Supercomputers/Systems/SK Telecom Titan
South Korea · operational · ai training
SK Telecom Titan
Also known as Titan (SK Telecom)
Operated by SK Telecom .
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
- 19.53 PFlop/s
- Rpeak
- 24.95 PFlop/s
- Rmax ÷ Rpeak
- 78.3%
- Cores
- 153,760
- 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 Apollo 6500 | Hewlett Packard Enterprise | - | Reported | top500.org |
| Accelerator | NVIDIA A100 SXM4 80GB Accelerators NVIDIA A100 SXM4 80 GB per TOP500; 1,040 A100 GPUs after the February 2023 doubling per RCR Wireless | NVIDIA | 1,040 reported | Reported | top500.orgrcrwireless.com |
| 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.org |
| Interconnect | NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds Infiniband HDR | NVIDIA | - | Reported | top500.org |
| Operating system | Red Hat Enterprise Linux Operating systems Linux distribution · Open source, commercial support subscription RHEL 8.6 | Red Hat | - | Reported | top500.org |
AI datacenter metrics
Figures beyond an accelerator count, each with who stated it. What can and cannot be compared across AI datacenters.
| Metric | Value | As of | Basis | Note | Source |
|---|---|---|---|---|---|
| Accelerators | 1,040 accelerators | 2023-02 | Operator stated | RCR Wireless: Titan doubled to 1,040 A100 GPUs and 17.1 petaflops (Feb 2023). | rcrwireless.com |
The read
SK Telecom's Titan AI supercomputer in Seoul, an HPE Apollo 6500 cluster used to train the company's Aidat language model. TOP500 lists AMD EPYC 7763 64C 2.45GHz hosts, NVIDIA A100 SXM4 80 GB accelerators, InfiniBand HDR and RHEL 8.6, with 153,760 cores, Rmax 19.53 and Rpeak 24.95 PFlop/s; it ranked 92nd in November 2022, 47th in June 2023 and 108th in June 2026. RCR Wireless (15 February 2023) reports a beta launch in May 2022 and a doubling of capacity to 1,040 NVIDIA A100 GPUs at 17.1 petaflops, a figure on a different basis from the HPL result and not recorded as a FLOPS field. TOP500 gives 2024 as the install year, which conflicts with the 2022 beta date, so the earlier date is used. The data centre location is not stated in either page fetched, so the site is left blank. The name is disambiguated from Oak Ridge's Titan.
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
- 2026-10-04 Added SK Telecom Titan (sk-telecom)
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 |
|---|---|---|
| top500.org | names system and part | AMD EPYC 7763 (Milan), NVIDIA A100 SXM4 80GB, NVIDIA InfiniBand HDR |
| rcrwireless.com | names system, part and count | NVIDIA A100 SXM4 80GB · counts: 1040 |
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.