58 lines
3.1 KiB
Markdown
58 lines
3.1 KiB
Markdown
# PSI HPC@CSCS
|
|
|
|
PSI has a long standing collaboration with CSCS for offering high end
|
|
HPC resources to PSI projects. PSI had co-invested in CSCS' initial
|
|
Cray XT3 supercomputer *Horizon* in 2005 and we continue to procure a share on the
|
|
CSCS flagship systems.
|
|
|
|
The share is intended for projects that by their nature cannot profit
|
|
from applying for regular [CSCS user lab allocation
|
|
schemes.](https://www.cscs.ch/user-lab/allocation-schemes).
|
|
|
|
We can also help PSI groups to procure additional resources based on
|
|
the PSI conditions - please contact us in such a case.
|
|
|
|
## Yearly survey for requesting a project on the PSI share
|
|
|
|
At the end of each year we prepare a survey process and notify all subscribed
|
|
users of the specialized **PSI HPC@CSCS mailing list** (see below) and the
|
|
merlin cluster lists, to enter their next year resource requests. Projects
|
|
receive resources in the form of allocations over the four quarters of the
|
|
following year.
|
|
|
|
The projects requests get reviewed and requests may get adapted to fit into the
|
|
available capacity.
|
|
|
|
The survey is done through ServiceNow, please navigate to
|
|
[Home > Service Catalog > Research Computing > Apply for computing resources at CSCS](https://psi.service-now.com/psisp?id=psi_new_sc_cat_item&sys_id=8d14bd1e4f9c7b407f7660fe0310c7e9)
|
|
and submit the form.
|
|
|
|
Applications will be reviewed and the final resource allocations, in case of
|
|
oversubscription, will be arbitrated by a panel within CSD.
|
|
|
|
### Instructions for filling out the 2026 survey
|
|
|
|
* We have a budget of 100 kCHF for 2026, which translates to 435'000 multicore node hours or 35'600 node hours on the GPU Grace Hopper nodes.
|
|
* multicore projects: The minimum allocation is 10'000 node hours, an average project allocation amounts to 30'000 node hours
|
|
* GPU projects: The minimum allocation is 800 node hours, an average project allocation is 2000 node hours.
|
|
* You need to specify the total resource request for your project in node hours, and how you would like to split the resources over the 4 quarters. For the allocations per quarter year, please enter the number in percent (e.g. 25%, 25%, 25%, 25%). If you indicate nothing, a 25% per quarter will be assumed.
|
|
* We currently have a total of 65 TB of storage for all projects. Additional storage
|
|
can be obtained, but large storage assignments are not in scope for these projects.
|
|
|
|
## CSCS Systems reference information
|
|
|
|
For 2025 we can offer access to [CSCS Alps](https://www.cscs.ch/computers/alps) Eiger (CPU multicore) and Daint (GPU) systems.
|
|
|
|
* [CSCS User Portal](https://user.cscs.ch/)
|
|
* Documentation
|
|
* [CSCS Eiger CPU multicore cluster](https://docs.cscs.ch/clusters/eiger/)
|
|
* [CSCS Daint GPU cluster](https://docs.cscs.ch/clusters/daint/)
|
|
|
|
## Contact information
|
|
|
|
* PSI Contacts:
|
|
* Mailing list contact: <psi-hpc-at-cscs-admin@lists.psi.ch>
|
|
* Marc Caubet Serrabou <marc.caubet@psi.ch>
|
|
* Derek Feichtinger <derek.feichtinger@psi.ch>
|
|
* Mailing list for receiving user notifications and survey information: <psi-hpc-at-cscs@lists.psi.ch> [(subscribe)](https://psilists.ethz.ch/sympa/subscribe/psi-hpc-at-cscs)
|