Supercomputers/Systems/T2K Tsukuba
Japan · decommissioned · hpc
T2K Tsukuba
Also known as T2K-Tsukuba; T2K Open Supercomputer (Tsukuba)
Operated by University of Tsukuba at Tsukuba Center for Computational Sciences .
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
- 77.28 TFlop/s
- Rpeak
- 95.38 TFlop/s
- Rmax ÷ Rpeak
- 81.0%
- Cores
- 10,368
- Power
- 671 kW
- Per watt
- 0.1 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 | - Cray Japan and Sumisho Computer Systems built the system on Appro Xtreme-X3 servers | Appro HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | hothardware.comtop500.org |
| CPU | - AMD quad-core Opteron 2.3 GHz, four sockets per node, 648 nodes | AMD | - | Reported | ccs.tsukuba.ac.jptop500.org |
| Interconnect | Mellanox InfiniBand DDR Interconnect Fat tree · 20 Gb/s per 4x port · About 2 microseconds quad-rail 4x DDR InfiniBand fat tree, 5.28 TB/s bisection bandwidth | Mellanox NVDANASDAQ
rolls up to NVIDIA | - | Reported | ccs.tsukuba.ac.jptop500.org |
The read
Tsukuba's member of the T2K Open Supercomputer Alliance, the 2008 trio of x86 clusters specified jointly by the Universities of Tsukuba, Tokyo and Kyoto. The Center for Computational Sciences describes 648 nodes of four quad-core 2.3 GHz AMD Opterons with 32 GB each, quad-rail DDR InfiniBand on a fat tree, 10,368 cores, 95.4 TFlop/s peak and 20.7 TB of memory, and says it was shut down in February 2014. Trade press reported in June 2008 that the three T2K systems entered operation, with the Tsukuba machine built by Cray Japan and Sumisho Computer Systems on Appro Xtreme-X servers (https://hothardware.com/news/t2k-open-supercomputer-now-in-operation). TOP500 lists 77.28 TFlop/s Rmax, 95.38 TFlop/s Rpeak and 671 kW for the 10,368-core configuration (https://www.top500.org/system/176215/); the first listing of the machine, with 10,000 cores and 76.46 TFlop/s, is https://www.top500.org/system/175783/, ranked 21st in June 2008 (the centre's page says 20th). Rmax rests on TOP500 alone, so the tier is reported.
Timeline
- 2008-06 List debut at #21 (June 2008) source
- 2008-06 First operational source
- 2014-02 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
Tsukuba Center for Computational Sciences
- Location
- Tsukuba, Japan
- Commissioned
- 1996
Change history
- 2026-10-09 Added T2K Tsukuba
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 |
|---|---|---|
| ccs.tsukuba.ac.jp | names system and part | Mellanox InfiniBand DDR |
| top500.org | names system and part | Mellanox InfiniBand DDR |
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.