interactive SLURM jobs.

• The GUI launcher starts with a window that prompts the user to select the desired version of the software to launch and guides them through the job submission process.

• The GUI launcher also makes it easy for users to view available accounts, the maximum wall times for each account, and the available computing resources with multiple help options at the bottom.

• After choosing the partition, the user is prompted to provide the computing resource they need for their job to run while adjusting the memory for them automatically.

• After requesting the resources the job will be submitted and waiting for slurm to allocate the computing resources.