Supercomputers/Systems/Voltage Park H100 Cloud
USA · operational · ai training
Voltage Park H100 Cloud
Also known as Voltage Park 24, 000 H100 AI cloud
Operated by Voltage Park .
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: 1263 of 1345 systems have a readable citation that names them, and 56 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 | - Dell PowerEdge XE9680 servers host the H100 GPUs | Dell Technologies | - | Reported | blocksandfiles.com |
| Accelerator | NVIDIA H100 SXM5 Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W Approximately 24,000 NVIDIA H100 80 GB SXM5 GPUs | NVIDIA | 24,000 reported | Reported | blocksandfiles.comtheregister.com |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds NVIDIA Quantum-2 InfiniBand interconnect | NVIDIA | - | Reported | blocksandfiles.com |
| Storage | VAST Data Platform Storage Disaggregated shared-everything VAST Data storage deployed across the majority of Voltage Park's data centers | VAST Data | - | Reported | blocksandfiles.com |
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 | 24,000 accelerators | 2025-06 | Operator stated | Approximately 24,000 NVIDIA H100 GPUs across six data centers in Washington, Texas, Virginia and Utah. | blocksandfiles.comtheregister.com |
The read
Voltage Park's own AI cloud of approximately 24,000 NVIDIA H100 80 GB SXM5 GPUs in Dell PowerEdge XE9680 servers interconnected with Quantum-2 InfiniBand, run by a wholly owned subsidiary of the Navigation Fund. The Register reported on 30 October 2023 that Voltage Park had acquired 24,000 H100s for about $500 million and planned sites in Texas, Virginia and Washington; Blocks and Files reported in June 2025 that the cloud spans six Tier 3+ data centers in Washington, Texas, Virginia and Utah, with VAST Data storage deployed across the majority of them. The fleet is split across several sites, so no single site is recorded. It is an owned fleet, not a rented allocation. No HPL or vendor FLOPS figure is published. The two outlets are independent but both relay company statements, so the tier is reported.
Timeline
- 2023-10-29 Announced source
Change history
- 2026-10-09 Added Voltage Park H100 Cloud
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 |
|---|---|---|
| blocksandfiles.com | names system, part and count | NVIDIA H100 SXM5, NVIDIA InfiniBand NDR (Quantum-2), VAST Data Platform · counts: 24000 |
| theregister.com | names system, part and count | NVIDIA H100 SXM5, VAST Data Platform · counts: 24000 |
Checked 2026-10-13 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.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.