Argentina · operational · hpc

Serafin

Operated by UNC / CCAD at CCAD (Universidad Nacional de Cordoba) .

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: -
Rpeak: -
Rmax ÷ Rpeak: -
Cores: 3,840
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	- 15 Supermicro AS-2124BT-HTR 2U chassis, 60 nodes	Supermicro SMCINASDAQ	-	Reported	supercomputo.unc.edu.ar
CPU	AMD EPYC 7532 CPUs x86-64 · 32C · 2.4 GHz · 7 nm · 200 W 120 AMD EPYC 7532 (32C) processors, two per node across 60 nodes	AMD AMDNASDAQ	120 reported	Reported	supercomputo.unc.edu.ar
Interconnect	NVIDIA InfiniBand HDR Interconnect Fat tree or dragonfly+ · 200 Gb/s per port · About 0.6 microseconds NVIDIA ConnectX-6 HDR100 InfiniBand per node via a Mellanox MQM8790-HS2F switch	NVIDIA NVDANASDAQ	-	Reported	supercomputo.unc.edu.ar

The read

A Supermicro cluster at the Universidad Nacional de Cordoba's high-performance computing centre (CCAD): 60 dual-socket nodes in 15 AS-2124BT-HTR chassis, 120 AMD EPYC 7532 processors of 32 cores each for 3,840 cores total, NVIDIA ConnectX-6 HDR100 InfiniBand, BeeGFS over node-local NVMe, and Rocky Linux. It joined CCAD's older Mendieta, Mulatona and Eulogia clusters, was delivered in February 2021 and publicly inaugurated in August 2021, named for a character in Roberto Fontanarrosa's comic strip Inodoro Pereyra, funded through Argentina's ANID/PAGE major-equipment programme at a cost of US$371,000. It has never been submitted to TOP500. CCAD's own current equipment page and the 2021 launch coverage disagree on peak performance (147.456 TFlop/s against 156 TFlop/s as announced at launch), so no Rmax or Rpeak is recorded; node and core counts agree across both and are recorded. No source we could fetch states a GPU count, and CCAD's own page describes no GPU acceleration on this cluster, so none is recorded as a component.

Timeline

2021-08 First operational source

Site & facility

CCAD (Universidad Nacional de Cordoba)

Location: Cordoba, Argentina
Commissioned: 2021

supercomputo.unc.edu.ar unc.edu.ar

Where sources disagree

We record conflicts instead of picking a winner quietly. All open questions.

Peak performance: 147.456 TFlop/s, CCAD's current equipment page against 156 TFlop/s, as announced at the August 2021 launch. The figure announced at Serafin's 2021 inauguration is about 6% higher than CCAD's own current technical page states for the deployed 60-node configuration; no Rpeak is recorded in the systems table.

Change history

2026-09-22 Added as an unranked system; sources disagree on peak performance so none is recorded

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
supercomputo.unc.edu.ar	names system, part and count	AMD EPYC 7532, NVIDIA InfiniBand HDR · counts: 120
unc.edu.ar	names system only	-

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.