Flopper.io Flopper.io
PE

Penguin Solutions Altus XE5318GTSv2

rack
8× H200
Penguin publishes no interconnect figure; expansion is 8x PCIe Gen5 x16 low profile and 4x PCIe Gen5 x16 FHHL
Power
—
kW Total
Memory
1128
GB Total

The Penguin Solutions Altus XE5318GTSv2 is a 5U GPU server carrying eight NVIDIA HGX H200 SXM5 modules of 141 GB each, hosted by AMD EPYC 9005/9004 Series processors with 8x PCIe Gen5 x16 low profile and 4x PCIe Gen5 x16 FHHL. Penguin's AMD-hosted Hopper workhorse, air cooled in 5U. 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.

Ask us where to buy this Spec sheet

We will point you at suppliers who have it. Free, and no signup.

Systems built on this design

11 configurations
NVIDIA HGX H200 Reference design baseboardASUS ESC N8-E11 rackDell PowerEdge XE9680 PowerEdge XE9680 rackGigabyte G593-SD1-LAX3 6NG593SD1DR000LBX3 rackHPE Cray XD670 rackHPE ProLiant Compute XD685 (HGX H200) rackPenguin Solutions Altus XE5318GTSv2 Viewing Altus XE5318GTSv2 rackPenguin Solutions Relion XE5318GTS Relion XE5318GTS rackQCT QuantaGrid D74F-7U D74F-7U rackQCT QuantaGrid D75F-7U D75F-7U rackSupermicro GPU SuperServer AS -8125GS-TNHR (HGX H200) AS -8125GS-TNHR rackSupermicro GPU SuperServer SYS-821GE-TNHR SYS-821GE-TNHR rack
Head to head: vs NVIDIA HGX H200vs ASUS ESC N8-E11vs Dell PowerEdge XE9680vs Gigabyte G593-SD1-LAX3vs HPE Cray XD670vs HPE ProLiant Compute XD685 (HGX H200)vs Penguin Solutions Relion XE5318GTSvs QCT QuantaGrid D74F-7Uvs QCT QuantaGrid D75F-7Uvs Supermicro GPU SuperServer AS -8125GS-TNHR (HGX H200)vs Supermicro GPU SuperServer SYS-821GE-TNHR

Built on the NVIDIA HGX H200 baseboard, so accelerator performance matches the reference. Chassis, CPU, memory and cooling are this vendor's own.

System Details

GPU Configuration

GPU Model: NVIDIA H200 SXM 141GB
GPU Count: 8 GPUs
Architecture: Hopper
Interconnect: Penguin publishes no interconnect figure; expansion is 8x PCIe Gen5 x16 low profile and 4x PCIe Gen5 x16 FHHL

System Specifications

Form Factor: rack
Total Power: —
Total Memory: 1128 GB
Memory Bandwidth: 38400 GB/s

Powered by NVIDIA H200

This system utilizes 8 × NVIDIA H200 SXM 141GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

700W

Per GPU Memory

141 GB

Process Node

4nm

Architecture

Hopper

View H200 Details →

Related Systems

HPE ProLiant Compute XD685 (HGX H200)

8× H200 SXM 141GB
2025

HPE Cray XD670

8× H200 SXM 141GB
2024

NVIDIA DGX H200

8× H200 SXM 141GB
2024

Gigabyte G593-SD1-LAX3

8× H200 SXM 141GB
2024

Supermicro GPU SuperServer AS -8125GS-TNHR (HGX H200)

8× H200 SXM 141GB
2024

Documentation & Resources

Official Datasheet

Penguin Solutions Altus XE5318GTSv2 technical specifications

Download PDF

Penguin Solutions GPU-Accelerated Servers

Penguin Solutions • Latest version

View Document ↗

Typical Use Cases

AI/ML Training
High-Performance Computing
Data Analytics

The Penguin Solutions Altus XE5318GTSv2 runs 8× H200 GPUs over Penguin publishes no interconnect figure; expansion is 8x PCIe Gen5 x16 low profile and 4x PCIe Gen5 x16 FHHL.