Supercomputers/Systems/ISM Supercomputer System I
Japan · operational · hpc
ISM Supercomputer System I
Also known as ISM Supercomputer System "I"; Supercomputer System for Statistical Science "I"; SGI ICE X (ISM)
Operated by Institute of Statistical Mathematics .
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
- 360.00 TFlop/s
- Rpeak
- 430.00 TFlop/s
- Rmax ÷ Rpeak
- 83.7%
- Cores
- 18,624
- Power
- 261.84 kW
- Per watt
- 1.4 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 | - SGI ICE X distributed-memory supercomputer, 400 nodes in the institute description | SGI HPENYSE
rolls up to Hewlett Packard Enterprise | - | Reported | ism.ac.jptop500.org |
| CPU | Intel Xeon E5-2697 v2 CPUs x86-64 · 12C · 2.7 GHz · 22 nm · 130 W Intel Xeon E5-2697 v2 12C 2.7GHz, two per node on 400 nodes | Intel | 800 estimated | Reported | ism.ac.jptop500.org |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond InfiniBand FDR | NVIDIA | - | Reported | top500.org |
| Storage | - DDN SFA12K-40, 2.5 PB RAID6 object storage on Lustre | DDN | - | Reported | ism.ac.jp |
| Operating system | SUSE Linux Enterprise Server Operating systems Linux distribution · Open source, commercial support subscription SUSE Linux Enterprise Server 11 | SUSE | - | Reported | ism.ac.jptop500.org |
The read
Supercomputer System for Statistical Science "I" of the Institute of Statistical Mathematics. The institute's own page describes a distributed-memory SGI ICE X of 400 nodes (two Xeon E5-2697v2 2.7 GHz CPUs, 128 GB each), 207 TFlop/s and 50 TB in total, SUSE Linux Enterprise Server 11, an SGI UV 20 front end, a hardware random-number sub-system and 2.5 PB of DDN SFA12K-40 Lustre storage. TOP500 lists a larger configuration of 18,624 cores (Xeon E5-2697v2 plus E5-2680v3 and Xeon Phi 5120D accelerators), 0.36 PFlop/s Rmax, 0.43 PFlop/s Rpeak and 261.84 kW, entering in November 2014 (301st), 265th in November 2015 and last seen in November 2016. The two sources describe different build-outs of the same installation, so the tier is reported. The machine room location is not on the pages fetched, so no site is recorded; the later ISM SGI 8600 and the Altix 3700 are separate rows in this dataset.
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 ISM Supercomputer System I
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 |
|---|---|---|
| ism.ac.jp | names system and part | Intel Xeon E5-2697 v2, SUSE Linux Enterprise Server |
| top500.org | names system and part | Intel Xeon E5-2697 v2, Mellanox InfiniBand FDR, SUSE Linux Enterprise Server |
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.