Kebnekaise GPU @ HPC2N

Kebnekaise GPU @ HPC2N (Compute resource)

SUPR uses JavaScript for certain functions. We cannot guarantee that you will be able to use the system with JavaScript disabled.

This resource is for access to the GPU nodes in Kebnekaise.

For CPU nodes see resource 'Kebnekaise'.

For large memory nodes see resource 'Kebnekaise Large Memory'.

Kebnekaise is a heterogeneous computing resource currently consisting of:

GPU nodes:
- 10 Intel® Xeon Gold 6132 Processor (Skylake), 2x14 cores, 192 GB/node
  - 2x NVidia® V100
- 2 AMD® EPYC 7413 (Zen3), 2x24 cores, 512 GB/node
  - 2x NVidia® A100 80G
- 1 AMD® EPYC 7413 (Zen3), 2x24 cores, 512 GB/node
  - 2x AMD® MI100

Compute nodes (separate resource):
- 51 Intel® Xeon Gold 6132 Processor (Skylake), 2x14 cores, 192 GB/node
- 1 AMD® EPYC 7763 (Zen3), 2x64 cores, 1024 GB/node
- To be decomissioned but on standby in case they are truly needed
  - 100 Intel® Xeon E5-2690v4 (Broadwell), 2x14 cores, 128 GB/node
Large Memory nodes (separate resource):
- 8 Intel® Xeon E7-8860v4 (Broadwell), 4x18 cores, 3072 GB/node
- On standby in case they are truly needed
  - 12 Intel® Xeon E7-8860v4 (Broadwell), 4x18 cores, 3072 GB/node

Notes:

GPU nodes are charged differently than ordinary computing nodes.
Access to the CPU nodes are handled through the 'Kebnekaise' resource.
Access to the Large Memory nodes are handled through the 'Kebnekaise Large Memory' resource.
New nodes will be procured on a semi-regular basis.

Open Round

At this moment, you can apply for access to this resource in the following round:

Access to HPC2N resources for researchers at HPC2N parter sites. NAISS Small rules apply.