Supercomputers/Systems/TUPAC
Argentina · operational · mixed
TUPAC
Also known as Tupac supercomputer; CSC-CONICET TUPAC
Operated by CSC-CONICET at CSC-CONICET (Polo Cientifico Tecnologico) .
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
- 54.42 TFlop/s
- Rmax ÷ Rpeak
- -
- Cores
- 3,712
- 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 |
|---|---|---|---|---|---|
| Accelerator | - NVIDIA Tesla M2090 (RISC2 table: 32*M2090) | NVIDIA | 32 reported | Reported | risc2-project.eu |
| CPU | AMD Opteron 6276 (Interlagos) CPUs x86-64 · 16C · 2.3 GHz · 32 nm · 115 W AMD Opteron 6276, 4 per node on 58 nodes | AMD | 232 reported | Reported | risc2-project.eu |
| Interconnect | Mellanox InfiniBand QDR Interconnect Fat tree · 40 Gb/s per 4x port InfiniBand QDR (two independent networks per CSC) | NVIDIA | - | Reported | csc.conicet.gov.ar |
The read
Argentina's CPU/GPU cluster at the Centro de Simulacion Computacional para Aplicaciones Tecnologicas (CSC) of CONICET in Buenos Aires, presented in March 2015 per the Spanish Wikipedia entry and funded by FONARSEC and CONICET. The RISC2 HPC observatory (April 2022) lists 58 nodes of four AMD Opteron 6276 CPUs plus 128 GiB with InfiniBand QDR, and 32 NVIDIA M2090 GPUs, 3,712 CPU cores and a 54.418 TFlop/s combined peak; CSC's own page states 48 TFlop/s, 4,096 Opteron cores and 16,384 CUDA cores on two InfiniBand QDR and two Ethernet networks with 72 TB of storage, so the figures differ slightly and the RISC2 table is used. The status date is the RISC2 listing; no more recent confirmation was fetched. It has never been on TOP500. The two pages come from different organizations but the CSC page gives no dates, so the tier is reported.
Timeline
- 2015-03 First operational source
Site & facility
Change history
- 2026-10-04 TUPAC added, the Opteron and Tesla M2090 cluster of CONICET's CSC in Buenos Aires, Argentina's GPU supercomputer from 2015
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 |
|---|---|---|
| csc.conicet.gov.ar | names system and part | AMD Opteron 6276 (Interlagos), Mellanox InfiniBand QDR |
| risc2-project.eu | names system and part | AMD Opteron 6276 (Interlagos) |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.