Supercomputers/Systems/Snellius (Phase 1 GPU)
Netherlands · operational · hpc
Snellius (Phase 1 GPU)
Also known as Snellius Phase 1 GPU; Snellius A100 nodes
Operated by SURF at SURF (Amsterdam Science Park) .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1581 systems have a readable citation that names them, and 64 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
- 3.60 PFlop/s
- Rpeak
- 6.01 PFlop/s
- Rmax ÷ Rpeak
- 59.9%
- Cores
- 12,960
- Power
- 205.35 kW
- Per watt
- 17.5 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.
How it compares
Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of Snellius (Phase 1 GPU), and those in Netherlands. Computed from our tables; a metric is ranked only when at least eight peers report it.
| Metric | Value | Against its era | Against its country |
|---|---|---|---|
| Rmax | 3.60 PFlop/s | #133 of 210 · bottom 38% | #6 of 11 · top 50% |
| Energy efficiency | 17.5 GF/W | #37 of 109 · top 33% | #4 of 8 · top 43% |
| Rmax ÷ Rpeak | 59.9 % | #176 of 207 · bottom 16% | #9 of 11 · bottom 21% |
| Power draw | 205 kW | #90 of 112 · bottom 21% | #7 of 9 · bottom 26% |
Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.
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 | - Lenovo ThinkSystem SD650-N V2 | Lenovo | - | Reported | top500.org |
| Accelerator | NVIDIA A100 SXM4 40GB Accelerators CUDA (Ampere GA100) · 6,912C · 1.41 GHz · 7 nm · 400 W 72 GPU nodes, each with an NVIDIA HGX A100 baseboard of 4 GPUs (288 A100, 40 GiB HBM each) | NVIDIA | 288 reported | Reported | visualization.surf.nltop500.org |
| CPU | - Intel Xeon Platinum 8360Y 36C 2.4GHz | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds Infiniband HDR | NVIDIA | - | Reported | top500.org |
| Operating system | CentOS Operating systems Linux distribution, RHEL-compatible rebuild · Open source, community-maintained CentOS Linux 8.4 | The CentOS Project | - | Reported | top500.org |
The read
The A100 GPU nodes of SURF's Snellius, which went live in December 2021 as part of Phase 1; SURF says Phase 1 alone had almost five times the performance of its predecessor Cartesius (cartesius). SURF's virtual tour counts 72 Phase 1 GPU nodes, each with an NVIDIA HGX A100 baseboard of four GPUs, for 288 A100 with 40 GiB of HBM each; it lists 5.6 PFlop/s theoretical for the Phase 1 GPUs, against 6.01 PFlop/s Rpeak on TOP500. TOP500 lists Lenovo ThinkSystem SD650-N V2 nodes with Xeon Platinum 8360Y 36-core 2.4 GHz, InfiniBand HDR and CentOS 8.4: Rmax 3.60 PFlop/s and 205.35 kW at measurement level 1. Its 12,960 'cores' equal 180 per node on 72 nodes, well above the 72 CPU cores per node, so TOP500's count apparently includes accelerator cores (inferred, not stated). TOP500 rank history is uneven, 403rd (November 2021), 439th, 254th, 176th (June 2023, its best), 210th, 240th (June 2024, the entry recorded here), 279th, 319th, 348th and 376th (June 2026). Sister rows: snellius-phase2-cpu and snellius-phase3-gpu.
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
SURF (Amsterdam Science Park)
- Location
- Amsterdam, Netherlands
- Commissioned
- 2021
Change history
- 2026-11-01 Added Snellius (Phase 1 GPU)
Nearby in scale
The machines closest to Snellius (Phase 1 GPU) by Rmax, smaller to larger.
Related systems
Same site
- Snellius (Phase 2 CPU) · Netherlands, operational, 2023
- Cartesius · Netherlands, decommissioned, 2013
- Huygens · Netherlands, operational, 2008
- Snellius (Phase 3 GPU) · Netherlands, operational
Same accelerator
- ABCI 2.0 · Japan, operational, 2021
- AlphaCentauri · Germany, operational, 2021
- Athena · Poland, operational, 2021
- Baskerville · United Kingdom, operational, 2021
- Berzelius · Sweden, operational, 2021
- Earth Simulator 4 · Japan, operational, 2021
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.