Supercomputers/Systems/Grete
Germany · operational · ai training
Grete
Also known as NHR-NORD@Goettingen Grete
Operated by GWDG .
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
- -
- Rpeak
- 5.46 PFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- -
- 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 | - MEGWARE GPU server system | MEGWARE | - | Reported | gwdg.de |
| Accelerator | - 36 initial nodes with four NVIDIA A100 40 GB each (quantity derived as 36 x 4) | NVIDIA | 144 reported | Reported | gwdg.degwdg.de |
| CPU | AMD EPYC 7513 (Milan) CPUs x86-64 · 32C · 2.6 GHz · 7 nm · 200 W 36 initial nodes with two AMD EPYC 7513 (quantity is nodes) | AMD | 36 reported | Verified | gwdg.degwdg.de |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds InfiniBand HDR, 2x 200 Gb/s per node on the A100 nodes | NVIDIA | - | Verified | gwdg.degwdg.de |
The read
GWDG's GPU cluster for the NHR-NORD@Goettingen centre, named after mathematician Grete Hermann. GWDG's 2022 press release announces the initial MEGWARE system as 36 nodes with two AMD EPYC 7513 CPUs and four NVIDIA A100 40 GB each on 2x200 Gb/s InfiniBand HDR, operational on 23 November 2022, ranked 470th on the TOP500 (1.83 PFlop/s) and 12th on the Green500. The current system page lists 106 nodes (3 Xeon Gold 6148 plus V100 nodes and 103 AMD EPYC plus A100 40/80 GB nodes), 6,840 CPU cores, 27.5 TB of GPU memory, 8 racks with direct liquid cooling and 5.46 PFlop/s stated peak (the page words it as LINPACK peak and cites 142nd on the November 2023 TOP500). A 2024 Grete Phase 3 with NVIDIA H100 is a separate TOP500 entry not included here. The TOP500 row is the November 2022 entry of the initial system.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-02 Added Grete (GWDG)
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 |
|---|---|---|
| gwdg.de | names system, part and count | AMD EPYC 7513 (Milan), NVIDIA InfiniBand HDR · counts: 36 |
| gwdg.de | names system and part | AMD EPYC 7513 (Milan), NVIDIA InfiniBand HDR |
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.