COMPUTE ATLAS Supercomputer supply-chain graph
1153 systems 482 sites Sourced data

Supercomputers/Systems/Katana

Australia · operational · mixed

Katana

Also known as UNSW Katana

Operated by UNSW Sydney .

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
-
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 -
Heterogeneous servers from Dell, Lenovo, Supermicro and XENON
Dell Technologies - Reported docs.restech.unsw.edu.au
Integrator -
Heterogeneous servers from Dell, Lenovo, Supermicro and XENON
Lenovo - Reported docs.restech.unsw.edu.au
Integrator -
Heterogeneous servers from Dell, Lenovo, Supermicro and XENON
Supermicro - Reported docs.restech.unsw.edu.au
Integrator -
Heterogeneous servers from Dell, Lenovo, Supermicro and XENON
XENON Systems
- Reported docs.restech.unsw.edu.au
Accelerator NVIDIA H200 SXM
Accelerators
CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W
NVIDIA H200 141GB, 80 GPUs in 13 nodes
NVIDIA - Reported docs.restech.unsw.edu.au
Accelerator NVIDIA L40S
Accelerators
CUDA (Ada Lovelace) · 350 W
NVIDIA L40S 48GB, 32 GPUs in 7 nodes
NVIDIA - Reported docs.restech.unsw.edu.au
Accelerator NVIDIA Tesla V100
Accelerators
CUDA (Volta GV100) · 5,120C · 1.53 GHz · 12 nm · 300 W
NVIDIA V100 32GB, 32 GPUs in 8 nodes
NVIDIA - Reported docs.restech.unsw.edu.au
Accelerator NVIDIA GH200 Grace Hopper Superchip
Accelerators
Arm CPU + CUDA GPU · TSMC 4N
NVIDIA GH200 480GB + 96GB, one node
NVIDIA - Reported docs.restech.unsw.edu.au

The read

UNSW Sydney's shared computational cluster, run by Research Technology Services and aimed at researchers with non-sensitive data. The current user documentation describes 180 nodes (145 CPU and 35 GPU) with nearly 12,000 CPU cores, 186 GPU cards and 26 PB of disk: 80 NVIDIA H200 (13 nodes), 32 L40S, 32 V100, 24 A100 40GB, 16 RTX Pro 6000, plus one H100 and one GH200. It runs Rocky Linux 8 with OpenPBS and is built from Dell, Lenovo, Supermicro and XENON servers. The figures have grown quickly: UNSW's facility page still lists 8,076 cores, 171 nodes and 117 GPUs, and a 2025 arXiv survey of Australian institutions gives 7,384 cores, 165 nodes and 117 GPUs, so the larger current-docs numbers are not independently confirmed. No FLOPS figure is published and no commissioning date was found.

Timeline

No dated events recorded.

Change history

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 sourceResultFound on the page
arxiv.orgdoes not name this system-
docs.restech.unsw.edu.aunames system and partNVIDIA H200 SXM, NVIDIA L40S, NVIDIA GH200 Grace Hopper Superchip

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

Further reading