Japan · operational · mixed

TSUBAME4.0

Operated by Institute of Science Tokyo at Institute of Science Tokyo, Ookayama .

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: 39.62 PFlop/s
Rpeak: 61.60 PFlop/s
Rmax ÷ Rpeak: 64.3%
Cores: 172,800
Power: 815.82 kW
Per watt: 48.6 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	- HPE Cray XD665	Hewlett Packard Enterprise HPENYSE	-	Reported	top500.org
Accelerator	NVIDIA H100 SXM5 94GB Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W Nvidia H100 SXM5 94Gb	NVIDIA NVDANASDAQ	-	Verified	top500.org t4.cii.isct.ac.jp
CPU	AMD EPYC 9004 (Genoa) CPUs x86-64 · 96C · 2.4 GHz · TSMC N5 · 360 W AMD EPYC 9654 96C 3.55GHz	AMD AMDNASDAQ	-	Verified	top500.org t4.cii.isct.ac.jp
Interconnect	NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds Infiniband NDR200	NVIDIA NVDANASDAQ	-	Verified	top500.org t4.cii.isct.ac.jp
Operating system	Red Hat Enterprise Linux Operating systems Linux distribution · Open source, commercial support subscription Redhat Linux	Red Hat IBMNYSE rolls up to IBM	-	Reported	top500.org

AI datacenter metrics

Figures beyond an accelerator count, each with who stated it. What can and cannot be compared across AI datacenters.

Metric	Value	As of	Basis	Note	Source
Accelerators	960 accelerators	2026-10	Operator stated	ISCT hardware page: 240 Cray XD665 nodes x 4 NVIDIA H100 SXM5 (product computed).	t4.cii.isct.ac.jp

The read

Successor to the now-decommissioned `tsubame-3` at the same institution (renamed from Tokyo Institute of Technology to Institute of Science Tokyo in 2024). ISCT's own hardware page independently confirms the EPYC 9654 / H100 SXM5 94GB / InfiniBand NDR200 node design in exact detail, including the otherwise-uncommon 94GB H100 memory capacity, but does not itself state Rmax/Rpeak/power/core totals, so those stay TOP500-sourced.

Timeline

2024 First operational source
2024-06 List debut at #25 (June 2024) 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

Institute of Science Tokyo, Ookayama

Location: Tokyo, Japan
Cooling: Direct liquid cooling
Commissioned: 2006

isct.ac.jp

Change history

2026-09-22 Gained a Green500 rank pointer (#25 on the 2024-06 list)

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
t4.cii.isct.ac.jp	names system and part	AMD EPYC 9004 (Genoa), NVIDIA H100 SXM5 94GB, NVIDIA InfiniBand NDR (Quantum-2)
top500.org	names system and part	AMD EPYC 9004 (Genoa), NVIDIA H100 SXM5 94GB, NVIDIA InfiniBand NDR (Quantum-2)

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.