Supercomputers/Systems/PACS-CS
Japan · decommissioned · hpc
PACS-CS
Also known as PACS-CS (7th generation PACS/PAX); Hitachi/Fujitsu PC cluster
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
- 10.35 TFlop/s
- Rpeak
- 14.34 TFlop/s
- Rmax ÷ Rpeak
- 72.2%
- Cores
- 2,560
- Power
- 545 kW
- Per watt
- 0.0 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 | - Hitachi was the main contractor for hardware and software | Hitachi | - | Reported | ccs.tsukuba.ac.jptop500.org |
| CPU | - Intel low-voltage Xeon 2.8 GHz, one per node, 2,560 nodes | Intel | - | Reported | top500.org |
| Interconnect | - three-dimensional hyper-crossbar network of trunked Gigabit Ethernet, network software by Fujitsu | Fujitsu | - | Reported | ccs.tsukuba.ac.jp |
The read
The seventh generation of Tsukuba's PACS/PAX series and the successor to CP-PACS: a PC cluster of 2,560 single-CPU nodes on a three-dimensional hyper-crossbar network built from trunked Gigabit Ethernet, designed around memory bandwidth. The Center for Computational Sciences says the full system started operation on 1 July 2006, was formally commissioned on 3 September 2007, achieved 10.35 TFlop/s in Linpack, ranked 34th on the June 2006 TOP500 list, was shut down in September 2011 and was followed by HA-PACS. Hitachi was the main contractor and Fujitsu developed network software. TOP500 independently records 2,560 cores, 10.35 TFlop/s Rmax, 14.34 TFlop/s Rpeak, 545 kW and the 34th place (https://www.top500.org/system/174675/).
Timeline
- 2006-06 List debut at #34 (June 2006) source
- 2006-07-01 First operational source
- 2011-09 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 PACS-CS
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 only | - |
| top500.org | 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.