Supercomputers/Systems/kakaocloud EPYC A100 Cluster
South Korea · upgraded · mixed
kakaocloud EPYC A100 Cluster
Operated by Kakao Enterprise .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1581 systems have a readable citation that names them, and 64 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
- 32.00 PFlop/s
- Rpeak
- 37.38 PFlop/s
- Rmax ÷ Rpeak
- 85.6%
- Cores
- 253,952
- 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.
How it compares
Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of kakaocloud EPYC A100 Cluster, and those in South Korea. Computed from our tables; a metric is ranked only when at least eight peers report it.
| Metric | Value | Against its era | Against its country |
|---|---|---|---|
| Rmax | 32.00 PFlop/s | #66 of 234 · top 28% | #6 of 32 · top 16% |
| Rmax ÷ Rpeak | 85.6 % | #53 of 231 · top 23% | #7 of 32 · top 19% |
Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.
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 | - Supermicro AS-4124GO-NART | Supermicro | - | Reported | top500.org |
| Accelerator | - NVIDIA A100 80GB | NVIDIA | - | Reported | top500.orgbyline.network |
| CPU | AMD EPYC 7763 (Milan) CPUs x86-64 · 64C · 2.45 GHz · 7 nm · 280 W AMD EPYC 7763 64C 2.45GHz | AMD | - | Reported | top500.orgbyline.network |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds Infiniband HDR200 | NVIDIA | - | Reported | top500.orgbyline.network |
| Operating system | Ubuntu Server Operating systems Linux distribution · Open source, free with paid support tiers Ubuntu 22.04.3 LTS | Canonical | - | Reported | top500.org |
Engineering, funding and access
What kakaocloud EPYC A100 Cluster runs on, who paid for it and how people get time on it. Each line carries who said it and when.
Hardware
| Item | Detail | As of | Basis | Source |
|---|---|---|---|---|
| Interconnect and topology | InfiniBand HDR200Byline reports InfiniBand; HDR200 is from TOP500 | 2024-05 | Press report | byline.networktop500.org |
Results
| Item | Detail | As of | Basis | Source |
|---|---|---|---|---|
| Notable result | 2024: debuted at 44th on the TOP500 at 21.21 PFlop/s with 89.7 percent Linpack efficiency, the only domestic cloud provider on the list | 2024-05 | Press report | byline.network |
The read
The larger of the two kakaocloud systems that entered the TOP500 in June 2024 (the other is kakaocloud-xeon-a100): a Supermicro AS-4124GO-NART cluster of AMD EPYC 7763 64-core hosts with NVIDIA A100 80 GB GPUs and HDR200 InfiniBand under Ubuntu 22.04.3. It debuted at rank 44 with 160,704 cores and 21.21 PFlop/s (Korean trade press: 89.7% Linpack efficiency, the only 160,000-core class system above 20 PFlop/s), then grew to 253,952 cores and 32.00 PFlop/s Rmax (37.38 PFlop/s Rpeak) from the November 2024 list, ranking 41st then and 74th on the June 2026 list; hence status upgraded, with the later figures recorded and the appearance row holding the 2024 debut. Neither source gives a GPU count or power figure. Kakao's AI-cluster peers in the dataset include nipa-kakao.
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
- 2026-10-18 Added kakaocloud EPYC A100 Cluster (Kakao Enterprise)
Nearby in scale
The machines closest to kakaocloud EPYC A100 Cluster by Rmax, smaller to larger.
Related systems
Same operator
- kakaocloud Xeon A100 Cluster · South Korea, operational
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.