HPE ProLiant Compute XD685 Air Cooling Server

ProLiant Compute XD685 Air Cooling Server

Reasons To Buy

Proven Expertise Worldwide to Support your Global Footprint and Growth Ambitions

The HPE ProLiant Compute XD685 is backed by a comprehensive portfolio of HPE Services to build, test, and deploy your solution with ongoing support, entirely available as stand-alone or consumption-based services to accelerate the deployment of your environment.
The robust and secure global supply chain from HPE and our proven ability to deploy massive custom AI clusters anywhere in the world with superior serviceability enable you to establish your presence globally with confidence.
Advisory and Professional Services, offering best-in-class methodologies and tools to enhance performance and increase return on investment (ROI), help you navigate the challenges of a large AI environment.
HPE Tech Care Service combines deep technical expertise with a leading-edge digital experience to provide personalized, prompt, and reliable product support so you can get the most out of your global IT investments.

Speed and Flexibility to Increase your Competitive Advantage

The HPE ProLiant Compute XD685 provides a choice of eight NVIDIA Blackwell Ultra (B300 HGX) GPUs, NVIDIA Blackwell (B200) GPUs, NVIDIA H200 Tensor Core GPUs, or AMD Instinct™ MI355X accelerators, delivering leading AI performance for training and tuning tasks.
A 5U direct liquid cooling (DLC) chassis helps you optimize pod deployment by leveraging a compact 8-node-per-rack arrangement.
HPE Performance Cluster Manager expedites initial cluster setup and software deployment and provides continued resilience and reliability.
HPE Factory Express services build, integrate, validate, fully test, and customize your solution in the factory, facilitating quicker on-site deployment.
You get flexibility and agility with a customized IT environment that lets you run the right workload at the right place, at the right time. Whether you want a solution on-premises or deploy a hybrid model, we offer flexible consumption and deployment models to suit your requirements and goals.

Sustainable and Safe AI Clusters to Increase Efficiencies and Protect Against Threats

The HPE ProLiant Compute XD685 offers a DLC option that helps you meet escalating power requirements and advance sustainability goals, leveraging HPE’s decades of expertise and market leadership in deploying the world’s largest liquid-cooled IT environments.
HPE iLO offers seamless server management with built-in security innovation to protect against escalating security threats.
Silicon root of trust from HPE, our fundamental approach to security, provides a Zero Trust architecture at the silicon level and helps enable completely server-essential firmware that is free from malware or compromised code.
HPE Financial Services allows you to turn aging infrastructure into capital in a secure, globally consistent, and environmentally responsible way, without disrupting your AI production.

Description

Do you need a highly performant, secure, and energy-efficient large AI cluster? The HPE ProLiant Compute XD685 leverages a purpose-built architecture and is designed to accelerate large, complex AI model training and tuning workloads. Powered by eight of the latest accelerators from NVIDIA or AMD, this server is optimized for large language model training, natural language processing, and multimodal training. Optional direct liquid cooling (DLC) helps organizations meet escalating power requirements, advance sustainability goals, and lower operational costs. Managed by HPE iLO, the server offers seamless management and robust security. HPE Services provides expertise for global installation and quick deployment of large AI clusters, offering stability and operational excellence.

What is new supporting point:
Support for eight NVIDIA Blackwell Ultra (B300 HGX) GPUs, eight NVIDIA Blackwell (B200) GPUs, eight NVIDIA H200 Tensor Core GPUs, or eight AMD Instinct™ MI355X accelerators.
Direct liquid cooling (DLC) available for all GPU options. Air-cooling available for NVIDIA H200 GPUs.
Modular 5U chassis for direct liquid cooling environments or 6U chassis for air cooled configurations.
24x DDR5 6400 RDIMMs and 12 channels per CPU DDR5 with ECC
Seamless management and built-in security features from HPE iLO 6
2x 5th Generation AMD EPYC™ Processors

Specifications

Memory
Internal memory type	DDR5-SDRAM
Storage
Supported RAID controllers	1 HPE MR416i-p Gen11 x16 Lanes 8GB Cache PCI SPDM Plug-in Storage Controller
Optical drive type	No
Ports & interfaces
Expansion slots	8x HHHL PCIe Gen5 x16, 3 x FHHL PCIe Gen5 x16
Optical drive
Optical drive type	No
Design
Optical drive type	No
Chassis type	Rack (5U)
Performance
Security features	HPE iLO 6
Management platform	Server: HPE iLO6, BMC, Redfish APIs, 1GbE LAN (for BMC only) Cluster: HPE Performance Cluster Manager
Power
Number of main power supplies	6
Power supply type	6x 3000W @54V Titanium Plus (N+1), 200-277 VAC input
Power supply	3000 W
Weight & dimensions
Dimensions (W x D x H) (imperial)	450.1 x 990.1 x 229.9 mm (17.7 x 39 x 9.05")
Weight	70 kg
Height	230 mm
Depth	990 mm
Management features
Management platform	Server: HPE iLO6, BMC, Redfish APIs, 1GbE LAN (for BMC only) Cluster: HPE Performance Cluster Manager
Technical details
Security features	HPE iLO 6
Number of processors supported	2
Chassis type	Rack (5U)
Weight (imperial)	70 kg (154.3 lbs)
Other features
Security features	HPE iLO 6
Number of processors supported	2
Weight (imperial)	70 kg (154.3 lbs)
Dimensions (W x D x H) (imperial)	450.1 x 990.1 x 229.9 mm (17.7 x 39 x 9.05")