Supermicro AS-8125GS-TNHR 2x AMD 9534 8x H100 80GB 1.5TB DDR5 2x 3.8TB NVMe

AMD EPYC Enterprise AI Platform with HGX H100 Baseboard This NVIDIA-certified Supermicro AS-8125GS-TNHR A+ Server delivers enterprise-grade AI performance with 8x H100 80GB SXM5 GPUs pre-integrated on HGX baseboard and dual AMD EPYC 9534 processors. Featuring 128 cores total, 1.5TB DDR5-4800 ECC memory (24x 64GB), and optimized storage with 1x 960GB M.2 boot drive plus 2x 3.8TB NVMe drives. Comprehensive networking includes 8x Mellanox ConnectX-7 400G OSFP, dual-port ConnectX-6 Dx 100GbE, and dual-port Intel X550 10GbE adapters. Enhanced with TPM 2.0 security module and enterprise management suite. Perfect for large language model training, distributed AI workloads, and scientific computing requiring maximum CPU core density with integrated GPU acceleration.

Qty

Price

Enterasource Warranty
Total price 1
Price on Request

What to Expect from Enterasource

Why choose Enterasource?

At Enterasource, we pride ourselves on delivering top-notch service to our valued customers. Here are just a few of the benefits you can expect when choosing us for your IT needs

Complimentary ground shipping on orders over $50 throughout the contiguous United States (exclusions apply to Hawaii and Alaska).

Efficient same-day shipping for in-stock orders placed before 3PM PST.

Rapid one business day quote response time.

Industry-leading next business day parts
replacement warranties to ensure your equipment stays up and running.

Our commitment to exceptional customer service and timely delivery make us the ideal choice for all your IT hardware needs. Choose Enterasource for a hassle-free experience and the highest quality product on the market.

Supermicro AS-8125GS-TNHR 2x AMD 9534 8x H100 80GB 1.5TB DDR5 2x 3.8TB NVMe
Manufacturer:
SuperMicro
Condition:
Refurbished
Availability:
Available by Quote
Too many options? We are happy to help you configure.
Ask a question
More Information
Description

AMD EPYC Enterprise AI Computing Platform

NVIDIA Certified A+ Server with 8x H100 80GB on HGX Baseboard and Dual AMD EPYC 9534 Processors

HGX H100 GPU Computing Platform

8x NVIDIA H100 80GB SXM5 GPUs integrated on single HGX baseboard for optimal performance

AMD EPYC 9534 Genoa Processing

2x AMD EPYC 9534 processors with 128 cores total and 256 threads for maximum parallel computing

Enterprise Memory Configuration

1.5TB DDR5-4800 (24x 64GB) ECC registered memory for large-scale AI workloads

Multi-Tier Network Architecture

8x 400G ConnectX-7, dual-port 100G ConnectX-6, and dual-port 10G Intel adapters

NVIDIA HGX H100 Baseboard Architecture

This configuration features 8x H100 80GB SXM5 GPUs pre-integrated on a single NVIDIA HGX baseboard, providing the ultimate in GPU-to-GPU communication performance and thermal efficiency for enterprise AI workloads.

GPU Architecture 8x H100 SXM5 80GB
HGX Integration Single baseboard unit
Total GPU Memory 640GB HBM3
GPU Interconnect NVLink 4.0 + NVSwitch

AMD EPYC + HGX H100 Performance Excellence

AI Workload Category AMD EPYC Advantage Enterprise Applications
Large Language Model Training 128 cores excel at data preprocessing and tokenization GPT models, enterprise chatbots, content generation platforms
Distributed AI Training Superior memory bandwidth and PCIe connectivity Multi-node training clusters, federated learning systems
High-Throughput Inference Massive parallel processing for batch operations Real-time recommendation engines, content moderation
Scientific Computing High core count ideal for simulation workflows Climate modeling, molecular dynamics, financial modeling

AMD EPYC 9534 System Architecture

AMD EPYC 9534 Genoa Platform

  • Dual Socket SP5 with AMD chipset integration
  • AMD EPYC 9534 processors (64 cores each, 128 total)
  • Base frequency 2.45GHz with 280W TDP per CPU
  • 5nm Zen 4 architecture with enhanced AI acceleration
  • 128 PCIe 5.0 lanes per CPU for maximum expansion

Enterprise DDR5 Memory System

  • 1.5TB (24x 64GB) DDR5-4800 2Rx4 ECC registered DIMMs
  • 24 DIMM slots with 12-channel memory per CPU
  • Up to 6TB maximum memory capacity support
  • Enterprise ECC protection with advanced RAS features
  • Optimized for memory-intensive AI preprocessing tasks

High-Performance Storage Configuration

M.2 Boot Drive:

1x 960GB PCIe Gen4 M.2 NVMe SSD for high-speed operating system and application storage

Primary Data Storage:

2x 3.8TB PCIe Gen4 2.5" NVMe SSDs for datasets, model checkpoints, and working storage

Total Storage Capacity:

8.56TB total NVMe storage with enterprise-grade performance and Gen4 speeds

Enterprise Multi-Tier Networking Architecture

Mellanox ConnectX-7 400G Fabric

8x ConnectX-7 400G OSFP adapters with InfiniBand/Ethernet dual-mode capability

PCIe Gen5 x16 interfaces providing 3.2TB/s aggregate bandwidth for distributed AI training

ConnectX-6 Dx 100GbE Management

1x dual-port ConnectX-6 Dx QSFP56 providing 200Gbps total management bandwidth

Dedicated high-speed management network isolated from training fabric for optimal performance

Intel X550 10GbE Administrative Network

1x dual-port Intel X550 10GbE RJ45 adapter delivering standard Ethernet connectivity for administrative access, IPMI management, and integration with existing network infrastructure. Provides reliable out-of-band management capabilities.

Enterprise Infrastructure & Security Platform

Redundant Power Architecture

  • 6x 3000W (3+3) Titanium Level 96% efficiency PSUs
  • 18kW total power capacity with full redundancy
  • Hot-swappable power supplies for zero-downtime maintenance
  • Advanced power monitoring and load balancing

Hardware Security Features

  • TPM 2.0 Module (SLB9670) hardware root of trust
  • AMD Platform Security Processor (PSP) integration
  • Secure Boot with cryptographic firmware verification
  • Silicon Root of Trust NIST 800-193 compliance

Supermicro Management Suite

  • IPMI 2.0 with full remote management capabilities
  • Supermicro Server Manager (SSM) enterprise platform
  • SuperDoctor 5 comprehensive health monitoring
  • SuperServer Automation Assistant (SAA) for deployment

Advanced Thermal Management

  • 10 heavy-duty fans with intelligent speed control
  • Optimized airflow for HGX H100 baseboard cooling
  • Real-time temperature monitoring with alerting
  • Data center-optimized for 24/7 continuous operation

Why Choose This AMD EPYC Configuration?

Maximum Core Density

Dual EPYC 9534 processors deliver 128 cores and 256 threads, providing exceptional parallel processing power for data-intensive AI preprocessing and multi-threaded workloads.

HGX Baseboard Integration

Pre-integrated HGX H100 baseboard eliminates integration complexity while providing optimal GPU-to-GPU communication and thermal management for maximum AI performance.

Comprehensive Network Fabric

Three-tier networking with 400G ConnectX-7, 100G ConnectX-6, and 10G Intel adapters provides complete connectivity for any data center environment.

Enterprise-Ready Platform

NVIDIA certified with TPM 2.0 security, redundant power, and comprehensive management features for mission-critical AI deployments in enterprise environments.

Part TypeServer
ManufacturerSuperMicro
MPNAS-8125GS-TNHR
ConditionRefurbished
Form FactorRackmount
Rack Units8U
2.5" Drive Bays18
CPU Slots2
Dimm Slots24
Memory Riser SupportNo
Drive Types Supported NVMe, SATA
GPU SupportYes
PCI-E Slots8
PCI-E Version5.0
Power Supply TypeRedundant Titanium
Power Supply Slots6
PSU Wattage3000W
System ManagementIPMI, SSM, SUM
TPM SupportYes
Optical Drive SupportNo
Write Your Own Review
You're reviewing:Supermicro AS-8125GS-TNHR 2x AMD 9534 8x H100 80GB 1.5TB DDR5 2x 3.8TB NVMe
Your Rating
Customer Questions
No Questions
Please, mind that only logged in users can submit questions
Product added to your quote
© 2025 Enterasource, LLC. All Rights Reserved.