Supercomputers/Systems/Ares
Poland · operational · hpc
Ares
Operated by Cyfronet AGH at Cyfronet AGH, Krakow .
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: 1092 of 1153 systems have a readable citation that names them, and 40 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
- 2.34 PFlop/s
- Rpeak
- 3.51 PFlop/s
- Rmax ÷ Rpeak
- 66.7%
- Cores
- 37,824
- Power
- 486.85 kW
- Per watt
- 4.8 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 | - Huawei servers (TOP500 manufacturer), 532 + 256 CPU servers and 9 GPU servers | Huawei | - | Reported | top500.orgcyfronet.pl |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 72 NVIDIA Tesla V100 cards | NVIDIA | 72 reported | Reported | cyfronet.pl |
| CPU | - Intel Xeon Platinum 8268 24C 2.9GHz (TOP500); Cyfronet cites Xeon Platinum and Gold servers | Intel | - | Reported | top500.orgcyfronet.pl |
| Interconnect | Mellanox InfiniBand EDR Interconnect Fat tree · 100 Gb/s per 4x port · About 0.9 microseconds InfiniBand EDR | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | top500.org |
| Operating system | CentOS Operating systems Linux distribution, RHEL-compatible rebuild · Open source, community-maintained CentOS | The CentOS Project | - | Reported | top500.org |
The read
Cyfronet AGH's Krakow CPU cluster with a small GPU partition: 532 nodes with 192 GB plus 256 nodes with 384 GB of Intel Xeon Platinum and Gold CPUs (37,824 cores), and nine GPU servers carrying 72 NVIDIA Tesla V100 cards, on InfiniBand EDR with over 11 PB of storage. TOP500 credits Huawei as manufacturer and lists a Xeon Platinum 8268 24C 2.9GHz part. Cyfronet's own page quotes over 4 PFlop/s theoretical (3.5 CPU, 0.5 GPU), and TOP500 gives 3.51 PFlop/s Rpeak for the CPU-only Linpack run. TOP500 first listed it in 2021 and its last listing was 487th in November 2024. Cyfronet still shows the system as operational. TOP500 is the only source for Rmax and core count, so the tier is reported.
Timeline
- 2021 First operational source
Site & facility
Change history
- 2026-10-01 Added Ares (Cyfronet AGH)
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 |
|---|---|---|
| top500.org | names system and part | Mellanox InfiniBand EDR, CentOS |
| cyfronet.pl | names system, part and count | NVIDIA Tesla V100, Mellanox InfiniBand EDR · counts: 72 |
Checked 2026-10-09 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.