Model SRS-GB200-NVL72
Liquid-cooled 72-GPU NVLink rack with Supermicro DLC and rack-level integration.
The Supermicro NVIDIA GB200 NVL72 SuperCluster is a liquid-cooled rack-scale system with 72 Blackwell GPUs and 36 Grace CPUs in one NVLink domain, delivered as an integrated rack with Supermicro direct liquid cooling.
Supermicro supplies the full solution, including networking, CDUs and onsite installation, and validates it at rack level before it ships.
Via 36 GB200 Grace Blackwell Superchips.
Up to 40% reduction in data center electricity cost.
1.8 TB/s GPU-to-GPU bandwidth across all 72 GPUs.
Up to 400 Gb/s Quantum-2 InfiniBand or Spectrum-X Ethernet compute fabric.
PCIe 5.0 local storage across the compute trays.
Consultation, parts, networking and onsite installation.
| Compute nodes | 18× 1U ARS-121GL-NBO |
|---|---|
| Enclosure | 48U, 19-inch |
| Dimensions (W × D × H) | 600 × 1068 × 2236 mm |
| GPUs | 72× NVIDIA B200 |
|---|---|
| CPUs | 36× NVIDIA Grace |
| GPU memory | Up to 13.4 TB HBM3e |
| System memory | Up to 17 TB LPDDR5X |
| GPU interconnect | 9× NVLink Switches, 4 ports per compute tray, 1.8 TB/s GPU-to-GPU |
| Compute fabric | Up to 400 Gb/s Quantum-2 InfiniBand or Spectrum-X Ethernet |
|---|---|
| In-band management | Up to 200 Gb/s with BlueField-3 DPUs |
| Out-of-band management | 1G / 10G Ethernet |
| Storage | 144 E1.S PCIe 5.0 drive bays |
| Power shelves | 8× 1U 33 kW (6× 5.5 kW PSUs each), 132 kW total |
|---|---|
| Cooling | 1× in-rack 250 kW CDU with redundant PSUs and dual hot-swap pumps |
| Alternative CDUs | In-row 1.3 MW; liquid-to-air 180 kW / 240 kW |
End-to-end liquid-cooled rack with 72 Blackwell Ultra GPUs — an exascale supercomputer in a single rack.
Next-generation rack-scale platform for agentic AI and reasoning, unifying 72 Rubin GPUs and 36 Vera CPUs in a single NVLink domain.
Turnkey, ready-to-deploy enterprise AI infrastructure built on the Vera Rubin platform and managed with NVIDIA Mission Control.
Tell us your workload and timeline for the Supermicro NVIDIA GB200 NVL72 SuperCluster. We reply within one business day with a configuration, pricing and lead time.