Supercomputers/Systems/Shirokane3
Japan · operational · hpc
Shirokane3
Also known as SHIROKANE3; HGC Shirokane3
Operated by University of Tokyo .
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
- 342.58 TFlop/s
- Rpeak
- 431.73 TFlop/s
- Rmax ÷ Rpeak
- 79.4%
- Cores
- 12,408
- Power
- 186.6 kW
- Per watt
- 1.8 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 | - HP Apollo 6000 Xl230/250 servers | Hewlett Packard Enterprise | - | Reported | top500.org |
| Accelerator | - Intel Xeon Phi 7120P, one in each of 24 thin nodes | Intel | 24 reported | Verified | top500.orgsupcom.hgc.jp |
| CPU | - Intel Xeon E5-2670 v3 12C 2.3GHz, 469 thin nodes, plus 6 fat nodes with Xeon E7-8857 v2 | Intel | - | Verified | top500.orgsupcom.hgc.jp |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond InfiniBand FDR | NVIDIA | - | Reported | top500.org |
| Storage | - 12 PiB home disk, 1 PiB archive disk and 21 PB tape archive | University of Tokyo | - | Reported | supcom.hgc.jp |
The read
Genome-analysis supercomputer of the Human Genome Center, Institute of Medical Science, University of Tokyo. The HGC system pages describe operation from April 2015 (a priority segment from January 2015) with 6 fat nodes (Xeon E7-8857 v2, 2TB each) and 469 thin nodes (24-core Xeon E5-2670 v3, 128GB each, 24 of them with a Xeon Phi 7120P), a 12 PiB home disk, 1 PiB archive disk plus 21 PB of tape, Red Hat Enterprise Linux, indirect evaporative cooling with an estimated PUE of 1.06 and about 422 TFlop/s peak. TOP500 lists the HPE Apollo 6000 Xl230/250 system as 12,408 cores on InfiniBand FDR, Rmax 342.6 and Rpeak 431.7 TFlop/s at 186.6 kW, ranked 193rd in June 2015, 288th in November 2015 and 400th in June 2016.
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
- 2026-10-09 Added Shirokane3
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 |
|---|---|---|
| supcom.hgc.jp | names system only | - |
| top500.org | names system and part | Mellanox InfiniBand FDR |
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
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.