Skip to content
HomeSolutionsDell TechnologiesPowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled)
Dell Technologies logo
Dell Technologies

PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled)

Air-cooled four-GPU 4U node for AI and HPC modeling

Full manufacturer warrantyAuthorized channel48-hour quote

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

PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled) — Dell Technologies enterprise hardware
Dell Technologies logo
PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled) hardware detail 1
PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled) hardware detail 2
PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled) hardware detail 3

Configuration at a Glance

GPU/Accelerator4x NVIDIA HGX H100 80GB 700W SXM5
GPU InterconnectNVLink, fully connected 4-GPU
CPU2x Intel Xeon Scalable, up to 64 cores each
MemoryUp to 32 DDR5 DIMM, up to 2TB

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

4x NVIDIA HGX H100 80GB 700W SXM5

Processor

2x Intel Xeon Scalable, up to 64 cores each

Memory

Up to 32 DDR5 DIMM, up to 2TB

Overview

The PowerEdge XE8640 is a 4U air-cooled server configured with four NVLink-connected NVIDIA HGX H100 SXM5 accelerators, designed for sites that need four-GPU AI and HPC performance without facility liquid cooling. Nexus Compute specifies, integrates, and tests each system, delivered with Dell ProSupport through authorized channels.

Who This Solution Is For

Teams needing four GPUs without liquid cooling
HPC modeling and simulation groups
Enterprises starting GPU consolidation
Departmental AI and research clusters

Business Benefits

No liquid infrastructure

Liquid-assisted air cooling lets the node deploy in standard data center rows without a facility water loop.

Balanced AI and HPC

Four NVLinked H100 GPUs handle both deep learning and double-precision simulation in one platform.

Deployment-ready

Systems arrive integrated, thermally validated, and firmware-baselined with Dell ProSupport coverage.

Typical Business Use Cases

1

HPC modeling and simulation

2

Mid-scale model training and fine-tuning

3

Mixed AI and scientific computing

4

Departmental GPU clusters

Industry Applications

HPCHigher Education & ResearchManufacturingHealthcare & Life Sciences

Technical Overview

The XE8640 fully interconnects four NVIDIA HGX H100 SXM5 accelerators over NVLink in a 4U chassis cooled by liquid-assisted air cooling, avoiding any facility liquid requirement. Dual current-generation Intel Xeon Scalable CPUs, DDR5 memory, and PCIe Gen5 expansion drive the GPUs, with iDRAC9 providing enterprise management.

GPU/Accelerator4x NVIDIA HGX H100 80GB 700W SXM5
GPU InterconnectNVLink, fully connected 4-GPU
CPU2x Intel Xeon Scalable, up to 64 cores each
MemoryUp to 32 DDR5 DIMM, up to 2TB
CoolingLiquid-assisted air cooling, hot-swap fans
Networking/FabricConnectX-7 NDR InfiniBand ready
Form Factor4U rack, air-cooled
ManagementiDRAC9 Enterprise
Warranty3-year Dell ProSupport

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

Can the XE8640 run in a standard air-cooled data center?

Yes, its liquid-assisted air cooling needs no facility water loop, so it deploys in conventional racks while still supporting four 700W H100 GPUs.

Is four GPUs enough for my training workload?

A four-GPU NVLink node fine-tunes and trains mid-sized models efficiently; for frontier-scale pretraining we would point you to the eight-GPU XE9680.

What deployment support is included?

Nexus Compute integrates and burn-in tests each unit and coordinates Dell ProSupport, so the server is production-ready on delivery.

Hardware Assistance

Configure the PowerEdge XE8640 (4x NVIDIA HGX H100, Air Cooled) 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.