HPE
Dell
EqualLogic
- All EqualLogic PS Series
- EqualLogic PS6610 Series
- EqualLogic PS6510 Series
- EqualLogic PS6500 Series
- EqualLogic PS6210 Series
- EqualLogic PS6110 Series
- EqualLogic PS6100 Series
- EqualLogic PS6010 Series
- EqualLogic PS6000 Series
- EqualLogic PS5000/PS5500 Series
- EqualLogic PS4210 Series
- EqualLogic PS-M4110 Series
- EqualLogic PS4110 Series
- EqualLogic PS4100 Series
- EqualLogic PS4000 Series
- EqualLogic PS3000 Series
Specs
-PLEASE CONTACT FOR CUSTOM CONFIGURATIONS. DISK SHELVES, COMPLETE CONFIGURATIONS & TURN KEY SYSTEMS AVAILABLE.
-SPARE PARTS AVAILABLE FOR EACH MODEL. PLEASE ASK FOR PRICING ON SPARE DISK, CONTROLLERS, POWER SUPPLIES, NICs, HBAs, SFPs, Cables, etc.
Quick Specs
What's New for Apollo 6500 Gen10 Plus
- Apollo d6500 Gen10 Plus Chassis | P19674-B21
- (2) Node Options | XL645d & XL675d
- (2) AMD 2nd and 3rd Gen EPYC™ Series Processors
- (32) 3200 MT/s DDR4 DIMMs
- (16) SFF Drives
- Up to (12) NVMe Supported
- Up to (16 PCIe) or (8 SXM4) GPUs Supported
- Up to (6) PSUs
- HPE NS204i Boot Supported
- Direct Liquid Cooling System fully integrated
- AMD Instinct™ MI00 with 2nd Gen Infinity Fabric™ Link
- NVIDIA H100 and AMD MI210 PCIe GPU support
- HPE iLO 5
Other Versions:
- HPE ProLiant Apollo 2000 G9 | G10
- HPE ProLiant Apollo 4200 G9 | G10 | G10+
- HPE ProLiant Apollo 4500 G9 | ---
- HPE ProLiant Apollo 6500 G9 | G10 | G10+
- HPE ProLiant Apollo pc40
- HPE ProLiant Apollo sx40
(Click to View)
Ideal for:
- Artificial Intelligence (AI)
- Big Data Analytics
- Data Analytics
- Deep Learning (DL)
- High-Performance Computing (HPC)
- Language Model
- Machine Learning (ML)
The HPE Apollo 6500 Gen10 Plus chassis is a 6U platform designed for extreme performance and scalability, purpose-built to support demanding workloads like HPC, AI, and deep learning. It houses up to two server nodes and supports flexible GPU configurations, enabling the integration of either PCIe or SXM GPU architectures. The chassis offers advanced cooling solutions, including Direct Liquid Cooling (DLC), which enhances power efficiency and ensures optimal thermal management even under intense computational loads.
The HPE ProLiant XL675d Gen10 Plus node is a dual-socket system optimized for maximum GPU density and performance. It supports up to 8 NVIDIA HGX A100 GPUs in the SXM configuration or up to 10 double-wide PCIe GPUs, making it a powerhouse for deep learning, AI training, and complex HPC simulations. Powered by dual AMD EPYC processors with up to 64 cores each, it delivers exceptional processing power paired with support for up to 4 TB of DDR4 memory.
The HPE ProLiant XL645d Gen10 Plus node is a single-socket system designed for flexible and scalable GPU deployments. It supports up to 8 single-wide or 4 double-wide PCIe GPUs, making it well-suited for machine learning inference, data analytics, and virtualized environments. With a single AMD EPYC processor offering up to 64 cores and support for up to 2 TB of DDR4 memory, the XL645d provides a balanced mix of performance and efficiency.
Looking to Customize?
Need assistance with customization? Our sales experts are here to help you.
Start Customizing Today!
Spec Sheet 4
| HPE Apollo 6500 Gen10 Plus System | |||
|---|---|---|---|
| HPE ProLiant XL675d Server | HPE ProLiant XL645d Server Node | ||
| Chassis | HPE Apollo d6500 Gen10 Plus Configure-to-order Chassis (6U Chassis) | ||
| Density / Scale | Dual Processor Server per chassis | Up to 2 Single Processor Server Nodes per chassis | |
| GPU | Up to 10 Double Wide PCIe or 16 Single Wide PCIe GPU | Up to 4 Double Wide PCIe or 8 Single Wide PCIe GPU per Server Node | |
| Choice between: NVIDIA HGX™ A100 8-GPU, AMD Instinct™ MI100 with 2nd Gen Infinity Fabric™, and other leading accelerators | Choice between: NVIDIA HGX™ A100 4-GPU, AMD Instinct™ MI100 with 2nd Gen Infinity Fabric™, and other leading accelerators | ||
| Interconnect | Support for up to six high speed fabric interconnects; whether Ethernet, Infiniband, or HPE Cray Slingshot | Support for up to three high speed fabric interconnects; whether Ethernet, Infiniband, or HPE Cray Slingshot | |
| Processor | Dual AMD 2nd Gen EPYC™ Series Processor per server, up to 280W | Single AMD 2nd & 3rd Gen EPYC™ Series Processor per Server Node, up to 280W | |
| Memory | 32 3200MT/s DDR4 Smart Memory | 8 3200MT/s DDR4 Smart Memory per Server Node | |
| Storage | Up to 16 SFF drives – SAS/SATA/NVMe (M.2 optional) | Up to 8 SFF drives – Max 6 NVMe per Server Node (M.2 optional) | |
| System Management | HPE Integrated Lights Out (iLO 5), HPE Performance Cluster Manager (HPCM), HPE Container Platform, HPE OneView, Integrated Rack Consolidation Module (RCM) | ||
| System Security | iLO 5 Silicon Root of Trust, iLO Advanced (Optional) | ||
| OS Support | HPE Cray OS, Microsoft Windows Server, Red Hat, Ubuntu, VMware | ||
| Power | Fully redundant power for all configurations with up to 6 3000W Platinum Hot Plug Power Supplies per chassis. | ||
| Power Capping available at the server and chassis level; Rack and Row level Power Capping available with Apollo Platform Manager Kit. Shared power infrastructure at the chassis level. | |||
| Cooling | 15 - 80mm dual rotor hot pluggable chassis fans | ||
| Direct Liquid Cooling System fully integrated, installed, and supported by HPE | |||
| Storage Controller | Embedded SATA; optional HPE E208i-a SR, P408i-a SR, and P816i-a SR series Smart Arrays | Embedded SATA; optional HPE E208e-p SR and P408e-p SR series Smart Arrays | |