Supercomputers/Systems/Hábrók
Netherlands · operational · hpc
Hábrók
Also known as Habrok
Operated by University of Groningen .
Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1263 of 1345 systems have a readable citation that names them, and 56 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- -
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- -
- Power
- -
- Per watt
- - GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - Dell Technologies delivered the cluster | Dell Technologies | - | Reported | wiki.hpc.rug.nl |
| Accelerator | - NVIDIA A100 40 GB, four per node, 6 GPU nodes | NVIDIA | 24 reported | Reported | wiki.hpc.rug.nl |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W NVIDIA V100 32 GB, two per node, 12 older GPU nodes | NVIDIA | 24 reported | Reported | wiki.hpc.rug.nl |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W AMD EPYC 7763 64-core 2.45 GHz, two per node, 132 standard and multi-node compute nodes | AMD | - | Reported | wiki.hpc.rug.nl |
| CPU | Intel Xeon Platinum 8358 CPUs x86-64 · 32C · 2.6 GHz · 10 nm · 250 W Intel Xeon Platinum 8358 32-core 2.6 GHz, two per node, 6 A100 GPU nodes | Intel | - | Reported | wiki.hpc.rug.nl |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond 100 Gbps Omni-Path network for 48 compute nodes and 6 GPU nodes | Intel | - | Reported | wiki.hpc.rug.nl |
The read
Hábrók, the University of Groningen's shared HPC cluster, successor to Peregrine and Millipede. The centre's cluster description says it was delivered by Dell and lists 84 standard and 48 multi-node compute nodes with two AMD EPYC 7763 (128 cores, 512 GB) each, 4 big-memory nodes with two Xeon Platinum 8380 and 4 TB, 6 GPU nodes with two Xeon Platinum 8358 and four NVIDIA A100 40 GB cards, 3 nodes with eight NVIDIA RTX Pro 6000 GPUs, 12 older Fujitsu-supplied nodes with two V100 each, H100 and L40S nodes, a 100 Gbps Omni-Path network for 48 compute and 6 GPU nodes, and 2.5 PB of Lustre scratch storage. The maintenance changelog shows it in active service in October 2026. No LINPACK result or start date is published on those pages.
Timeline
No dated events recorded.
Change history
- 2026-10-09 Added Hábrók
Source check
We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.
| Cited source | Result | Found on the page |
|---|---|---|
| wiki.hpc.rug.nl | names system, part and count | AMD EPYC 7763 (Milan), Omni-Path Architecture · counts: 24 |
| wiki.hpc.rug.nl | names system, part and count | AMD EPYC 7763 (Milan), Intel Xeon Platinum 8358, Omni-Path Architecture · counts: 24 |
Checked 2026-10-13 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
Inside a node
Latency-optimised versus throughput-optimised processors, NUMA, and the end of the host-to-device copy.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.