Skip to content

Join the list

To become part of the list of Jupyter services for German researchers you have to provide a yaml-file with the information listed in the table below. You can check out existing yaml-files in the data folder. In the same folder there is also a template you can base on. To add a logo of your service provide one in the assets folder. When everything is ready, please add a pull request to the main branch of this repository. If you need further assistance or have other questions, do not hesitate to open an issue.

YAML Key                                 Required? Description                 Example
title yes The service name. If too general, should include a provider or project context. NFDI4Ing JupyterHub
provider yes All service providers. University of Stuttgart, Technical University of Darmstadt
service_url yes Link address to the entrypoint for users. https://jupyter.nfdi4ing.de
service_url_requirement no If the service_url is not public available, the requirements to reach it can be stated here. VPN of University of Stuttgart
support yes Email address where to get help. fokus@izus.uni-stuttgart.de
health_api_url no Healthcheck URL green / orange / red
documentation_url no Link address to service specific documentation for users.
target_group_open_for yes Roles, domain, NFDI consortia, University, etc. Researchers of NFDI4Ing
restricted yes Whether any researcher in Germany can use the service false
login_process yes How is the login performed; Login-URL Login via DFN-AAI & eduGAIN
features yes What the service offers.
-- version yes Version of JupyterHub; classic notebook or lab view? 1.x JupyterHub (classic view)
-- programming_languages no List of supported programming languages (kernels). Python, Julia, R
-- environments no List of supported environments (e.g., Python virtualenv) configured as kernels. Python-Keras, Python-PyTorch
-- environments_info no Link to information about the supported environments, if not included under documentation_url.
-- extensions no List of pre-installed JupyterHub extensions.
-- proxy_apps no List of server-proxy featured applications. MATLAB IDE
-- install yes Whether it is allowed to install further packages, kernels, extensions. true
-- shared_folder yes Whether there is a folder to share documents with others. false
-- persistent_storage yes Whether files can survive the docker session. true
-- misc no List of other features you find relevant. dynamic image creation
technicals no Information about the technology stack used.
-- platform no The underlying system below JupyterHub. HPC; Openstack, ...
-- deployment no The way Jupyter is installed on the system. Docker; Kubernetes
-- deployment_url no Link to more (technical) information or a (Git) repository about the deployment or other administration tasks.
-- login_attributes no List of required attributes, entitlements or memberships for login. SAML:persistend_id, SAML:display_name, SAML:email
-- hardware_location no Where the resources (compute / storage) are located. Germany; Europe
-- misc no List of other technicals you find relevant.
resources yes Information about available resources on the instance.
-- default_server_user no Default number of Jupyter servers per user. 1
-- max_server_user yes Maximum number of Jupyter servers per user. 5
-- default_cpu no Default number of CPU for a Jupyter server. 1
-- max_cpu yes Maximum number of CPU for a Jupyter server. 4
-- total_cpu yes Guarantied total number of CPUs for the whole Jupyterhub instance. 50
-- burst_total_cpu yes Scaled total number of CPUs for the whole JupyterHub instance. Normally not exclusive, depends on other factors. 50
-- default_cpu_time no Default CPU time for a Jupyter server. 1 h
-- max_cpu_time yes Maximum CPU time for a Jupyter server. 72 h
-- default_memory no Default amount of memory for a Jupyter server. 100 MB
-- max_memory yes Maximum amount of memory for a Jupyter server. 4 GB
-- total_memory yes Guarantied total amount of memory for the whole Jupyterhub instance. 500 GB
-- burst_total_memory yes Scaled total amount of memory for the whole JupyterHub instance. Normally not exclusive, depends on other factors. 600 GB
-- default_gpu no Default number of GPU for a Jupyter server. 0
-- max_gpu yes Maximum number of GPU for a Jupyter server. 0
-- total_gpu yes Guarantied total number of GPUs for the whole Jupyterhub instance. 10
-- burst_total_gpu yes Scaled total number of GPUs for the whole JupyterHub instance. Normally not exclusive, depends on other factors. 10
-- default_disk no Default amount of disk space (maybe temporary) for a Jupyter server. 10 GB
-- max_disk yes Maximum amount of disk space (maybe temporary) for a Jupyter server. 10 GB
-- default_persistent_disk no Default amount of disk space that survives a Jupyter session. 2 GB
-- max_persistent_disk yes Maximum amount of disk space that survives a Jupyter session. 2 GB
-- total_disk yes Guarantied total amount of disk space for the whole Jupyterhub instance. 50 TB
-- burst_total_disk yes Scaled total amount of disk space for the whole JupyterHub instance. Normally not exclusive, depends on other factors. 50 TB
usage no Statistical information about the Jupyterhub instance.
-- average_daily_sessions yes Average number of Jupyter sessions per day. 10