Navigation and service

UNICORE Access to the Supercomputers at Forschungszentrum Jülich

This page is for users of the Forschungszentrum Jülich supercomputers who want to use UNICORE (UNiform Interface to COmputing REsources) to access the supercomputer resources.

  • UNICORE allows you to run jobs, access and transfer data and run workflows on HPC systems in Jülich (and possibly in other HPC centres as well), and to integrate these functionalities into your applications.
  • UNICORE primarily offers RESTful APIs that can be used from programming languages such as Python and environments such as Jupyter Notebooks running on Jupyter-JSC.
  • A full-featured command line client (UCC) is available that you can download and install on your local workstation. UCC is a Java application and requires a Java runtime environment (version 1.8 or later).

The prerequisite for using UNICORE is that you have an account at JSC for at least one of the HPC systems. The authentication can be done with your webservice (JUDOOR) account, or one of the other accepted authentication methods, such as OAuth tokens from the Human Brain Project.


  • UCC is available from Sourceforge. You will find the latest version at the top of the list. The installation and configuration is described in the UNICORE UCC manual, which is also part of the installation package.

  • For directly using the RESTful APIs we provide a helper library PyUNICORE that can be installed from PyPI with „pip install -U pyunicore“

Jülich-specific Settings for UCC

  • You can use your JUDOOR username/password to authenticate. In the UCC preferences file, set
    username=<your judoor username>
  • Define a truststore containing at least the dfn-verein global_certificate_signer_chain
  • The registry address containing links to the Jülich HPC systems is


Contact in case of questions and problems.