Supercomputers/Systems/KISTI-6
South Korea · under construction · mixed
KISTI-6
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
- -
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- -
- Power
- -
- Per watt
- - 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 Supercomputing EX4000 | Hewlett Packard Enterprise | - | Reported | finance.yahoo.comintelligentcio.com |
| Accelerator | NVIDIA GH200 Grace Hopper Superchip Accelerators Arm CPU + CUDA GPU · TSMC 4N NVIDIA GH200 GPUs | NVIDIA | - | Reported | evertiq.com |
| Storage | HPE Cray ClusterStor E2000 Storage HPE Cray E2000 with NVMe and NL-SAS | Hewlett Packard Enterprise | - | Reported | finance.yahoo.com |
The read
South Korea's sixth national supercomputer, an HPE Cray EX4000 with NVIDIA GH200 and H200 GPUs and AMD EPYC CPUs. Installation finished in August 2026 and official service is targeted for December 2026. Its sources disagree on almost every headline figure: peak is given as 588, 600 and 614 PFlop/s, GPUs as 8,496 against node counts that do not reconcile, and cooling as fully fanless direct liquid against 96% liquid with rear-door exchangers, so no peak, GPU count or cooling edge is recorded. HPE's own release is reachable here only through a Business Wire copy, so the tier is reported.
Timeline
- 2025-05 Announced source
Site & facility
Where sources disagree
We record conflicts instead of picking a winner quietly. All open questions.
- Peak performance: 600 PFlop/s (HPE release copy) against 614 PFlop/s (DigitalToday). A third source gives 588; no peak is recorded.
- Cooling: 100% fanless direct liquid cooling (HPE) against 96% liquid cooled with rear-door exchangers. No cooling edge is recorded.
Change history
- 2026-09-20 Planned system added; its sources disagree on peak, GPUs and cooling
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.
One cited source was readable and does not mention this system; another could not be read because the host returned a bot challenge. The unreadable one may well be fine, so this is weaker evidence than a corroborated row and much stronger than an unsupported one. See how this check works.
| Cited source | Result | Found on the page |
|---|---|---|
| finance.yahoo.com | could not be fetched | - |
| digitaltoday.co.kr | does not name this system | NVIDIA GH200 Grace Hopper Superchip |
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
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
Storage and I/O
Parallel file systems, why checkpointing dominates the write load, and the metadata failure mode nobody expects.