NVIDIA HGX H200 vs Penguin Solutions Relion XE5318GTS
Accelerator configuration, aggregate memory, bandwidth and power, side by side.
Specifications
| Specification | NVIDIA HGX H200 | Penguin Solutions Relion XE5318GTS |
|---|---|---|
| Vendor | NVIDIA | Penguin Solutions |
| Part number | — | Relion XE5318GTS |
| Form factor | baseboard | rack |
| Released | 2024 | — |
| Accelerator | 8x H200 | 8x H200 |
| Aggregate memory | 1.1 TB | — |
| Aggregate bandwidth | 38.4 TB/s | — |
| System power | 8.0 kW | — |
| Interconnect | NVSwitch + NVLink | Penguin publishes no interconnect figure; expansion is 12x PCIe Gen5 x16 low profile and 1x PCIe Gen4 x16 low profile |
Rows shown in grey are identical between the two.
NVIDIA HGX H200
HGX platform for cloud providers with H200 GPUs
Vendor product pagePenguin Solutions Relion XE5318GTS
The Penguin Solutions Relion XE5318GTS is a 5U GPU server carrying eight NVIDIA HGX H200 SXM5 modules of 141 GB each, hosted by 5th and 4th Gen Intel Xeon Scalable processors with 12x PCIe Gen5 x16 low profile and 1x PCIe Gen4 x16 low profile. The Intel-hosted Hopper machine, and the most expansion-heavy server in the range with thirteen slots. Penguin publishes only rack height, host processor, expansion slots and the accelerator for its servers, so no power draw, total memory, interconnect bandwidth or throughput figure is shown here; its full datasheets are hosted off its own domain and were not used as a source.
Vendor product page