Supercomputers/Systems/Hydra (MPCDF)
Germany · operational · hpc
Hydra (MPCDF)
Also known as Max Planck supercomputer Hydra; RZG Hydra
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: 1263 of 1345 systems have a readable citation that names them, and 56 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
- 1.28 PFlop/s
- Rpeak
- 1.46 PFlop/s
- Rmax ÷ Rpeak
- 87.7%
- Cores
- 65,320
- Power
- 1.26 MW
- Per watt
- 1.0 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 | - IBM iDataPlex DX360M4 | IBM | - | Reported | top500.orgmpg.de |
| Accelerator | NVIDIA Tesla K20X Accelerators CUDA (Kepler GK110) · 2,688C · 0.732 GHz · 28 nm · 235 W NVIDIA Kepler K20X GPUs, 676 cards in 338 nodes | NVIDIA | 676 reported | Reported | mpg.detop500.org |
| CPU | - Intel Xeon E5-2680v2 10-core 2.8 GHz Ivy Bridge | Intel | - | Reported | top500.orgmpg.de |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Mellanox InfiniBand FDR14 | NVIDIA | - | Reported | mpg.detop500.org |
The read
The Max Planck Society's main supercomputer at the Rechenzentrum Garching (now MPCDF), put into operation in October 2013 according to the society's 2013 research report. It combined an Ivy Bridge partition (Xeon E5-2680v2, 10 cores at 2.8 GHz) with a Sandy Bridge phase from 2012, 676 NVIDIA Kepler K20X GPUs in 338 nodes and 24 Xeon Phi coprocessors in 12 nodes, all on InfiniBand FDR14, for about 2.6 PFlop/s aggregate peak. TOP500 lists the main iDataPlex DX360M4 Ivy Bridge part (system 178197) at 65,320 cores, 1,283.31 TFlop/s Rmax, 1,463.17 TFlop/s Rpeak and 1.26 MW, 28th in June 2014 and on eleven lists to November 2018; the GPU part (178199) and the Sandy Bridge part (177741) are separate entries. The figures here cover that main partition only. The status date is the last list appearance.
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
- 2026-10-09 Added Hydra (MPCDF)
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 |
|---|---|---|
| mpg.de | names system, part and count | Mellanox InfiniBand FDR · counts: 676 |
| top500.org | does not name this system | Mellanox InfiniBand FDR |
Checked 2026-10-13 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.