Supercomputers/Systems/Genius (Garching)
Germany · operational · hpc
Genius (Garching)
Also known as RZG Blue Gene/P; MPI/IPP Blue Gene/P
Operated by MPCDF at MPCDF Garching .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1540 systems have a readable citation that names them, and 64 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
- 21.91 TFlop/s
- Rpeak
- 27.85 TFlop/s
- Rmax ÷ Rpeak
- 78.7%
- Cores
- 8,192
- Power
- 63 kW
- Per watt
- 0.3 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 Blue Gene/P, two racks, 2,048 compute nodes | IBM | - | Verified | docs.mpcdf.mpg.detop500.org |
| CPU | IBM PowerPC 450 850MHz CPUs PowerPC · 4C · 0.85 GHz · 90 nm IBM PowerPC 450 850 MHz, four per compute node, 8,192 cores | IBM | 8,192 verified | Verified | docs.mpcdf.mpg.detop500.org |
| Interconnect | IBM BlueGene 3D torus Interconnect 3D torus, plus separate collective and barrier networks · 175 MB/s per link per direction 3D torus plus separate global communication and synchronization networks | IBM | - | Reported | docs.mpcdf.mpg.de |
| Operating system | IBM Compute Node Kernel Operating systems Microkernel, no dynamic scheduling · Proprietary CNK/SLES 9 per TOP500; lightweight kernel on compute nodes, SLES 10 on the front end per RZG | IBM | - | Reported | top500.orgdocs.mpcdf.mpg.de |
The read
IBM Blue Gene/P at the Garching computing centre of the Max Planck Society (then RZG, now MPCDF), which its December 2007 bulletin calls the first such system at a customer site worldwide. Installed in September 2007 and in production from October 2007 after tuning, it started as two racks of 32 node boards, 2,048 four-core compute nodes (8,192 cores, 850 MHz PowerPC 450, 512 MB per core, 4 TB total) plus 16 I/O nodes on a 10 Gbit/s Ethernet link, with a 3D torus and separate collective and barrier networks, an 8-processor Power5 front end called genius.rzg.mpg.de, LoadLeveler scheduling and GPFS. The bulletin reports 21.9 TF/s LINPACK on 8,192 cores and plans a third rack in January 2008 to lift peak from 27 to 40 TF/s; TOP500 lists the 8,192-core configuration at 21.91 TFlop/s Rmax, 27.85 peak and 63 kW, ranked 41st in November 2007 and never re-listed under this entry, so the later expansion is not in the recorded figures. The Max Planck release of April 2008 states a 40 teraflop/s peak for the Blue Gene/P. Sister machine at the same centre: vip-mpcdf. Not the KU Leuven system genius-kuleuven, which only shares the name.
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
- 2026-10-17 Genius (Garching) added from its source pages
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 |
|---|---|---|
| docs.mpcdf.mpg.de | does not name this system | - |
| top500.org | names system, part and count | IBM PowerPC 450 850MHz · counts: 8192 |
Checked 2026-10-18 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.