Official ETH Zürich HPC documentation for Euler: connections, batch system, hardware, software, and services.
Euler
About Empa collaboration with ETH Zürich
Section titled “About Empa collaboration with ETH Zürich”Euler is the central high-performance computing cluster of ETH Zürich, open to research groups across all disciplines.
Empa currently has an allocation on specific nodes of the Euler cluster and provides these resources to its researchers. Since we are still in a test phase, the details of this arrangement with ETH Zürich are not fixed yet. The Scientific IT team at Empa is responsible for managing the collaboration with ETH Zürich. We provide Empa researchers with technical support for compute-related problems and long-term projects.
New Euler users
Section titled “New Euler users”To get started, contact us through the contact page. Since access to Euler for Empa users is managed by us, we will set up your access and provide you with the necessary instructions. The steps below walk you through creating your Euler account and logging in for the first time.
- Make sure you have an Edu-ID that is affiliated with Empa. If you already have an Edu-ID affiliated with Empa, you can use it. If not, you need to create one. Note that an Edu-ID account cannot be created using an Empa e-mail address. First open an Edu-ID account using a private e-mail address, then link the Empa affiliation using the EduID Empa app.
- Once this is done, go to the ETH Zürich service portal and log in with your Edu-ID.
- Register for a service and select Euler HPC.
- Once we have approved your request, you will receive a link that allows you to set a password. Your Euler account is then generated.
- After that, connecting to Euler should work. If you are connected to the Empa network, either directly or via VPN, you can connect to Euler directly. Note that the first-time login is different from the follow-up logins: you will receive a verification code to your e-mail address that must be entered at the prompt. For details, see the user account documentation.
- Once you have access to the shell, try to run your first job:
Terminal window srun date
Platform overview
Section titled “Platform overview”Euler consists of many different types of compute nodes, from general-purpose CPU nodes to GPU nodes with NVIDIA GPUs.
Empa does not buy individual nodes for specific users. Instead, Empa currently has a fixed allocation on a specific set of nodes, which is shared between all Empa users. Since we are still in a test phase, the size and composition of this allocation may change.
Our current allocation covers CPU resources, including large-memory nodes. GPU nodes are not part of our allocation at the moment. If your workloads need GPUs, the CSCS supercomputers are the right resource for you. If you are unsure whether your workload fits within our allocation, please contact us.
Regardless of node type, Empa users have access to:
- Storage: file systems for home, work, and scratch data.
- Software stack: a large library of pre-installed software, from Python and R to engineering and simulation applications.
- Interactive services such as JupyterHub, RStudio server, Code server (VS Code in the browser), and Globus for large-scale data transfer.
Note that the use of Empa software licenses on Euler is currently being established, so some licensed software cannot be run on Euler yet. Open-source software and self-compiled codes are available.
For a complete overview of the cluster, please visit the ETH Zürich HPC documentation.
Support
Section titled “Support”Empa Scientific IT is your first point of contact for general questions about compute, getting access to Euler, and project setup. See the Compute support page for more details. We are also there to help you with technical problems, for example setting up your software environment or getting your jobs to run on Euler.
If you have an Euler-specific technical problem that requires direct support from ETH Zürich, please contact the ETH Zürich HPC support. For system status and maintenance windows, check the Euler system status page.
Resources and links
Section titled “Resources and links”The most important Euler resources for Empa users:
Tutorial covering your first steps on Euler, from login to running your first computations.
How to connect to Euler from the command line and set up your SSH configuration.
Submit and manage computational jobs with Slurm on Euler.
Interactive tool to help you build the right Slurm submission line for your job.
Login nodes, CPU nodes, GPU nodes, storage, and network of the Euler cluster.
Pre-installed software on Euler, including Python, R, MATLAB, machine learning frameworks, and engineering tools.
Interactive Jupyter notebooks on Euler, directly from your browser.
VS Code in your browser, running on Euler.
Current status of the Euler cluster, maintenance windows, and announcements.
Frequently asked questions about using Euler.
Contact information for the ETH Zürich HPC support team.