8 H100 GPU 2 EPYC AI SYSTEM
- Powered by 8 NVIDIA H100, A100, L40, A40, RTX6000, RTXA5000
- AMD EPYC 9004 processors with Dual 128 Zen 4c cores
About
XINMATRIX® 8 GPU Deep Learning System is a scalable and configurable 4U Rackmount System with up to 8 Double Width NVIDIA GPU. Dual AMD EPYC 9004 Processor. The amount of GPU, CPU, System Memory and Storage are configurable according to requirements.
SYMMATRIX provides full range of AI GPU Systems from 1 to 10 GPU custom to your requirements.
Specification
Specification
GPU
8 NVIDIA H100, A100, L40, A40, RTX6000, RTXA5000
Processor / System Bus
2 x Socket (LGA 6096)
AMD EPYC™ 9004 series(up to 400W)
4th Generation AMD EPYC Processors with AMD 3D V-Cache technology
Memory
Total Slots:
32 DIMM SLOTS
8TB
Expansion Slots
8 x PCIe slots (Gen5 x16 link, FH,FL) for dual-slot GPU cards
Drive Bays
NVME, SATA, SAS
Networking
2 x 10GbE LAN ports
Optional up to 400Gb Ethernet/Infiniband
1 x Management Port
Dimensions
4U Rackmount
Depth 786 x W 438 x H 177 mm
Power Supply
3+1 Redundant 1600W 80 PLUS Platinum Power Supply
Environment
Operation temperature: 10℃ ~ 35℃
Non operation temperature: -40℃ ~ 70℃
Non operation humidity: 20% ~ 90% ( Non condensing)
You May Also Like
Related products
-
8 GPU 2 EPYC DEEP LEARNING AI SERVER
SKU: SMX-GS4845More Information- GPU : 8 NVIDIA A100, V100, RTXA6000, RTX8000, A40
- NVLINK : 4 NVLINK
- CPU: 128 CORES (2 AMD EPYC ROME)
- PCIe Gen 4.0 support
- System Memory: 4 TB (32 DIMM)
- Type A: 12 3.5" SATA/NVMe U.2 Hotswap bays
- Type B: 24 2.5" SATA/SAS NVMe U.2 Hotswap bays
-
4 GPU 1 XEON DEEP LEARNING AI SERVER
SKU: SMXB5631G88More Information- GPU: 4 NVIDIA A100, V100, RTXA6000, A40, RTX8000, T4
- CPU: 28 CORES (1 Intel Xeon Scalable)
- System Memory: 1.5 TB (12 DIMM)
- STORAGE: 2 2.5" SSD, 2 NVMe M.2 SSD
- 1U Rackmount
-
10 GPU 2 XEON DEEP LEARNING AI SERVER
SKU: SMX-R4051More Information- GPU: 10 NVIDIA H100, A100, L40, A40, RTX6000, 2-slot GPU
- CPU: 2 4th Generation Intel Xeon Scalable Processors
- System Memory: 8 TB (32 DIMM)
- STORAGE: NVME
Our Partners
Previous
Next