Uvation MarketplaceMarketplace

HPE Compute XD690
AI Server

The HPE Compute XD690 AI Server combines dual Intel Xeon 6 processors with P-cores and eight NVIDIA B300 GPUs in a 10U air-cooled rack server. With DDR5 memory, NVMe storage support, and high-speed ConnectX-8 networking, it provides a platform for GPU-accelerated AI computing.

8× NVIDIA B300 GPUs
GPUs per rack
Dual Intel Xeon 6 Processors
Processors per rack
Up to 3 TB DDR5 Memory
Memory per rack

Compute, Memory and
Accelerated Processing

Built around NVIDIA B300 accelerators and Intel Xeon 6 processing, the XD690 combines GPU resources, expandable memory, and NVMe storage in a rack-mounted system.

NVIDIA B300 GPU Architecture

Integrates eight NVIDIA B300 SXM GPUs, each with 288 GB of memory.

High-Capacity DDR5 Memory

Supports 32 DDR5 DIMMs with memory capacity of up to 3 TB.

NVMe Storage Support

Accommodates up to eight E1.S EDSFF NVMe SSDs for primary storage.

Eight-GPU NVIDIA B300 Configuration

The XD690 integrates eight NVIDIA B300 SXM GPUs within a single 10U server platform, supported by dual Intel Xeon 6 processors and high-capacity DDR5 memory.

HPE Compute XD690 AI Server in a data hall
8× NVIDIA B300 SXM
GPUs
288 GB
Memory per GPU
Up to 3 TB DDR5
Memory
Up to 8× E1.S EDSFF
NVMe SSDs

Workloads

A GPU-accelerated server platform for AI computing, combining NVIDIA B300 accelerators, Intel Xeon 6 processing, high-capacity memory, and NVMe storage.

AI Model Development

Eight B300 GPUs linked by NVLink give transformer and diffusion models room to train in a single 10U server.

GPU-Accelerated Computing

288 GB of HBM3e per GPU and dual 64-core Xeon 6 CPUs support complex models that combine multiple data types.

AI Workload Execution

Eight 800 Gb/s ConnectX-8 SuperNICs and NVMe storage support high-throughput inference pipelines across multi-node clusters.

Engineered to dominate

Designed for Dense AI Infrastructure

A 10U platform combining GPU acceleration, server processing, high-speed networking, and integrated power hardware.

Exploded view of the HPE Compute XD690 AI Server
8× ConnectX-8
Network adapters
12× 3,200 W
Hot-plug PSUs
Silicon Root of Trust
and Secure Boot
Server 24.04 LTS
Canonical Ubuntu
Exploded view of the HPE Compute XD690 AI Server
8× ConnectX-8
Network adapters
12× 3,200 W
Hot-plug PSUs
Silicon Root of Trust
and Secure Boot
Server 24.04 LTS
Canonical Ubuntu

The rack, itemized.

SUBSYSTEM

SPECIFICATION / CONFIGURATION DETAILS

Processor

2x Intel® Xeon® 6 processors with P-cores

Operating System

Canonical Ubuntu 24.04 LTS

Chipset

N/A

Accelerators / GPU

8× NVIDIA B300 SXM; 288 GB per GPU

Memory

32 DDR5 DIMMs; up to 3 TB

Primary Storage

2 × M.2 NVMe

Secondary Storage

8 × E1.S NVMe

Security

Silicon Root of Trust; TPM; Secure Boot; signed firmware

Chassis

10U air-cooled chassis

Power Supply

12× 3,200 W hot-plug PSUs

Ports

1 × USB 3.0 Type-A; 1 × Micro-USB 2.0 debug; 1 × Mini DisplayPort; 1 × Reset port

Slots / PCIe Architecture

2× FHHL DPU slots or 2× HHHL NIC slots

Bezel

N/A

Form Factor

10U rack server

Network Card

8× ConnectX-8; management Ethernet

Warranty

3/3/3 warranty - three years parts, three years labor, and three years onsite support

Front view of the HPE Compute XD690 AI Server

Power & Fabric

Redundant Power

Twelve 3,200 W Titanium power supplies in an N+N configuration support 14.3 kW continuous and 16.4 kW peak draw.

East-West Fabric

Eight 800 Gb/s ConnectX-8 SuperNICs on the GPU board scale training and inference over InfiniBand XDR or Spectrum-X Ethernet.

North-South Networking

Two PCIe slots take BlueField-3 DPUs or 400 Gb/s adapters to carry north-south data center traffic.

Nuclear-Ready, Plug and Play AI Factories

Bypass grid constraints with self-contained, high-density inference ready infrastructure directly adjacent to clean, base load nuclear power.

Global Deployment Engine

Any Infra, Anywhere in the world.

Global Shipping and Deployment Capability treats advanced infrastructure as mobile, modular freight.

Packaged complex computing nodes, cooling manifolds, and power distribution units into certified, standard form factors.

360° Service Ecosystem

Unifying Legacy with Next-Gen Compute.

Vertically integrated service ecosystem that serves as an operational bridge between legacy environments and next-generation, AI-native infrastructure.

Finance and Leasing

Capital-as-a-Service for High-Density Compute

Convert upfront capital expenditures (CapEx) into predictable operating expenses (OpEx).

Uvation allows enterprises and sovereign entities to deploy multi-megawatt AI factories and advanced GPU clusters without liquidity constraint.