4 GPU 1 XEON DEEP LEARNING AI SERVER
- 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
About
XINMATRIX® GPU Deep Learning System is a scalable and configurable. The amount of GPU, CPU, System Memory and Storage are configurable according to requirements.
Deep learning is one of the fastest growing segments in the machine learning/artificial intelligence field. It uses algorithms to model high-level abstractions of data in order to gain meaningful insight for practical application. Such data manipulation has application in various fields, such as computer vision, speech recognition and language processing, and audio recognition.
Specification
Specification
GPU
Options:
4 A100, V100, RTXA6000, A40, RTX8000, T4
4 A100, V100, RTXA6000, A40, RTX8000, T4
CPU
28 CORES
(One 28-Core Intel Xeon Scalable)
System Memory
12 DIMM slots
1.5TB (12 x 128GB DDR4-2933 ECC LRDIMM)
Storage
2X Hot-swappable 2.5" drive bays
SATA, SSD/HDD
2x NVMe M.2
Network
2X 10GbE ports + 1X IPMI
Dual-Port InfiniBand, 100Gb/s Ethernet or other high speed PCIe card (Optional)
System Weight
~ 15 KG
System Dimension
1U Rackmount
885mm(D) x 1U(H) x 19”(W)
Maximum Power Requirements
1,600Watts (200-240Vac input)
Redundant (1+1) PFC 80-PLUS Platinum
Operating Temperature Range
10° C ~ 35° C (50° F ~ 95° F)
Support & Warranty
Three Years on-site parts and services, NBD 8x5
You May Also Like
Related products
-

8 GPU H200, RTX Pro 6000, Xeon
SKU: 4UXGM-GNR2More Information- 4U NVIDIA MGX™ with 8x H200 or RTX Pro 6000 GPU
- Dual Socket E2 (LGA 4710), supports Intel® Xeon® 6700P-series, 6500P-series, and 6700E-series processors
- 16+16 DIMM slots (2DPC); supports DDR5 RDIMM, MRDIMM
- 16 Hot-swap E1.S (PCIe5.0 x4) drive bays
- 3+1, 80-PLUS Titanium, 3200W CRPS
- 2 RJ45 (1GbE) by Intel® i350
- Remote Management (IPMI)
-

2 NVIDIA GH200 GRACE HOPPER
SKU: GPU ARS-221GL-NHIRMore Information- High density 2U 1-node GPU system with two Integrated NVIDIA GH200
- NVIDIA® NVLink® high speed interconnect between both GH200 Super Chips
- Up to 1248GB of coherent memory (system as a whole), including 960GB LPDDR5X and 288GB of HBM3e for LLM applications
- 4x PCIe 5.0 X16 supporting one NVIDIA Bluefield 3 and two connectX-7 cards
-

10 GPU, H200, RTX Pro 6000, EPYC
SKU: 4U10G-TURIN2/RFMore Information- 4U Rackmount with 10X H200 NVL, RTX PRO 6000
- Dual Socket SP5 (LGA 6096), supports AMD EPYC™ 9005/9004 (with AMD 3D V-Cache™ Technology) and 97x4 series processors
- 12+12 DIMM slots (1DPC), supports DDR5 RDIMM, RDIMM-3DS
- 16 hot-swap 2.5" NVMe (PCIe5.0 x4) drive bays or 24 Hot-swap 2.5" SATA/SAS* drive bays *Additional RAID/HBA card required
Our Customers





























Previous
Next
