The detailed documentation can be found on this link. It contains detailed information about the hardware and software of the cluster. There is also an example on how to run your code on the cluster using the workload manager (Slurm) and the container virtualization software (Singularity).

In 2024-06 login nodes are available

In 2024-07 four of the nodes of the TCML were exchanged with different hardware. All 4 new nodes have 8 x 2080Ti  GPUs - instead of 4 x 1080Ti. Both GPU types have 11 GB RAM. Also newer CPU processors and better SSD. Thanks to the ML-Cloud for giving us this servers.

In 2025-04 four of the nodes of the TCML were exchanged with new hardware. All 4 new nodes have 8 x L40S GPUs with 48 GB of GPU-RAM. This upgrade was founded by 8 Professors and in the beginning only members of those groups are allowed to use the L40S hardware. In the beginning, L40S usage is additional limited to jobs with max 7 days calculating time.

If you are in the group of  …  and want to use L40S please contact  …
and tell him/her your TCML user name and maybe the reason

The next maintenance is scheduled

from Monday, 2026-10-05 until Friday, 2026-10-16 (up to 2 week)

Account creation while maintenance is not possible, will be delayed.

TCML has 2 types of accounts

Regular accounts with a login like “last name” or “first letter of first name + last name”.
Course accounts for lectures like “studnnn” where nnn is a number.
Course accounts have a lifetime of the course duration and will be deleted end of March (for winter courses) or end of September (for summer courses)

Tutors can request a batch of accounts if they send us a list: 
Name of the lecture/course
First Name; Last Name; Email address

Feedback from tutors regarding the performance of the hardware for course xy with nn students is highly encouraged for future capacity planning.

Deleting an account:
If your work is finished, please backup your data including singularity-images and send us an email.

We don't like accounts which are unused or used for storing data without usage of GPUs and will cleanup regularly. In that case you will receive an email with further details. Account without valid email address will be deleted. If email address change: send us an email with new communication email address.

Unsubscribe tcml-users (organizational information) or tcml-discuss (technical issues) can be done via listserv or send us an email. tcml-users mailing list is recommended for active users.

please use you university email address like
@uni-tuebingen.de  or  @student.uni-tuebingen.de  or  @ai.uni-tuebingen.de  or  mpi-address  or  similar
for applying.
If you want to communicate (also tcml-user@listserv…) with an alternative email address you can use the comments field.

If you don't get an answer within 3 working days, please send a reminder.

To use the cluster, the users must login to one of 3 available login node and execute their tasks from there. The login nodes have the following IP addresses or aliases:
IP address: 134.2.17.166
Alias: login1.tcml.uni-tuebingen.de

IP address: 134.2.17.248
Alias: login2.tcml.uni-tuebingen.de

IP address: 134.2.17.202
Alias: login3.tcml.uni-tuebingen.de

use ssh in the following way:

$ ssh username@login3.tcml.uni-tuebingen.de

ssh is only available from inside of the university network (134.2.x.y)

remote development only allowed on login3

We use the data provided to us only to create a user account on the cluster and to contact you with issues concerning the cluster. We wont share given data with anyone. Administrators are allowed to delete your data after warning you beforehand. As long as it does not contradict to the former three sentences, the privacy policy of the Eberhard Karls Universitaet Tuebingen holds.

If you have any question, complaint, suggestion of improvement or anything else feel free to contact us via tcml-contactspam prevention@listserv.uni-tuebingen.de.