Supercomputers/Systems/HiPerGator AI
USA · operational · ai training
HiPerGator AI
Also known as HiPerGator 4.0 AI SuperPOD
Operated by University of Florida .
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
- 17.49 PFlop/s
- Rpeak
- 19.00 PFlop/s
- Rmax ÷ Rpeak
- 92.1%
- Cores
- 81,648
- 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 B200 SuperPOD, 63 DGX B200 systems | NVIDIA | 63 reported | Verified | top500.orgnews.ufl.edu |
| Accelerator | NVIDIA B200 SXM Accelerators CUDA (Blackwell) · TSMC 4NP · 1,000 W 504 NVIDIA Blackwell B200 GPUs (180 GB) in 63 DGX B200 nodes | NVIDIA | 504 reported | Reported | it.ufl.edu |
| Accelerator | NVIDIA A100 (in DGX A100) Accelerators 140 NVIDIA DGX A100 systems (original 2021 SuperPOD, being replaced by DGX B200) | NVIDIA | - | Reported | news.ufl.edu |
| CPU | Intel Xeon Platinum 8570 CPUs x86-64 · 56C · 2.1 GHz · Intel 7 Intel Xeon Platinum 8570 56C 2.1GHz | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds NVIDIA Quantum-2 InfiniBand NDR400, 400 Gb/s per GPU | NVIDIA | - | Verified | top500.orgnews.ufl.edu |
| Storage | Lustre Storage Parallel filesystem, separate metadata and object servers Two-tier Lustre: 36 PB capacity tier and 11 PB all-flash DDN tier (over 70 PB total distributed storage) | DDN | - | Reported | it.ufl.edu |
| Operating system | Red Hat Enterprise Linux Operating systems Linux distribution · Open source, commercial support subscription RHEL 9.5 | Red Hat | - | Reported | top500.org |
The read
The University of Florida's NVIDIA DGX SuperPOD AI system within its HiPerGator supercomputer. The original SuperPOD, announced in July 2020 and opened to the state university system in January 2021, was 140 NVIDIA DGX A100 systems on HDR InfiniBand (UF news release), a configuration UF reports ranked 22nd on the June 2021 TOP500 and 2nd on the Green500. UF's Board of Trustees approved a 24 million dollar upgrade, and UF now lists an AI SuperPOD of 63 NVIDIA DGX B200 nodes with 504 B200 GPUs on Quantum-2 NDR InfiniBand, with first delivery in 2025. The HPL figures here (17.49 PFlop/s Rmax, 19.00 PFlop/s Rpeak, 81,648 cores) are TOP500's current entry for the DGX B200 system (Xeon Platinum 8570 hosts, Infiniband NDR400, RHEL 9.5). UF's storage is a two-tier Lustre system (36 PB capacity tier and an 11 PB all-flash DDN tier). The rest of HiPerGator (about 60,000 AMD EPYC cores, L4 and RTX PRO 6000 GPUs) is not modelled here. The UF data center location was not stated in the sources fetched, so no site is attached.
Timeline
- 2020-07 Announced source
- 2021 First operational source
- 2025-11 List debut at #106 (November 2025) source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-01 Added HiPerGator AI (University of Florida)
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 |
|---|---|---|
| it.ufl.edu | names system, part and count | NVIDIA B200 SXM, Lustre · counts: 504 |
| top500.org | names system and part | NVIDIA B200 SXM, Intel Xeon Platinum 8570, NVIDIA InfiniBand NDR (Quantum-2), NVIDIA A100 (in DGX A100) |
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.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.