Supercomputers/Systems/ARIS
Greece · operational · hpc
ARIS
Also known as GRNET ARIS; Greek national supercomputer
Operated by GRNET at ARIS Data Centre (Athens) .
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
- 170.00 TFlop/s
- Rpeak
- 190.00 TFlop/s
- Rmax ÷ Rpeak
- 89.5%
- Cores
- 8,520
- Power
- 154 kW
- Per watt
- 1.1 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 | - IBM NeXtScale nx360M5 | IBM | - | Reported | top500.org |
| Accelerator | - 44 GPU nodes x 2 NVIDIA Tesla K40m (quantity derived as 44 x 2) | NVIDIA | 88 reported | Reported | doc.aris.grnet.gr |
| Accelerator | - 18 Phi nodes x 2 Intel Xeon Phi 7120P (quantity derived as 18 x 2) | Intel | 36 reported | Reported | doc.aris.grnet.gr |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 1 ML node with 8 NVIDIA Volta V100 GPUs | NVIDIA | 8 reported | Reported | doc.aris.grnet.gr |
| CPU | - 426 thin nodes with two 10-core Intel Xeon E5-2680v2 at 2.8 GHz (8,520 cores, quantity is nodes) | Intel | 426 reported | Reported | top500.orgdoc.aris.grnet.gr |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond InfiniBand FDR (56 Gb/s) fat tree; TOP500 lists FDR14 | NVIDIA | - | Reported | doc.aris.grnet.grtop500.org |
The read
Greece's national HPC system run by GRNET, an IBM NeXtScale cluster that debuted at 468th on the June 2015 TOP500 with 8,520 cores (426 thin nodes with two 10-core Xeon E5-2680v2), 0.17 PFlop/s Rmax, 0.19 PFlop/s Rpeak and 154 kW on InfiniBand FDR14. The ARIS documentation lists 532 nodes in total: 426 thin nodes, 44 GPU nodes with two Tesla K40m each, 18 Phi nodes with two Xeon Phi 7120P each, 44 fat nodes with 512 GB and one ML node with eight V100 GPUs, all on FDR InfiniBand (56 Gb/s fat tree) with GPFS storage. The TOP500 figures and core count cover the thin-node partition only. GRNET later ran an equipment upgrade project from May 2022 to June 2025 that kept capacity while cutting energy use by more than half, and describes ARIS as in use since 2015 at the data centre of the Ministry of Education building. The newer DAEDALUS system is a separate row.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Change history
- 2026-10-04 Added ARIS, GRNET
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 |
|---|---|---|
| doc.aris.grnet.gr | names system, part and count | NVIDIA Tesla V100, Mellanox InfiniBand FDR · counts: 8 |
| top500.org | names system, part and count | Mellanox InfiniBand FDR · counts: 8 |
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.