Supercomputers/Systems/AOBA-S
Japan · operational · hpc
AOBA-S
Operated by Cyberscience Center, Tohoku University at Tohoku University Cyberscience Center .
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: 1092 of 1153 systems have a readable citation that names them, and 40 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
- 17.22 PFlop/s
- Rpeak
- 19.82 PFlop/s
- Rmax ÷ Rpeak
- 86.9%
- Cores
- 64,512
- Power
- 1.39 MW
- Per watt
- 12.4 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 | - NEC SX-Aurora TSUBASA C401-8 | NEC | - | Reported | top500.orghpci-office.jp |
| Accelerator | NEC SX-Aurora TSUBASA Vector Engine Accelerators NEC vector · 8C · 1.6 GHz · 16 nm · 300 W Vector Engine Type 30A 16C 1.6GHz, 8 per node, 504 nodes | NEC | 4,032 reported | Reported | hpci-office.jptop500.org |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds Infiniband NDR200 | NVIDIA | - | Reported | top500.orghpci-office.jp |
The read
The vector subsystem of Tohoku University's AOBA supercomputer, an NEC SX-Aurora TSUBASA system of 504 nodes, each with one vector host and eight Vector Engines, 4,032 in all, on InfiniBand. It was 49th on the November 2023 TOP500 and 126th in June 2026. The HPCI resource page gives service from August 2023, which is the start date used. It quotes 21.05 PFlop/s for the whole system, adding the host CPUs to the vector engines' 39.28 TFlop/s per node, while TOP500's 19.82 PFlop/s matches the vector engines alone, so the two are different scopes rather than a conflict. The host CPU model is not recorded because a Tohoku paper on the test cluster names an AMD EPYC and the HPCI page says Intel. TOP500 is the only source for Rmax, Rpeak and core count, so the tier is reported.
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
- 2026-09-22 Gained a Green500 rank pointer (#75 on the 2024-06 list)
- 2026-09-20 Added with a rank pointer on the edition where it ranked highest
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 |
|---|---|---|
| hpci-office.jp | names system, part and count | NEC SX-Aurora TSUBASA Vector Engine, NVIDIA InfiniBand NDR (Quantum-2) · counts: 4032 |
| top500.org | names system and part | NEC SX-Aurora TSUBASA Vector Engine, NVIDIA InfiniBand NDR (Quantum-2) |
Checked 2026-10-09 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
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.