Apply to this round for a project on the Berzelius SuperPOD at NSC. Berzelius is not a NAISS system. It is a direct donation to NSC from the Knut and Alice Wallenberg foundation (KAW). Projects with financial support from KAW should have priority according to the donation letter.
To apply, you must be a scientist in Swedish academia, at the level of PhD student or higher.
At present, almost exclusively projects with financial support from KAW are granted due to the high demand from such proposals.
Deadlines and Decisions
Handling of projects requesting a default allocation:
Proposals requesting a default allocation will normally be handled within a week.
Handling of proposals requesting more than a default allocation:
Monthly evaluation of proposals during the year.
Proposals submitted at the latest on the 15th will undergo review during the same month.
July and December have different schedules.
Proposals submitted after June 15 will be processed in August. i.e. with a first possible allocation starting September 1st for larger than default requests!
This round is open for proposals until 2025-01-01 00:00.
|
Resource |
Centre |
Upper
Limit |
Available |
Unit |
Note |
|
|
Berzelius Compute |
NSC |
28 800 |
345 600 |
GPU-h/month |
Applications are normally evaluated during the last week each month
|
|
|
Submit your proposal at least one week before the end of a month to be considered for an allocation from the first of the following month. Received proposals will be evaluated against each other and time that become available as project ends at the end of a month will be allocated to the proposed projects accordingly.
Berzelius is an NVIDIA SuperPOD consisting of 94 DGX-A100 nodes, sporting a total of 752 NVIDIA A100 GPUs.
The SuperPOD uses the SLURM resource manager and job scheduler. The original 60 DGX-A100 nodes have 8x NVIDIA A100 GPUs (40GB), 128 CPU cores (2x AMD Epyc 7742), 1 TB of RAM and 15 TB of NVMe SSD local disk. The 34 newer DGX-A100 nodes have 8x NVIDIA A100 GPUs (80GB), 128 CPU cores (2x AMD Epyc 7742), 2 TB of RAM and 30 TB of NVMe SSD local disk. High performance central storage is available using 4x AI400X and 2x AI400X2 from DDN, serving 1.5 PB of storage space to all nodes of the cluster. All DGX-A100 GPUs have dedicated Mellanox HDR InfiniBand HBAs, that is, there are 8 Mellanox HDR HBAs per DGX-A100 node, connected in a full bisection bandwidth, fat-tree topology.
|
|
|
|
|
Berzelius Storage |
NSC |
— |
931 322 |
GiB |
Applications are normally evaluated during the last week each month.
|
|
|
Submit your proposal at least one week before the end of a month to be considered for an allocation from the first of the following month. Received proposals will be evaluated against each other and time that become available as project ends at the end of a month will be allocated to the proposed projects accordingly.
High performance central storage is available using 4x AI400X and 2x AI400X2, serving a total of 1.5 PB of storage to all nodes of the cluster via a dedicated InfiniBand interconnect. Aggregate read IO performance is 320 GB/s from the central storage and the dedicated data interconnect bandwidth per node is 25 GB/s.
NSC centre storage (as available on Tetralith) is not accessible on Berzelius.
|
|
|
