Supercomputers/Systems/Draco
Germany · operational · hpc
Draco
Also known as MPCDF DRACO
Operated by MPCDF at MPCDF Garching .
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
- 790.70 TFlop/s
- Rpeak
- 1.04 PFlop/s
- Rmax ÷ Rpeak
- 76.4%
- Cores
- 28,128
- 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 | - Lenovo NeXtScale nx360M5 cluster | Lenovo | - | Reported | top500.orgmpcdf.mpg.de |
| Accelerator | - 106 nodes with two PNY GTX980 GPUs each | NVIDIA | 212 reported | Reported | mpcdf.mpg.de |
| CPU | Intel Xeon E5-2698 v3 CPUs x86-64 · 16C · 2.3 GHz · 22 nm · 135 W Intel Xeon E5-2698v3 16C 2.3GHz, about 880 nodes of 32 cores | Intel | 1,760 estimated | Reported | mpcdf.mpg.detop500.org |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond InfiniBand FDR14 network | NVIDIA | - | Verified | mpcdf.mpg.detop500.org |
The read
Draco is the HPC extension cluster installed in May 2016 at the Max Planck Computing and Data Facility: about 880 Haswell Xeon E5-2698 nodes of 32 cores at 2.3 GHz, 106 of them with two PNY GTX980 GPUs, on InfiniBand FDR14 with 1.5 PB of disk. TOP500 ranks it 160th in November 2016 (best rank) with 28,128 cores, 0.79 PFlop/s Rmax and 1.04 PFlop/s Rpeak, falling to 424th in June 2018. MPCDF says a January 2017 expansion with 64 Broadwell nodes bought by the Fritz Haber Institute brought the total to 30,688 cores and 1.12 PFlop/s peak, so the TOP500 figures describe the original Haswell part.
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
- 2026-10-07 Added Draco, a mid-list TOP500 system (2016-11 160)
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 | Intel Xeon E5-2698 v3, Mellanox InfiniBand FDR |
| mpcdf.mpg.de | names system and part | Intel Xeon E5-2698 v3, Mellanox InfiniBand FDR |
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.