DeltaAI node architecture
DeltaAI is operated by NCSA · University of Illinois in United States. This profile covers the documented compute inventory.
Cluster interconnect: HPE Slingshot 11 · 4 × 200 Gb/s. Topology: Not verified.
GH200 ×4
- Node count
- 152
- CPU
- NVIDIA Grace · 72 cores
- CPUs per node
- 4
- Accelerator
- NVIDIA Hopper (GH200)
- Physical accelerators per node
- 4
- Host / unified memory
- 480 GB LPDDR5
- Accelerator memory
- 96 GB HBM3 per GPU
- Local accelerator links
- NVLink · all GPU pairs
- CPU-to-accelerator link
- NVLink-C2C
- Storage
- 3.9 TB local
Scope and uncertainty
152 compute nodes: 132 NSF/NAIRR-funded and 20 Illinois Computes-funded. Four Grace-Hopper superchips per node.
Not verified means a detail has not been confirmed in this catalog. Diagrams are schematic; node indices are illustrative, not physical hostnames. Data checked 13 September 2026.
Sources
Browse all cluster architecture pages →
CLUSTER