Supercomputers/Systems/Topaze
France · operational · hpc
Topaze
Also known as CCRT Topaze
Operated by CEA at Very Large Computing Centre (TGCC) .
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
- 3.26 PFlop/s
- Rpeak
- 4.34 PFlop/s
- Rmax ÷ Rpeak
- 75.1%
- Cores
- 110,592
- Power
- -
- Per watt
- - 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 | - BullSequana XH2000 | Atos | - | Verified | top500.orgglobenewswire.com |
| Accelerator | - 48 accelerated nodes with NVIDIA A100 GPUs | NVIDIA | - | Reported | globenewswire.com |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W AMD EPYC 7763 64C 2.45GHz, 864 nodes | AMD | 864 reported | Verified | top500.orgglobenewswire.com |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds InfiniBand HDR100 | NVIDIA | - | Reported | top500.org |
| Storage | - DDN storage, about 3 PB at 280 GB/s | DDN | - | Reported | globenewswire.com |
The read
Atos BullSequana XH2000 system at the CEA TGCC computing centre in Bruyeres-le-Chatel, co-designed by Atos and CEA DAM for the CCRT industrial research partnership. The record uses the TOP500 Topaze-cpu entry (110,592 cores of AMD EPYC 7763 on InfiniBand HDR100, 3.26 PFlop/s Rmax, 4.34 PFlop/s Rpeak), which ranked 201st in June 2023 (entry recorded here) and 400th in June 2026. The June 2021 press release describes the whole machine as 864 EPYC Milan nodes, 48 A100 GPU nodes and four large-memory nodes, around 8.8 PFlop/s peak, with about 3 PB of DDN storage at 280 GB/s, shared by CCRT partners including Safran, EDF, Michelin and Thales.
Timeline
- 2021-06 Announced source
- 2021-06 First operational source
- 2023-06 List debut at #201 (June 2023) source
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 Topaze, CEA CCRT
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 | AMD EPYC 7763 (Milan), NVIDIA InfiniBand HDR |
| globenewswire.com | names system, part and count | AMD EPYC 7763 (Milan) · counts: 864 |
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.