Supercomputers/Systems/QPACE3
Germany · operational · hpc
QPACE3
Also known as QPACE 3
Operated by Universitaet Regensburg .
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
- 447.09 TFlop/s
- Rpeak
- 766.77 TFlop/s
- Rmax ÷ Rpeak
- 58.3%
- Cores
- 18,432
- Power
- 77.35 kW
- Per watt
- 5.8 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 | - Fujitsu PRIMERGY CX1640 M1 servers | Fujitsu | - | Reported | top500.org |
| CPU | - Intel Xeon Phi 7210 64C 1.3GHz (Knights Landing), 352 nodes in phase 1 and 320 nodes in phase 2 | Intel | - | Verified | top500.orgarxiv.org |
| Interconnect | Omni-Path Architecture Interconnect Fat tree · 100 Gb/s per port · About 1 microsecond Intel Omni-Path fabric | Cornelis Networks | - | Verified | top500.orgarxiv.org |
The read
Lattice QCD cluster of the Regensburg lattice group (RQCD), successor to the Knights Corner based QPACE 2. The Regensburg paper on DD-alphaAMG describes QPACE 3 as installed in two phases: phase 1 of 352 nodes, each with a 64-core Xeon Phi 7210 (Knights Landing) at 1.3GHz, 16GB MCDRAM and 48GB DDR4, and phase 2 of 320 nodes with 96GB DDR4, on a 100 Gbit/s Intel Omni-Path fabric; it reports the system ranked 5th on the November 2016 Green500. TOP500 lists it under the site Fujitsu Technology Solutions GmbH as a PRIMERGY CX1640 M1 cluster of 18,432 Xeon Phi 7210 cores, Rmax 447.09 and Rpeak 766.77 TFlop/s at 77.35 kW, ranked 376th in November 2016 (the benchmarked partition is smaller than the full installation).
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
- 2026-10-09 Added QPACE3
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 |
|---|---|---|
| arxiv.org | names system and part | Omni-Path Architecture |
| top500.org | names system and part | Omni-Path Architecture |
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.