Supercomputers/Systems/Joule 3.0
USA · operational · hpc
Joule 3.0
Also known as NETL Joule 3.0
Operated by National Energy Technology Laboratory at NETL HPC Facility (Morgantown) .
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
- 4.86 PFlop/s
- Rpeak
- 6.11 PFlop/s
- Rmax ÷ Rpeak
- 79.5%
- Cores
- 107,648
- 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 | - Koi Computers (TOP500 vendor) | Koi Computers | - | Reported | top500.org |
| Accelerator | - 72 GPUs, H100 | NVIDIA | 72 reported | Reported | hpc.netl.doe.govhpc.netl.doe.gov |
| CPU | AMD EPYC 9004 (Genoa) CPUs x86-64 · 96C · 2.4 GHz · TSMC N5 · 360 W AMD EPYC 9534 64C 2.45GHz | AMD | - | Verified | top500.orghpc.netl.doe.gov |
| Interconnect | - InfiniBand, 200-400 GB/s network | NVIDIA | - | Reported | hpc.netl.doe.gov |
The read
The third generation of the Joule supercomputer at the DOE National Energy Technology Laboratory in Morgantown, West Virginia, refreshed in June 2024 at a cost of 21.66 million dollars for a stated near 40 percent capacity gain. TOP500 lists a Koi Computers system with AMD EPYC 9534 64C 2.45GHz processors over InfiniBand running Linux, Rmax 4.86 PFlop/s, Rpeak 6.11 PFlop/s and 107,648 cores; it first appeared on the June 2024 list at rank 181. NETL's pages independently state 107,648 CPU cores, AMD EPYC 9534 processors, 6.11 PFLOPS peak, 72 GPUs (H100), 481 TB of memory, over 19.5 PB of storage and a 200-400 GB/s InfiniBand network, and name 215th on the November 2024 TOP500. NETL cites 4.82 PFLOPS of additional GPU peak that is not recorded here.
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
- 2026-10-02 Added Joule 3.0
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 |
|---|---|---|
| hpc.netl.doe.gov | names system only | - |
| top500.org | names system and part | AMD EPYC 9004 (Genoa) |
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.