Supercomputers/Systems/AIST Super Cluster P-32
Japan · operational · hpc
AIST Super Cluster P-32
Also known as AIST Super Cluster (P-32 subsystem)
Operated by AIST .
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
- 6.16 TFlop/s
- Rpeak
- 8.80 TFlop/s
- Rmax ÷ Rpeak
- 69.9%
- Cores
- 2,200
- 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 | - IBM-built AIST Super Cluster P-32 | IBM | - | Reported | top500.orgaist.go.jp |
| CPU | - AMD Opteron 2.0 GHz, 2,200 processors | AMD | 2,200 reported | Reported | top500.org |
| Interconnect | - Myrinet | Myricom | - | Reported | top500.org |
The read
P-32 subsystem of the AIST Super Cluster at the Grid Technology Research Center of AIST. TOP500 lists IBM, 2,200 Opteron 2.0 GHz cores on Myrinet under Linux, Rmax 6,155 and Rpeak 8,800 GFlop/s, ranked 19th in June 2004 and 28th in November 2004, staying on the list until November 2007 (rank 458, https://www.top500.org/system/173309/). AIST's June 2004 announcement says the whole AIST Super Cluster was installed at the end of March 2004 and has a combined 14.6 TFlop/s peak, 9.6 TB main memory and Linux, built from three subsystems (P-32, M-64 and F-32) joined by 10 Gigabit Ethernet so they can run separately or together; M-64 emphasises large memory per computer and F-32 independent parallel work. The announcement gives no node or processor counts, so the Opteron, Myrinet and 2,200-core figures rest on TOP500, and the 8.8 TFlop/s Rpeak is only the P-32 share of the 14.6 TFlop/s total.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-17 Added AIST Super Cluster P-32
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 |
|---|---|---|
| aist.go.jp | names system only | - |
| top500.org | names system only | - |
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.