Attention: Confluence is not suitable for the storage of highly confidential data. Please ensure that any data classified as Highly Protected is stored using a more secure platform.
If you have any questions, please refer to the University's data classification guide or contact ict.askcyber@sydney.edu.au. When using in-built AI tools ensure you are in compliance with the University's responsible AI guardrails
Apollo GPU cluster
Welcome to the user guide for the Apollo GPU cluster, the University’s new dedicated GPU platform managed out of Sydney Informatics Hub.
See also: Apollo GPU Cluster Usage Policies
Fee-for-service applies to Apollo usage. Rates for University of Sydney researchers are highly competitive and subsidised by the University. A limited free tier is also available to each Chief Investigator to enable trial. See ‘Rates and billing’ page.
Info Sessions
Request an information session for your area.
Access Apollo
To work on Apollo, you’ll need to provision access for your chosen DashR project and complete an onboarding form. Our rates and billing information is also provided here:
Technical specifications
The Apollo GPU cluster is managed using Kubernetes. Three master nodes are configured for managing the cluster and the DGX nodes function as Kubernetes worker nodes. The Run:AI scheduler is installed on Kubernetes for workload scheduling and AI workload orchestration.
The Platform is currently running NVIDIA Base Command Manager with DGX OS 7.
Compute
| H200 GPUs | VRAM (Gb) | CPU cores | RAM (Gb) | Disk (Tb) |
|---|---|---|---|---|---|
Per H200 GPU | 1 | 141 | 14 | 275 | 3.8 |
Per DGX node | 8 | 1128 | 112 | 2200 | 30.7 |
Whole cluster | 24 | 3384 | 336 | 6600 | 92.2 |
Storage
1 Pb DDN EXAscaler parallel filesystem
Connectivity: 8x 200GbE active optical cables
Controller: DDN ES400NVX2-NDR200-SE
Drives: 48 x 30.72TB QLC NVMe G4 4K SSD drives (total)
Get support
For enquiries and support please use the https://sydney.au1.qualtrics.com/jfe/form/SV_5mXyhFZsPIwZDBs
Who’s using Apollo?
As at 14th October 2026, we've got 540 researchers across 191 research projects onboarded to Apollo, across 7 faculties. By faculty:
Engineering & IT (214 researchers, 92 projects);
Medicine & Health (160 researchers, 45 projects);
Science (144 researchers, 42 projects);
Business (9 researchers, 5 projects);
Arts & Social Sciences (7 researchers, 4 projects);
Conservatorium of Music (2 researchers, 2 projects);
Architecture, Design and Planning (4 researchers, 1 project).
Acknowledge Apollo in your publications
Your acknowledgement enables us to provide these services.
Standard Acknowledgement:
The authors acknowledge the use of the Apollo GPU cluster, a service provided by the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.
Apollo Cover Art Acknowledgement
The cover art for Apollo GPU cluster was created by Dr. Eden Zhang, Data Scientist at Sydney Informatics Hub.