Supercomputers/Systems/Magellan ALCF
USA · operational · hpc
Magellan ALCF
Also known as Magellan (Argonne)
Operated by Argonne National Laboratory at Argonne Leadership Computing Facility .
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
- 38.66 TFlop/s
- Rpeak
- 43.00 TFlop/s
- Rmax ÷ Rpeak
- 89.9%
- Cores
- 4,032
- Power
- 132 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 iDataPlex, 504 dual-socket compute servers | IBM | 504 reported | Reported | top500.orgscience.osti.gov |
| Accelerator | - 266 Nvidia 2070 GPU cards in 133 GPU servers | NVIDIA | 266 reported | Reported | science.osti.gov |
| CPU | - Intel Nehalem dual quad-core 2.66 GHz (Xeon X5550 per TOP500), 2 per server | Intel | 1,008 estimated | Reported | top500.orgscience.osti.gov |
| Interconnect | Mellanox InfiniBand QDR Interconnect Fat tree · 40 Gb/s per 4x port QDR InfiniBand | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | top500.orgscience.osti.gov |
The read
Argonne half of the DOE Magellan cloud testbed, built to test whether private clouds suit mid-range DOE science workloads; the NERSC half is magellan-nersc. The March 2011 ASCAC status slides describe 504 IBM iDataPlex servers (dual quad-core 2.66 GHz Nehalem, 24 GB RAM, QDR InfiniBand), 4,032 cores and about 40 TF peak, plus 200 active-storage nodes with 40 TB of SSD, 15 big-memory servers and 133 GPU servers carrying 266 Nvidia 2070 cards. Timeline in the slides: core system deployed Jan-Feb 2010, acceptance Feb 2010, cloud user access Mar 2010, Nimbus in Jun 2010, OpenStack in Dec 2010, and the Magellan cloud projects scheduled to end Sep 2011; the project was funded under ARRA. The TOP500 entry covers only the 4,032 compute cores (the same count as the 504 servers): Rmax 38.66 TF at rank 286 in Nov 2010 (31.31 TF, rank 292, in Jun 2010) against 43.0 TF Rpeak, versus the slides' rounded 40 TF peak. The GPU, active-storage and big-memory servers are not in the Linpack figure.
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
Argonne Leadership Computing Facility
- Location
- Lemont, IL, USA
- Cooling
- Direct liquid cooling (warm water)
- Commissioned
- 2006
Change history
- 2026-10-17 Magellan ALCF 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 |
|---|---|---|
| science.osti.gov | does not name this system | - |
| top500.org | names system and part | Mellanox InfiniBand QDR |
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.