Supermicro Systems Featuring NVIDIA HGX H100/H200 GPUs
With upcoming support for Blackwell GPUs
Construct Your Tailored AI Solutions from the Largest NVIDIA MGX™ Systems Portfolio
Learn More
Full-Stack Data Center Acceleration for Generative AI, Graphics, and Video Applications
Learn More
New Super Quiet, NVIDIA AI Enterprise Software Pre-loaded System with Four A100 PCIe GPUs Supporting a Wide Range of AI & HPC Workloads
Learn More
- HGX Systems
- MGX Systems
- PCIe GPU Systems
- NVIDIA Networking
- Test Drive
- 3rd Gen OVX
- NVIDIA-Certified Systems™
- Resources
Next Leap of AI Infrastructure is Here
Today’s advanced AI models are changing our lives rapidly. Accelerated compute infrastructure is evolving at unprecedented speed in all market segments. Flexible, robust, and massively-scalable infrastructure with next-generation GPUs is enabling a new chapter of AI.
In close partnership with NVIDIA, Supermicro delivers one of the broadest selections of NVIDIA-Certified systems providing the most performance and efficiency from small enterprises to massive, unified AI training clusters with the new NVIDIA H100 and H200 Tensor Core GPUs.
Together, we achieve up to nine times the training performance of the previous generation for some of the most challenging AI models, cutting a week of training time into just 20 hours. Supermicro systems with the H100 PCIe, HGX H100 GPUs, as well as the newly announced HGX H200 GPUs, bring PCIe 5.0 connectivity, fourth-generation NVLink and NVLink Network for scale-out, and the new NVIDIA ConnectX®-7 and BlueField®-3 cards empowering GPUDirect RDMA and Storage with NVIDIA Magnum IO and NVIDIA AI Enterprise software.
Supermicro HGX Systems
8U NVIDIA HGX H100/H200 8-GPU System
Suited for Today’s Largest-Scale AI Training Models and HPC, Featuring Superior Thermal Capacity with Reduced Accoustics, More I/O, and Vast Storage
- GPU: NVIDIA HGX H100/H200 8-GPU with up to 141GB HBM3e memory per GPU
- GPU-GPU Interconnect: 900GB/s GPU-GPU NVLink interconnect with 4x NVSwitch – 7x better performance than PCIe
- CPU: Dual 4th/5th Gen Intel Xeon® or AMD EPYC™ 9004 series processors
- Memory: Up to 32 DIMM slots: 8TB DDR5-5600
- Storage: Up to 16x 2.5" Hot-swap NVMe drive bays
- NIC: High-performance 400 Gb/s NVIDIA BlueField®-3 or NVIDIA ConnectX®-7
- Networking: Scale out with NVIDIA Quantum-2 InfiniBand or NVIDIA Spectrum-X™ Ethernet
- Watch TECHTalk
- Featured Products: SYS-821GE-TNHR, AS -8125GS-TNHR
4U Liquid-Cooled NVIDIA HGX H100/H200 8-GPU System
Highest density and efficiency with D2C liquid cooling for both GPUs and CPUs to optimize performance and energy cost
- GPU: NVIDIA HGX H100/H200 8-GPU with up to 141GB HBM3e memory per GPU
- GPU-GPU Interconnect: 900GB/s GPU-GPU NVLink interconnect with 4x NVSwitch – 7x better performance than PCIe
- CPU: Dual 4th/5th Gen Intel Xeon® or AMD EPYC™ 9004 series processors
- Memory: Up to 32 DIMM slots: 8TB DDR5-5600
- Storage: Up to 16x 2.5" Hot-swap NVMe drive bays
- NIC: High-performance 400 Gb/s NVIDIA BlueField®-3 or NVIDIA ConnectX®-7
- Networking: Scale out with NVIDIA Quantum-2 InfiniBand or NVIDIA Spectrum-X™ Ethernet
- Direct-to-chip liquid cooling to support CPUs up to 350W and GPUs up to 700W
- Featured Products: SYS-421GE-TNHR2-LCC, AS -4125GS-TNHR2-LCC
Supermicro’s NVIDIA MGX™ Systems
NVIDIA MGX Systems with NVIDIA GH200 Grace Hopper Superchip
Integrated CPU, GPU, and Coherent Memory System for Cloud-Scale AI Inference
- CPU/GPU: Up to 2x 72-core Grace Arm Neoverse V2 CPU + H100 Tensor Core GPU in a single chip via NVIDIA GH200 Grace Hopper Superchip in 1U
- Memory (onboard): CPU: 480GB LPDDR5X GPU: up to 144GB HBM3e
- Storage: 8x Hot-swap E1.S drives and 2x M.2 NVMe drives
- NIC: High-performance 400 Gb/s NVIDIA BlueField®-3 or NVIDIA ConnectX®-7
- Networking: Scale out with NVIDIA Quantum-2 InfiniBand
- Learn More
NVIDIA MGX PCIe GPU Systems
Modular Building Block Platform Supporting Today’s and Future GPUs, CPUs, and DPUs
- GPU: Up to 4 NVIDIA double-width PCIe GPUs, including H100 PCIe, H100 NVL PCIe, L40S, and more
- CPU: NVIDIA Grace™ CPU Superchip, or 4th/5th Gen Intel® Xeon® Scalable processor
- Memory: Up to 480GB integrated LPDDR5X DRAM (with Grace CPU Superchip) or up to 2TB 4800MT/s ECC DDR5 DRAM (with Intel CPU)
- Storage: Up to 8 Hot-Swap E1.S NVMe
- NIC: High-performance 400 Gb/s NVIDIA BlueField®-3 or NVIDIA ConnectX®-7
- Networking: Scale out with NVIDIA Quantum-2 InfiniBand
- Learn More
Supermicro PCIe GPU Systems
Supermicro 4U or 5U PCIe 10-GPU System
Suited for Today’s Largest Scale AI Training Models and HPC, featuring superior thermal capacity with reduced accoustics, More I/O, and Vast Storage
- GPU: Up to 10 PCIe 5.0 GPUs including NVIDIA H100 PCIe, H100 NVL, or NVIDIA L40S
- CPU: Dual 4th/5th Gen Intel Xeon® or AMD EPYC™ 9004 series processors
- Memory: Up to 32 DIMM slots: 8TB DDR5-5600
- Cooling: Optimized thermal capacity to support CPUs up to 350W and GPUs up to 700W with air cooling
- NIC: High-performance 400 Gb/s NVIDIA BlueField®-3 or NVIDIA ConnectX®-7
- Networking: Scale out with NVIDIA Quantum-2 InfiniBand or NVIDIA Spectrum-X™ Ethernet
- Watch TECHTalk
- Learn More
Supermicro 4U 4-GPU Workstation
Optimized for 3D Metaverse Collaboration, Data Scientists, and Content Creators. Available in both Rackmount and Workstation Form Factors
- GPU: Up to 4x double width, full length GPUs, including NVIDIA L40S and NVIDIA RTX 6000 Ada Generation
- CPU: Dual 4th/5th Gen Intel Xeon® Scalable Processors
- Memory: 16 DIMM slots, up to 4TB DDR5-5600
- Expansion: 4 PCIe 5.0 x16 (double-width) slots, 3 PCIe 5.0 x16 (single-width) slots
- Storage: 8x 3.5" Hot-swap SATA/NVMe/SAS drive bays
- Featured Product: SYS-741GE-TNRT
Ethernet and InfiniBand for Scale-Out AISuperClusters
Supermicro: Your One-Stop Shop for AI Infrastructure, powered by NVIDIA Technology
Supermicro is a global leader in architecting and deploying comprehensive rack-scale solutions tailored for AI and HPC. Our solutions are designed to meet the most demanding customer requirements to ensure optimal performance, scalability, and efficiency. NVIDIA Networking Platforms serve as the nervous system for compute infrastructure. Both NVIDIA Spectrum-X and Quantum-2 InfiniBand are available today to meet your demands.
1:1 GPU to NIC 400Gb/s with Non-blocking
NVIDIA Quantum-2 InfiniBand
NVIDIA Quantum-2 (QM9700)
NVIDIA ConnectX®-7
NVIDIA Spectrum-X Ethernet
NVIDIA Spectrum-4 (SN5600)
NVIDIA BlueField®-3
Test Drive Supermicro Systems with NVIDIA GPUs
Get remote bare-metal access via Supermicro’s JumpStart program
Sign Up Now
NVIDIA OVX™ systems powered by Supermicro drive industrial revolution enabling digital twins
Learn More
NVIDIA-Certified Systems™ by Supermicro
With the continued rollout of advanced applications and workloads, customers require manageable, secure, and scalable servers for their data centers. Supermicro's compelling lineup of high-performance servers supporting NVIDIA GPUs and DPUs includes a growing number of NVIDIA-Certified Systems, with many more currently undergoing the certification process. Each server/GPU configuration earns its own certification.
Learn More
Featured Supermicro NVIDIA Certified and Qualified Systems
SYS-821GE-TNHR
NVIDIA-Certified with HGX H100 8-GPU |
NVIDIA ConnectX®-7 and BlueField®-3 support |
5th/4th Gen Intel® Xeon® Scalable processors |
32 DIMM slots, up to 8TB DDR5-5600 |
Learn more
AS -4125GS-TNRT
NVIDIA-Qualified with H100 PCIe GPUs |
NVIDIA ConnectX®-7 and BlueField®-3 support |
Dual-Socket, AMD EPYC™ 9004 Series Processors |
24 DIMM Slots, up to 6TB DDR5-4800 |
Learn more
SYS-421GE-TNHR2-LCC
NVIDIA-Certified with HGX H100 8-GPU |
NVIDIA ConnectX®-7 support |
5th/4th Gen Intel® Xeon® Scalable processors |
Direct-to-chip Liquid Cooling |
Learn more
AS -8125GS-TNHR
NVIDIA-Certified with HGX H100 8-GPU |
NVIDIA ConnectX®-7 and BlueField®-3 support |
Dual-Socket, AMD EPYC™ 9004 Series Processors |
24 DIMM Slots, up to 6TB DDR5-4800 |
Learn more
SYS-521GE-TNRT
NVIDIA-Qualified with H100 NVL, H100 PCIe, and L40S |
NVIDIA ConnectX®-7 support |
5th/4th Gen Intel® Xeon® Scalable processors |
32 DIMM slots, up to 8TB DDR5-5600 |
Learn more
ARS-221GL-NR
NVIDIA-Qualified with H100 NVL and L40S |
NVIDIA ConnectX®-7 support |
NVIDIA Grace™ CPU Superchip with 144 Cores |
Up to 480GB LPDDR5X onboard memory |
Learn more
Resources
Applied Digital Builds Massive AI Cloud With Supermicro GPU Servers
Applied Digital Offers Users the Latest In Scalable AI and HPC Infrastructure For AI Training and HPC Simulations with Supermicro High Performance Servers
Genesis Cloud Selects Supermicro's Powerful AI Servers For New Offerings
Genesis Cloud Offers AI and ML Services Using Supermicro Advanced AI Training System Featuring NVIDIA HGX H100 GPUs
Supermicro Shows Liquid-Cooled NVIDIA H100 Delta Next at ISC 2023
By Patrick Kennedy -
Supermicro showed off a large number of systems at ISC 2023. One of them particularly will be impactful. Supermicro had a liquid-cooled 8x NVIDIA H100 Delta-Next platform at the show and it was different than we expected.
See the product review
Accelerated Compute Infrastructure for AI and HPC Workloads with Supermicro Solutions Featuring the NVIDIA® H100 GPU
Supermicro Infrastructure for Demanding AI/ML and HPC Environments
Supermicro TECHTalk: AI Training Powered by Supermicro 8U 8GPU Server & New NVIDIA H100 GPUs
Requirements for AI training and machine learning is growing exponentially. Supermicro provides the accelerated compute infrastructure for increasing the performance of AI training by 9 times over current products available.
Play the video
Supermicro TECHTalk: Media and Graphics powered by Supermicro 4U 10 GPU server
The new Supermicro servers that incorporate the NVIDIA L40 GPUs create the optimum system for driving AI media and graphics applications.
Play the video
Supermicro TECHTalk: High-Density AI Training/Deep Learning Server
Our newest data center system packs the highest density of advanced NVIDIA Ampere GPUs with fast GPU-GPU interconnect and 3rd Gen Intel® Xeon® Scalable processors. In this TECHTalk, we will show how we enable unparalleled AI performance in a 4U rack height package.
Play the video
Supermicro TECHTalk: The Lowest Latency, Highest Performance Peer-to-Peer GPU System in 2U
In the first episode of our new Supermicro Tech Talks series, we take a look at our 2U system featuring the NVIDIA HGX A100 4 GPU board with four direct-attached NVIDIA A100 Tensor Core GPUs using PCI-E 4.0 for maximum performance and NVIDIA NVLink™ for high-speed GPU-to-GPU interconnects.
Play the video
Supermicro TECHTalk: Large-Scale Data Center GPU Solution
In this episode of Supermicro TECHTalk, we reveal one of our newest systems, our A+ 2U 2-Node GPU Server, powered by 3rd Gen AMD EPYC processors. This system delivers unrivaled energy efficiency and flexibility for a new performance standard of video streaming, cloud gaming, and social networking applications.networking applications.
Play the video
SuperMinute: 4U System with HGX A100 8-GPU
For the most demanding AI workloads, Supermicro builds the highest-performance, fastest-to-market servers based on NVIDIA A100™ Tensor Core GPUs. With the newest version of NVIDIA® NVLink™ and NVIDIA NVSwitch™ technologies, these servers can deliver up to 5 PetaFLOPS of AI performance in a single 4U system.
Play the video
SuperMinute: 2U System with HGX A100 4-GPU
The new AS -2124GQ-NART server features the power of NVIDIA A100 Tensor Core GPUs and the HGX A100 4-GPU baseboard. The system supports PCI-E Gen 4 for fast CPU-GPU connection and high-speed networking expansion cards.
Play the video
Supermicro SuperMinute: 2U 2-Node Server
Supermicro's breakthroughmulti-node GPU/CPU platform is unlike any existing product in the market. With our advancedBuilding Block Solutions® design and resource-saving architecture, this system leverages the most advanced CPU and GPU engines along with advanced high-density storage in a space-saving form factor, delivering unrivaled energy-efficiency and flexibility.
Play the video
Supermicro TECHTalk: 5 PetaFLOPS of AI Performance in a Single 4U Server
In this episode of Supermicro Tech Talks, we take a detailed look at our 4U System with HGX A100 8-GPU, ideal for large-scale deep learning training and neural network model applications.
Play the video
Delivering AI to Enterprises, with Supermicro Systems, NVIDIA AI Enterprise, and VMware
Supermicro Systems Supporting Scalable Virtualized AI Solutions
Supermicro, Cloudera, & NVIDIA Accelerate Enterprise Data Analytics
How Supermicro Ultra Servers and Cloudera Data Platform Extract Deep Insights from Massive Amounts of Data
Maximizing AI Development & Delivery with Virtualized NVIDIA A100 GPUs
Supermicro systems with NVIDIA HGX A100 offer a flexible set of solutions to support NVIDIA Virtual Compute Server (vCS) and NVIDIA A100 GPUs, enabling AI developments and delivery to run small and large AI models.
Unleash the Future of Innovation with HPC & AI
Transforming businesses with a new class of Supermicro and NVIDIA high-performance solutions
Deep Learning Benchmarks on Supermicro’s 4U 8 GPU System Based on Dual 3rd Gen AMD EPYC™ Processors
Demonstrating the performance of the Supermicro AS -4124GS-TNR, a 4U dual-processor 8 GPU server with up to 8TB of memory, and 160 Lanes of PCI-E 4.0, shows the generation over generation performance improvements of the new 3rd Gen AMD EPYC 7003 Series Processors on Deep Learning benchmarks
IDLab (Ghent University and the University of Antwerp) Implements Supermicro's Most Powerful Server, Cuts Down AI Research Experiment Time From Nearly Seven Hours Down To 40 Minutes
Supermicro’s new 10U GPU server using NVIDIA HGX-2 GPU baseboards helps leading Belgian research team improve calculation times by 10X which leads to faster AI development
AI Workloads at Scale
Kubernetes Cluster with Supermicro’s Systems with AMD EPYC™ 7002 processors
Supermicro® System Combines AMD EPYC™ Processors and NVIDIA GPUs to Achieve Consistent Deep Learning Performance with Linear Scaling
Learn about the benchmark results for Supermicro AS - 2023US-TR4, an AMD EPYC™ CPU based Supermicro high-end enterprise GPU server platform.