Supermicro GPU Products

A global leader in high performance, high efficiency server technology and innovation, Supermicro has been developing and supporting end-to-end green computing solutions, cloud computing, enterprise IT, big data, HPC, and embedded markets since 1993.

Advantages of Supermicro Server

Powerful computing power

Compared to traditional CPU servers, advanced GPU accelerators are capable of simultaneously processing large amounts of data and complex computing tasks, enabling them to perform well in handling complex tasks such as artificial intelligence, deep learning, and high-performance computing.

Low power consumption and high energy efficiency

It can provide better performance with lower power consumption under the same computing task. The server can automatically adjust its power supply according to its load situation, further improving energy utilization efficiency.

Application scenarios

Whether it is neural network training and inference in deep learning, or scientific research and engineering simulation in high-performance computing, they can better adapt to different workloads and maximize the performance advantages of GPU servers.

image

Application scope

Supermicro GPU servers are mainly used in fields such as AI and Deep Learning, High-performance Computing, Data Analysis And Mining, Graphics Rendering And Visual Effects, Finance, Healthcare, Cloud Computing And Virtualization.

  • AI and Deep Learning
  • High performance computing
  • Practice IT Management
  • Financial sector
  • Data Analysis and Data Mining
  • Cloud computing and virtualization
FEATURED RACKMOUNT GPU SERVER

H200 GPU SuperServer SYS-821GE-TNHR

Key Features
1.5th/4th Gen Intel® Xeon® Scalable processor support
2.32 DIMM slots Up to 8TB: 32x 256 GB DRAM Memory Type: 5600MTs ECC DDR5
3.8 PCIe Gen 5.0 X16 LP
2 PCIe Gen 5.0 X16 FHHL Slots, 2 PCIe Gen 5.0 X16 FHHL Slots (optional)
4.Flexible networking options
5.2 M.2 NVMe for boot drive only
16x 2.5" Hot-swap NVMe drive bays (12x by default, 4x optional)
3x 2.5" Hot-swap SATA drive bays
Optional: 8x 2.5" Hot-swap SATA drive bays
6.10 heavy duty fans with optimal fan speed control
7.Optional: 8x 3000W (4+4) Redundant Power Supplies, Titanium Level
6x 3000W (4+2) Redundant Power Supplies, Titanium Level

image

Supermicro Products

image

NVIDIA HGX B200 8-GPU

  • Key Features
    Dual 5th/4th Gen Intel® Xeon® Scalable processors
    8x NVIDIA B200 SXM GPUs, 1.4TB HBM3e GPU Memory Space
    32x DIMM slots, DDR5 up to 5600MT/s(1DPC), 4800MT/s(2DPC)
    8x PCIe 5.0 x16 LP, 2x PCIe 5.0 x16 FHHL slots
    10x Front Hot-swap 2.5" NVMe U.2 , 2x NVMe M.2
    2x 10GbE RJ-45, 1x Dedicated BMC/IPMI
    6x 5250W Redundant (3+3) Power Supplies Titanium Level Efficiency
image

NVIDIA GB200 NVL72 SuperCluster

  • Supermicro’s GB200 NVL72 solution represents a breakthrough in AI computing infrastructure, combining Supermicro’s end-to-end liquid-cooling technology. It enables up to 25x performance improvement at the same power envelope compared to previous NVIDIA Hopper generations, while reducing data center electricity costs by up to 40%. The system integrates 72 NVIDIA Blackwell GPUs and 36 NVIDIA Grace CPUs in a single rack, delivering exascale computing capabilities through NVIDIA’s most extensive NVLink™ network to date, achieving 130 TB/s of GPU communications.
image

NVIDIA DGX B300

  • GPUs:NVIDIA Blackwell Ultra GPUs
    CPU:Dual Intel® Xeon® Processors
    Total GPU Memory:2.3TB
    Performance:72 PFLOPS FP8 training/144 PFLOPS FP4 inference
    Networking:8x OSFP ports serving 8x single-port NVIDIA ConnectX-8 VPI (Up to 800Gb/s NVIDIA InfiniBand/Ethernet)
    2x dual-port QSFP112 NVIDIA BlueField-3 DPU (Up to 800Gb/s NVIDIA InfiniBand/Ethernet)
    Management Network:1GbE onboard NIC with RJ45
    1GbE RJ45 Host baseboard management controller (BMC)
    Storage: OS: 2x 1.9TB NVMe M.2
    Internal storage: 8x 3.84TB NVMe E1.S
    Power Consumption:~14kW
image

NVIDIA DGX GB300

  • GPU: 72x NVIDIA Blackwell Ultra GPUs in Grace Blackwell Ultra Superchips
    CPU Cores: 2592
    GPU Memory HBM3e: 20.1TB
    Total Fast Memory: 37.9TB
    Performance: 1400 petaFLOPS of FP4 AI performance
                            700 petaFLOPS of FP8 AI performance
                            350 petaFLOPS of FP16 AI performance
    Networking: 72x OSFP single-port NVIDIA ConnectX®-8 VPI with 800Gb/s NVIDIA InfiniBand
                            18x dual-port NVIDIA BlueField®-3 VPI with 200Gb/s NVIDIA InfiniBand and Ethernet
    NVIDIA NVLink Switch System: 9x L1 NVIDIA NVLink Switches
    Management Network: Host baseboard management controller (BMC) with RJ45

Why Supermicro?

Supermicro is dedicated to providing the quality, reliability, and environments that their consumer base expects. The team at Supermicro complies to meticulous design implementation, manufacturing standards, and ISO standards to safeguard their product line to industry standards.

A profound relationship with the industry's top semiconductor manufacturers ensures customers early access to cutting-edge technology and enables Supermicro to provide their customers a broad portfolio of systems and solutions that meet the demands of the future of the data-driven business world.

Supermicro stands behind their product line with a thorough warranty and replacement policy. Customers have access to Supermicro's global partners and technical support teams 24/7 that guarantees to resolve your issues.

LET’STALK

Are you ready?

Discuss about your next project.

Leave your contact information and our cloud experts will respond promptly to meet your specific needs.