Kebnekaise

Details:
[ Compute (Skylake-SP) nodes | Compute nodes (AMD Zen3) | Compute nodes (AMD Zen4) | Largemem nodes | GPU nodes (V100) ]
[ GPU nodes (A100) | GPU nodes (A40) | GPU nodes (MI100 AMD) | GPU nodes (A6000) | GPU nodes (L40s) | GPU nodes (H100) | GPU nodes (A40) ]

Kebnekaise is the current supercomputer at HPC2N. It is named after the massif of the same name, which has some of Sweden's highest mountain peaks (Sydtoppen and Nordtoppen). Just as the massif, the supercomputer Kebnekaise is a system with many faces. 

Kebnekaise is being continuously updated with new hardware as old hardware is retired.

Kebnekaise timeline

• 2016: it was delivered by Lenovo and installed during the summer (except for the largemem nodes, all the old Broadwell and K80 nodes have been decommissioned)
• November 7, 2016: it was opened up for general availability
• 2017: 36 nodes with the (then) new generation of Intel Xeon Phi, also known as Intel Knights Landing (KNL), were installed during spring (now decommissioned)
• 2018: extended with 52 Intel Xeon Gold 6132 (Skylake) nodes, as well as 10 NVidian V100 (Volta) GPU nodes.
• 2023: extended with 2 dual NVIDIA A100 GPU nodes and one many-core AMD Zen3 CPU node.
• 2024: extended with
  • 8 nodes with 2x128 cores Zen4 CPUs
  • 2 nodes with 2x24 cores Zen3 CPUs, 496 GB of usable memory, 2xA100 GPUs
  • 1 node with 2x24 cores Zen3 CPUs, 496 GB of usable memory, 2xMI100 AMD GPUs
  • 1 node with 2x24 cores Zen4 CPUs, 310 GB of usable memory, 2xA6000 GPUs
  • 10 nodes with 2x24 cores Zen4 CPUs, 310 GB of usable memory, 2xL40s GPUs
  • 2 nodes with 2x48 cores Zen4 CPUs, 620 GB of usable memory, 4xH100 SXM5 GPUs
• 2024 (fall):
  • 1 node with 2x64 cores Zen4 CPUs, ~740 GB of usable memory, 8xA40 GPUs
  • 2 nodes with 2x64 cores Zen4 CPUS, ~740 GB of usable memory, 6xL40s GPUs

Kebnekaise Celebration and HPC2N Open House was held 30 November 2017.

There is local scratch space on each node, which is shared between the jobs currently running. Connected to Kebnekaise is also our parallel file system Ransarn (where your project storage is located), which provide quick access to files regardless of which node they run on. For more information about the different file systems that are available on our systems, read the Filesystems and Storage page.

All nodes are running Ubuntu Jammy (22.04 LTS). We use EasyBuild to build software and we also use a module system called Lmod. We are constantly improving the portfolio of installed software. The software page currently lists only a few of the installed software packages. Please log in to Kebnekaise (regular: kebnekaise.hpc2n.umu.se or ThinLinc: kebnekaise-tl.hpc2n.umu.se) for a list of all available software packages. Note that some packages may be different on the AMD nodes. Login to kebnekaise-amd.hpc2n.umu.se to check the software on those.

NOTE: There is a special login node for the A100 GPUs that is AMD Zen3 (AMD EPYC 7313) and with 1 A100 card: kebnekaise-amd (for ThinLinc: kebnekaise-amd-tl). It is also running Ubuntu Jammy 22.04 like the A100 nodes, and is recommended for when you are using the A100 GPUs, or any other node running Ubuntu Jammy, as it allows you to see which software is available on them.

Kebnekaise is using SLURM for job management and scheduling.

Do note that running all 28 cores with lots of AVX (on the normal CPUs) will limit the clock to at absolute maximum 2.9 GHz per core, and probably no more than 2.5.

The AVX clock frequency != the rest of the CPUs clock frequency and has a lower starting point and lower max boost.

Detailed node Info

Compute nodes, Skylake-SP

Architecture is Intel Xeon Gold 6132 (Skylake-SP).

Each core has:

The memory is shared in the whole node, but physically 96 GB is placed on each NUMA island. The memory controller on each NUMA node has 6 channels.

The Intel Xeon Gold 6132 has two AVX-512 FMA units per core.

Some more information can be found here and here.

Thus it is possible to run 32 double precision or 64 single precision floating point operations per second per clock cycle within the 512-bit vectors, as well as eight 64-bit and sixteen 32-bit integers, with up to two 512-bit fused-multiply add (FMA) units.

Compute nodes, AMD Zen3

Architecture is AMD Zen3 (AMD EPYC 7763 64-Core)

 - The CPU-only node have 2 CPU sockets with 64 cores each and 1TB of memory (or 8020MB/core usable)

Compute nodes, AMD Zen4

Architecture is AMD Zen4 (AMD EPYC 9754 128-Core)

- The CPU-only nodes have 2 CPU sockets with 128 cores each and 768 GB of memory (or 2516MB/core usable)

Large memory nodes

There are 18 cores on each of the 4 NUMA islands. The cores on each NUMA island share 768 GB memory, but have access to the full 3072 GB on the node. The memory controller on each NUMA island has 4 channels.

GPU nodes, V100

We have 10 nodes with NVidia V100 (Volta) GPUs. Each CPU core is identical to the cores in the Skylake compute nodes and in addition to that the nodes each have 

• 2 V100 GPUs each with
  • 2x5120(CUDA) cores
  • 2x640(Tensor) cores

One V100 GPU is located on each NUMA island.

GPU nodes, A100

 - The GPU enabled nodes (AMD EPYC 7413 24-Core) have 2 CPU sockets with 24 cores each, i.e. 48 in total and 512GB memory (or 10600MB/core usable)

References and further information

• Intel Xeon Processor E5-2690 v4 (Intel homepage)
• Xeon E5-2690 v4
• Intel Xeon E5-2690 v4 at CPU World
• Information about Intel Xeon E5-2600 at thenextplatform
• Detailed specs for Intel Xeon E5-2600 at Microway
• Intel Xeon E7-8860v4
• NVidia V100

The information that were used in creating the images for the compute node and largemem node on this page are generated with the lstopo command.