Supercomputers/Systems/Anselm
Czechia · decommissioned · hpc
Anselm
Also known as IT4Innovations Anselm; Anselm Bull cluster
Operated by IT4Innovations at IT4Innovations, Ostrava .
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
- 73.00 TFlop/s
- Rpeak
- 94.00 TFlop/s
- Rmax ÷ Rpeak
- 77.7%
- Cores
- -
- Power
- 77 kW
- Per watt
- 0.9 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 | - Bull supplied the system (built by Atos) | Bull | - | Reported | ostrava.czdocs-old.e-infra.cz |
| Accelerator | - NVIDIA Tesla Kepler K20m GPUs, 23 nodes | NVIDIA | 23 reported | Reported | docs-old.e-infra.cz |
| CPU | - Intel Sandy Bridge E5-2665 8-core 2.4 GHz, 180 compute nodes | Intel | 180 reported | Reported | docs-old.e-infra.cz |
| Interconnect | Mellanox InfiniBand QDR Interconnect Fat tree · 40 Gb/s per 4x port InfiniBand QDR fully non-blocking fat tree | NVIDIA | - | Reported | docs-old.e-infra.cz |
The read
The first supercomputer of the Czech IT4Innovations centre at VSB-Technical University of Ostrava, built by Bull (now Atos) and christened on 24 May 2013. The archive page says it ran from the second quarter of 2013 to the first quarter of 2021, with 209 compute nodes, 94 TFlop/s theoretical peak, 77 kW power draw and an original cost of 90 million CZK. The hardware overview lists 180 regular nodes with two Sandy Bridge E5-2665 8-core CPUs, 23 nodes with NVIDIA Tesla K20m GPUs, 4 nodes with Xeon Phi 5110P and 2 fat nodes with 512 GB, on a non-blocking InfiniBand QDR fat tree with Lustre /home (360 TB) and /scratch (146 TB), 15.1 TB RAM in total and 73 TFlop/s maximum Linpack. It was retired into an exhibition at Lower Vitkovice. Salomon and Karolina are recorded separately.
Site & facility
IT4Innovations, Ostrava
- Location
- Ostrava, Czechia
- Cooling
- Direct liquid cooling
- Commissioned
- 2013
Change history
- 2026-10-09 Added Anselm
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 |
|---|---|---|
| docs-old.e-infra.cz | names system only | - |
| ostrava.cz | names system only | - |
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.