Computing resources

 

HAICORE: computing resources for the Helmholtz AI community

Access to computing resources is key for the Helmholtz AI community to accelerate innovative AI applications.

The “Helmholtz AI computing resources” (HAICORE) provide easy and low-barrier GPU access to the entire AI community within the Helmholtz Association.

HAICORE provides two usage modalities:
1. Ad-hoc usage for an initial playing with the data and/or AI methods
2. Lightweight projects of up to 5,000 GPU hours per half year

For ease of access, computing services are delivered through web portals and traditional ssh login.

 

How can you get access to the HAICORE resources?


Ad-hoc usage

For ad-hoc usage with up to 10 GPU/h per day - GPUs implemented at KIT

Registration for HAICORE@KIT can be done via FeLS.


During your login process, first select Helmholtz AAI as your organization then you can select your respective Helmholtz institution.
Further information can be found here.

In order to use the system, make sure that you have the KIT-specific HPC-VPN turned on. You may access the system either via regular ssh connection or Jupyter Notebooks in the browser.

Here are the links to the respective documentations:
VPN / HAICORE / Jupyter

 

Need help? Open a Ticket at our support portal.

-------------

After a positive formality check, an account will be generated and you will be informed that you now have access to the system. Please note that due to an overflow of work, only 5 people per week will be granted access.

-------------

Any work performed on the HAICORE@KIT partition should be appropriately cited

"This work was supported by the Helmholtz Association's Initiative and Networking Fund on the HAICORE@KIT partition."


 

Lightweight projects

For lightweight project usage with up to 5,000 GPU/h per half year  - GPUs implemented at FZJ

The HAICORE resource allocation is set up through the voucher system, compute projects must be submitted there. These are the required steps:

  1. Create an account at JuDoor
  2. Submit your HAICORE proposal

-------------

Make sure that your code makes good use of the GPUs: On Juwels Booster, one of the fastest supercomputer in Europe, your job will be assigned at least one full compute node with four GPUs. If you use only one, the usage of four will be deducted from your account.

More information about Juwels

-------------

Any work performed on the HAICORE@FZJ partition should be appropriately cited

"This work was supported by the Helmholtz Association's Initiative and Networking Fund on the HAICORE@FZJ partition."


 

Bigger projects?

If a proposal exceeds 5,000 GPU hours, it will be redirected to the already well established tier 1 and tier 2 project allocation process. The Helmholtz AI consultant teams can provide support for the application process via the voucher system.

 

 

 

Questions?

The HAICORE ramp-up policy contains all relevant information for accessing the resources during HAICORE’s ramp-up phase. If you have any further questions, we are happy to help! Send us an e-mail to contactnoSp@m@helmholtz.ai.