Euler for D-MATH users
The D-MATH is a shareholder of the HPC cluster Euler. All our resources are in the single share “ls_math” (which is also the default for our members).
Important: all users need to request the access supported by a member (Professor, Senior Scientist) of D-MATH.
Note: On 7 January 2026, the resources of “es_math” have been transferred to “ls_math”. This will lead to a more efficient Euler usage for our department.
Computing resources
The “ls_math” share provides access to 1’504 CPU cores, 5’248 GB of system RAM, 108 GPUs, and 3’968 GB of VRAM. These resources consist of:
| Nodes | Type | GPU Info | System Info | Expiration |
|---|---|---|---|---|
| 5 | medium-memory high-throughput nodes | – | 128 cores, 256 GB memory | 2028-01-31 |
| 16 | high-end GPU resources | 2x NVIDIA RTX 3090 with 24 GB VRAM | 16 cores, 48 GB memory | 2026-10-31 |
| 10 | high-end GPU resources | 2x NVIDIA RTX 4090 with 24 GB VRAM | 16 cores, 48 GB memory | 2028-04-30 |
| 2 | high-end GPU resources | 2x NVIDIA RTX 4090 with 24 GB VRAM | 16 cores, 48 GB memory | 2028-09-30 |
| 16 | high-end GPU resources | 2x NVIDIA RTX 4090 with 24 GB VRAM | 16 cores, 48 GB memory | 2030-02-28 |
| 4 | professional GPU resources | 1x NVIDIA Tesla A100 with 80 GB VRAM | 8 cores, 80 GB memory | 2026-10-31 |
| 16 | professional GPU resources | 1x NVIDIA RTX Pro 6000 Blackwell Server Edition with 96 GB VRAM | 8 cores, 96 GB memory | 2030-02-28 |
Storage resources
On Euler there are different types of storage (see personal storage, global scratch or local scratch, …). If you need storage of type work (optimised for large files, 10 MB to 10 TB and beyond) or of type project (very fast, optimised for a lot of small files, e.g. AI, Machine Learning), please let us know with the request form below.
Note that the current shared size of our work storage is 400 TB and of our project storage 3 TB (if you need more, contact us).
Important: This is a shared resource and if you exceed the quota the whole group is affected, so please check regularly the quota (space and number of files) with the lsquota command.
E.g.:
$ lquota /cluster/work/math
+-----------------------+-------------+----------+-------------+-------------+
| Storage location: | Quota type: | Used: | Soft quota: | Hard quota: |
+-----------------------+-------------+----------+-------------+-------------+
| /cluster/work/math | space | 337.8 TB | 400.0 TB | 440.0 TB |
| /cluster/work/math | files | 22748313 | 100000000 | 110000000 |
+-----------------------+-------------+----------+-------------+-------------+
$ lquota /cluster/project/math
+-----------------------+-------------+----------+-------------+-------------+
| Storage location: | Quota type: | Used: | Soft quota: | Hard quota: |
+-----------------------+-------------+----------+-------------+-------------+
| /cluster/project/math | space | 1.16 TB | - | 3.30 TB |
| /cluster/project/math | files | 4336126 | - | 6600006 |
+-----------------------+-------------+----------+-------------+-------------+
Request access to our resources
If you need access to our share or if you need storage, please submit the following form:
Using the share
List the shares you are allowed to use with the command my_share_info, e.g.:
$ my_share_info
You are a member of the following shareholder groups on Euler:
ls_math
If you have more than one share available, select the desired one while submitting your computation to the slurm batch system, as follow:
$ sbatch -A <share_name> [other options] --wrap "command"
For example:
$ sbatch -A ls_math -G 1 ./my_cuda_program
Help and support
Read the PDF Slides “First Steps on Euler” (courtesy of Dr. Roger Käppeli).
Students of D-MATH and CSE may also contact Dr. Roger Käppeli of SAM for Help, but first, please read the slides. For more detailed information or for technical problems, see the official documentation of the HPC cluster or contact the cluster support team:
URL: https://docs.hpc.ethz.ch
E-mail: cluster-support@id.ethz.ch
Useful links
Posted on
in Computing


