Supercomputers/Systems/Fran
USA · operational · mixed
Fran
Also known as ARL DSRC Fran; TI-24 Fran
Operated by ARL DSRC .
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
- 8.83 PFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- 173,760
- 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 Cray EX4000 system | Hewlett Packard Enterprise | - | Reported | centers.hpc.miltop500.org |
| Accelerator | NVIDIA L40S Accelerators CUDA (Ada Lovelace) · 350 W NVIDIA L40S, one per visualization node, 24 nodes | NVIDIA | 24 reported | Reported | centers.hpc.mil |
| Accelerator | - NVIDIA H100 NVL, two per AI/ML node, 32 nodes | NVIDIA | 64 reported | Reported | centers.hpc.mil |
| CPU | AMD EPYC 9004 (Genoa) CPUs x86-64 · 96C · 2.4 GHz · TSMC N5 · 360 W AMD EPYC 9654 Genoa 96C 2.4 GHz, 792 standard, 24 large-memory, 24 visualization and 32 AI/ML nodes | AMD | - | Reported | centers.hpc.miltop500.org |
| Interconnect | HPE Slingshot-11 Interconnect Dragonfly · 200 Gb/s per port · Around 2 microseconds HPE Slingshot-11 interconnect | Hewlett Packard Enterprise | - | Reported | top500.org |
| Operating system | Red Hat Enterprise Linux Operating systems Linux distribution · Open source, commercial support subscription Red Hat Enterprise Linux 9 | Red Hat | - | Reported | centers.hpc.mil |
The read
The newest unclassified supercomputer of the Army Research Laboratory DoD Supercomputing Resource Center, an HPE Cray EX4000 on AMD EPYC 9654 Genoa processors. The DoD HPCMP centers pages list 792 standard compute nodes, 24 large-memory nodes, 24 visualization nodes (one NVIDIA L40S each), 32 AI/ML nodes (two NVIDIA H100 NVL each) and 30 Persistent System Framework nodes, 173,760 compute cores, 434 TB of usable memory and a rated peak of 8.826 PFlop/s, scheduled with Slurm under a stated end of life of 30 December 2030. The center announced on its systems-news page that Fran became available for allocated use on 1 October 2026 (the announcement text gives 173,184 cores, so the HPCMP pages differ slightly on the core total; the 173,760 figure is the one on the system description). TOP500 lists a Fran entry (HPE Cray EX, EPYC 9654, Slingshot-11, 151,680 cores) at rank 309 on the June 2026 list with Rmax 4.87 PFlop/s, which looks like a partial-system run; that Rmax is not recorded here because it does not describe the full machine. Only the operator peak is recorded; tier reported because the operator figures come from a single HPCMP source family and TOP500 describes a different configuration.
Timeline
- 2026-10-01 First operational source
Change history
- 2026-10-09 Added Fran
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 |
|---|---|---|
| centers.hpc.mil | could not be fetched | - |
| top500.org | names system, part and count | AMD EPYC 9004 (Genoa), HPE Slingshot-11 · counts: 24 |
Checked 2026-10-13 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.