Supercomputers/Systems/CSIRO GPU Cluster
Australia · upgraded · hpc
CSIRO GPU Cluster
Also known as CSIRO Nitro GPU cluster
Operated by CSIRO .
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
- 335.30 TFlop/s
- Rpeak
- 472.50 TFlop/s
- Rmax ÷ Rpeak
- 71.0%
- Cores
- 6,875
- Power
- 550 kW
- Per watt
- 0.6 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 | - XENON Systems Nitro G16 3GPU GPU cluster, integrated by XENON | XENON Systems | - | Reported | top500.orgitnews.com.au |
| Accelerator | - NVIDIA Tesla K20m GPUs (final TOP500 configuration); earlier Tesla S1070 and Tesla 2050 generations | NVIDIA | - | Reported | top500.orgitnews.com.au |
| CPU | - Intel Xeon E5-2650 8C 2GHz, 268 CPUs in 134 nodes (2012 configuration) | Intel | 268 reported | Reported | itnews.com.autop500.org |
| Interconnect | Mellanox InfiniBand FDR Interconnect Fat tree · 56 Gb/s per 4x port · About 1 microsecond Mellanox FDR InfiniBand | NVIDIA | - | Reported | top500.org |
The read
CSIRO's GPU-accelerated cluster built by the Australian integrator XENON Systems and upgraded in place over several years. iTnews reported its launch on 23 November 2009 as 128 dual-Xeon E5462 nodes (1,024 cores) with 64 Tesla S1070 units (256 GPUs), then in April 2012 as 134 compute nodes with 268 eight-core Xeon E5-2650 CPUs and 390 Tesla 2050 cards for a targeted 120 to 130 TFlop/s, a roughly A$1 million upgrade; the pre-upgrade machine drew 95 kW and ranked 11th on the Green500. TOP500's page shows the later Nitro G16 configuration, Xeon E5-2650 8C 2GHz with Tesla K20m and FDR InfiniBand, 6,875 cores, Rmax 335.30 and Rpeak 472.50 TFlop/s, 550 kW, listed from November 2010 (rank 146) to June 2016 (rank 411); the core count there counts GPU cores, so it differs from the node-level counts above. Because the hardware changed several times the status is recorded as upgraded and the figures are those of the final TOP500 configuration.
Timeline
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Change history
- 2026-10-09 Added CSIRO GPU Cluster
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 |
|---|---|---|
| top500.org | names system and part | Mellanox InfiniBand FDR |
| itnews.com.au | blocked our fetch | - |
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.