Supercomputers/Systems/Kuma
Switzerland · operational · mixed
Kuma
Also known as EPFL Kuma
Operated by EPFL at SCITAS (EPFL Lausanne) .
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
- 12.31 PFlop/s
- Rpeak
- 21.66 PFlop/s
- Rmax ÷ Rpeak
- 56.8%
- Cores
- 49,728
- Power
- 291.92 kW
- Per watt
- 42.2 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 | - Lenovo ThinkSystem SR675 V3 | Lenovo | - | Reported | top500.org |
| Accelerator | NVIDIA H100 SXM5 94GB Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 84 nodes x 4 NVIDIA H100 SXM5 94GB (plus 20 nodes of 8 L40S) | NVIDIA | 336 reported | Verified | epfl.chtop500.org |
| CPU | AMD EPYC 9004 (Genoa) CPUs x86-64 · 96C · 2.4 GHz · TSMC N5 · 360 W AMD EPYC 9334 32C 2.7GHz | AMD | - | Reported | top500.org |
| Interconnect | - ConnectX-7 Infiniband (EPFL cites 2x200Gb HDR per node) | NVIDIA | - | Reported | top500.orgepfl.ch |
The read
The GPU cluster of EPFL's SCITAS research computing unit in Lausanne. EPFL lists 104 GPU nodes: 84 with four NVIDIA H100 SXM5 94GB GPUs each (336 H100s) and 20 with eight NVIDIA L40S GPUs each, on 200Gb InfiniBand with all-flash scratch. TOP500 lists it as a Lenovo ThinkSystem SR675 V3 system with AMD EPYC 9334 hosts, 49,728 cores, Rmax 12.31 PFlop/s, Rpeak 21.66 PFlop/s and 291.9 kW (207 kW in the optimised run); it first ranked 102nd on the November 2024 list and was 164th on June 2026. EPFL's page gives no production date.
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
- 2026-10-02 Added Kuma, EPFL's H100 cluster, 102nd on the November 2024 TOP500
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 | NVIDIA H100 SXM5 94GB, AMD EPYC 9004 (Genoa) |
| epfl.ch | names system and part | NVIDIA H100 SXM5 94GB |
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
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.