Supercomputers/Systems/NVIDIA DGX SuperPOD (V100)
USA · operational · ai training
NVIDIA DGX SuperPOD (V100)
Also known as DGX SuperPOD DGX-2H
Operated by NVIDIA at NVIDIA (undisclosed US site) .
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
- 9.44 PFlop/s
- Rpeak
- 11.21 PFlop/s
- Rmax ÷ Rpeak
- 84.2%
- Cores
- 127,488
- 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 | - NVIDIA DGX-2H servers | NVIDIA | - | Reported | top500.orgforums.developer.nvidia.com |
| Accelerator | NVIDIA Tesla V100 Accelerators CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W 1,536 NVIDIA Tesla V100 SXM3 GPUs in 96 DGX-2H servers | NVIDIA | 1,536 reported | Reported | forums.developer.nvidia.comtop500.org |
| CPU | Intel Xeon Platinum 8174 CPUs x86-64 · 24C · 3.1 GHz · 14 nm · 240 W Xeon Platinum 8174 24C 3.1GHz | Intel | - | Reported | top500.org |
| Interconnect | Mellanox InfiniBand EDR Interconnect Fat tree · 100 Gb/s per 4x port · About 0.9 microseconds Mellanox InfiniBand EDR | NVIDIA | - | Reported | top500.org |
| Operating system | Ubuntu Server Operating systems Linux distribution · Open source, free with paid support tiers Operating System: Ubuntu 18.04.01 | Canonical | - | 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 | 1,536 accelerators | 2019-06 | Operator stated | NVIDIA: 96 DGX-2H servers containing 1,536 Tesla V100 SXM3 GPUs | forums.developer.nvidia.com |
The read
NVIDIA's first DGX SuperPOD: 96 DGX-2H servers holding 1,536 Tesla V100 SXM3 GPUs on Mellanox EDR InfiniBand. NVIDIA's DGX SuperPOD post says it earned the 22nd spot on the June 2019 TOP500, which matches the TOP500 ranking history (rank 22 in June 2019, 20 in November 2019, 199 in June 2026 at 9.44 PFlop/s). TOP500 gives the installation year as 2020 and no power figure. This is a separate TOP500 entry from the DGX SaturnV and Selene rows.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Change history
- 2026-10-09 Added NVIDIA DGX SuperPOD (V100)
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 |
|---|---|---|
| forums.developer.nvidia.com | names system, part and count | NVIDIA Tesla V100 · counts: 1536 |
| top500.org | names system and part | NVIDIA Tesla V100, Intel Xeon Platinum 8174, Mellanox InfiniBand EDR, Ubuntu Server |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Inside a node
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.
Programming a supercomputer
The software stack beneath the application, and why the OS is now a sovereignty signal as much as a technical choice.