Skip to content
HomeSolutionsGPU ServersSupermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX)
Nexus ComputeNew

Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX)

Liquid-cooled 8-GPU H100 density that cuts cooling power and noise.

Full manufacturer warrantyAuthorized channel48-hour quote

We help you choose, configure, and deliver the right system — no obligation.

Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX) — Nexus Compute enterprise hardware
Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX) hardware detail 1
Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX) hardware detail 2
Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX) hardware detail 3

Configuration at a Glance

GPU / Accelerator8x NVIDIA H100 80GB SXM5 (640GB HBM3 total)
GPU Interconnect4x NVSwitch, 900GB/s all-to-all NVLink
CoolingDirect-to-chip liquid cooling (GPUs + NVSwitch), 4 loops
CPUDual Intel Xeon Scalable (4th/5th Gen), up to 64 cores each

Tailored per engagement. Full technical overview below.

Configuration Options

Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.

GPU / Accelerator

8x NVIDIA H100 80GB SXM5 (640GB HBM3 total)

Processor

Dual Intel Xeon Scalable (4th/5th Gen), up to 64 cores each

Memory

Up to 8TB DDR5-5600 ECC RDIMM (32 DIMM)

Storage

Up to 16x 2.5" hot-swap NVMe + 2x M.2 boot

Overview

This direct-liquid-cooled build of the Supermicro 8U HGX H100 platform routes coolant over all eight SXM5 GPUs and the NVSwitches to sustain peak clocks while reducing data-center cooling power. Nexus Compute specifies the cooling loops and CDU integration, validates the system, and warranty-backs it through authorized Supermicro channels.

Who This Solution Is For

Operators packing high GPU density into power- or thermally-constrained racks
Facilities pursuing lower PUE and reduced cooling overhead
Sustained-load training shops needing stable peak clocks
Sites adding rack-level liquid cooling to existing rooms

Business Benefits

Lower cooling power

Direct-to-chip liquid cooling can cut data-center cooling energy by up to roughly 40% versus air at equivalent density.

Sustained peak clocks

Liquid blocks hold GPU and NVSwitch temperatures down so the node maintains performance under continuous load.

Integrated and validated

We coordinate manifold, CDU, and leak-checked loop integration so the node arrives ready to commission.

Typical Business Use Cases

1

Continuous high-utilization model training

2

High-density GPU racks in constrained facilities

3

Sustainability-driven AI infrastructure builds

4

Liquid-cooled cluster nodes

Industry Applications

AI & Machine LearningHPCHigher Education & ResearchTelecomSaaS & Software

Technical Overview

The platform mounts the NVIDIA HGX H100 8-GPU SXM5 baseboard with four dual-GPU cold plates plus a liquid-cooled NVSwitch block across four coolant loops. Dual Intel Xeon Scalable CPUs and up to 8TB DDR5 sit in the 8U chassis with PCIe Gen5 fabric slots and Titanium redundant power.

GPU / Accelerator8x NVIDIA H100 80GB SXM5 (640GB HBM3 total)
GPU Interconnect4x NVSwitch, 900GB/s all-to-all NVLink
CoolingDirect-to-chip liquid cooling (GPUs + NVSwitch), 4 loops
CPUDual Intel Xeon Scalable (4th/5th Gen), up to 64 cores each
MemoryUp to 8TB DDR5-5600 ECC RDIMM (32 DIMM)
StorageUp to 16x 2.5" hot-swap NVMe + 2x M.2 boot
Networking / Fabric8x PCIe Gen5 x16 LP; 400G InfiniBand / 25GbE optional
Form Factor8U rackmount, liquid-cooled
Power8x 3000W (4+4) Titanium redundant PSUs

Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.

Warranty, Support & Fulfillment

Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.

Enterprise Warranty

Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.

Authorized Channel

Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.

Lead Time & Deployment

48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.

Nationwide Fulfillment

Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.

Frequently Asked Questions

Do we need a CDU or facility water?

These nodes connect to a rack or row coolant distribution unit; we help specify the CDU and confirm whether facility water or a closed loop fits your site.

Why choose liquid over the air-cooled version?

Liquid cooling lowers cooling energy and sustains peak GPU clocks under continuous load, which matters most for dense racks and long training runs.

Is servicing more complex?

Loops are quick-disconnect and leak-checked before shipment; we document maintenance procedures so your team can service nodes safely.

Hardware Assistance

Configure the Supermicro 8x H100 SXM5 Direct Liquid-Cooled (8U HGX) with Nexus Compute

Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.