Supercomputers/Systems/Pleiades
USA · decommissioned · hpc
Pleiades
Operated by NASA at NASA Advanced Supercomputing Facility .
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
- 5.95 PFlop/s
- Rpeak
- 7.09 PFlop/s
- Rmax ÷ Rpeak
- 83.9%
- Cores
- 241,108
- Power
- 4.41 MW
- Per watt
- 1.4 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 | - SGI ICE X | SGI HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | nas.nasa.govnas.nasa.gov |
| CPU | Intel Xeon E5-2680 v4 CPUs x86-64 · 14C · 2.4 GHz · 14 nm · 120 W Xeon E5-2680v4 14C 2.4GHz | Intel | - | Reported | nas.nasa.govnas.nasa.gov |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Infiniband FDR | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | nas.nasa.govnas.nasa.gov |
| Storage | Lustre Storage Parallel filesystem, separate metadata and object servers Lustre-based filesystems | SGI HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | nas.nasa.gov |
| Cooling | Air cooling (CRAC/CRAH) Cooling Room-level air handling · Practical ceiling around 15-20 kW per rack SGI ICE X: air cooled with in-rack water coils | SGI HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | nas.nasa.govnas.nasa.gov |
| Operating system | Tri-Lab Operating System Stack Operating systems Linux distribution, RHEL-derived, HPC-tuned · Open source, LLNL-maintained Operating System: Tri-Lab Operating System Stack (TOSS) | Lawrence Livermore National Laboratory | - | Reported | nas.nasa.gov |
The read
Almost every system on this site was built, run and retired as a unit. Pleiades was upgraded in place for more than a decade, with successive processor generations added and older racks removed while the machine kept running and kept its name. That makes it genuinely awkward for a dataset built around systems as fixed objects, and it is the reason our schema carries era bounds on component edges at all. It is also an SGI machine that outlived SGI: the supplier at build was acquired by HPE in 2016, and the system carried on. NASA's HECC page for the system is headed as decommissioned December 2025 and gives a final configuration of 153 racks, 10,728 nodes and 232,416 cores at 6.91 PFlop/s peak and 5.95 PFlop/s LINPACK, built from Broadwell, Haswell, Ivy Bridge and Sandy Bridge nodes. A NAS knowledge-base notice (published 25 November 2025, last updated 7 January 2026) states that all Pleiades nodes have been turned off, that the Electra-based Broadwell nodes remain, and that the Pleiades front ends stay in service as the front ends of Aitken, Electra and Endeavour. The recorded date 2026-01 is the month the notice reports the shutdown complete, while the system page says December 2025, so the exact month of final shutdown is uncertain. TOP500 lists the 241,108-core run (5.95 PFlop/s Rmax, 7.11 PFlop/s Rpeak, 4,407 kW) and shows its last appearance on the November 2025 list at rank 249, with no June 2026 entry. The recorded cores and peak are the TOP500 run, which is larger than NASA's final configuration.
Timeline
- 2008 Announced source
- 2008-11 List debut at #3 (November 2008) source
- 2008-11 First operational source
- 2026-01 Decommissioned source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
NASA Advanced Supercomputing Facility
- Location
- Moffett Field, CA, USA
- Cooling
- Air and liquid hybrid
- Commissioned
- 1987
Change history
- 2026-10-02 Marked Pleiades decommissioned
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 |
|---|---|---|
| nas.nasa.gov | names system and part | Intel Xeon E5-2680 v4, Mellanox InfiniBand FDR, Tri-Lab Operating System Stack, Lustre |
| nas.nasa.gov | names system only | - |
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.