4U GPU Supermicro Rack Server SYS-420GP-TNAR With Dual Intel Xeon Scalable Processors
Price:
Negotiable
MOQ:
1pc
Delivery Time:
5-7 work days
Brand:
Supermicro
Product Description
Supermicro SYS-420GP-TNAR – 4U Dual Intel Xeon GPU Server for AI Training & HPC
The Supermicro SYS-420GP-TNAR is a high-density 4U rack-mount server purpose-built for AI training, deep learning, high-performance computing (HPC), and large-scale data analytics. It delivers exceptional performance, scalability, and reliability for compute-intensive workloads, making it an ideal choice for modern data centers and AI research labs.
Key Features
- Dual 3rd Gen Intel® Xeon® Scalable Processors (Socket P+/LGA-4189)
- Supports up to 40 cores/80 threads per CPU, 60MB cache, and 270W TDP (air-cooled)
- Intel® C621A chipset for enterprise-grade stability
- Massive Memory Capacity
- 32 DIMM slots (16 per CPU)
- Up to 8TB ECC DDR4 (3200/2933/2666 MT/s, RDIMM/LRDIMM)
- Supports Intel® Optane™ Persistent Memory 200 series (up to 8TB PMEM, 12TB total with DRAM)
- Unmatched GPU Acceleration
- Up to 8× NVIDIA HGX A100 SXM4 GPUs (80GB) onboard
- 10× PCIe Gen 4.0 x16 low-profile slots for flexible GPU/accelerator expansion
- Optimized for NVIDIA NVLink® high-speed GPU interconnect
- Flexible Storage Options
- 6× 2.5” hot-swap NVMe/SATA/SAS drive bays (front-accessible)
- 2× M.2 NVMe/SATA slots (boot drives only, onboard)
- RAID 0/1/5/10 support for data redundancy
- Enterprise Networking & Management
- AIOM/OCP 3.0 support for modular, high-speed networking
- 2× 10GbE RJ45 ports (Intel® X550-AT2, optional)
- Integrated IPMI 2.0 + KVM/IP for remote management
- Supermicro SuperDoctor® 5, SuperCloud Composer® for system monitoring
- Efficient Cooling & Power
- 4× heavy-duty, variable-speed fans (optimal thermal control, high airflow)
- 4× 2200W redundant Titanium-level (96%+) power supplies (2+2 redundancy)
- Optional 3000W PSUs for higher GPU power demands (SYS-420GP-TNAR+ variant)
Specifications Summary
| Component | Details |
|---|---|
| Form Factor | 4U Rack-mount |
| CPU | Dual Socket P+ (LGA-4189), 3rd Gen Intel Xeon Scalable |
| Chipset | Intel C621A |
| Memory | 32 DIMMs, up to 8TB DDR4 + 8TB Optane PMEM |
| GPU | 8× HGX A100 SXM4; 10× PCIe 4.0 x16 slots |
| Storage | 6× 2.5” NVMe/SATA/SAS; 2× M.2 |
| Power | 4× 2200W Titanium Redundant (2+2) |
| Cooling | 4× High-Performance Variable-Speed Fans |
| Dimensions (W×H×D) | 482.6 × 177.8 × 736.6 mm (19” × 7” × 29”) |
Ideal Applications
- AI & Deep Learning Training: Large language models (LLMs), computer vision, neural network training
- High-Performance Computing (HPC): Scientific computing, finite element analysis, computational fluid dynamics
- Big Data Analytics: Distributed computing frameworks (Spark, Hadoop), real-time data processing
- Virtualization & Cloud: High-density virtual machines, containerization (Kubernetes), cloud infrastructure
Similar Products
Get in Touch
Have questions about our products or want to discuss a custom order? Our team is ready to help you.
Company
Beijing Hongkun Jiasheng Technology Co., Ltd.
Location
Jin Yan Long Building, Changping District, Beijing, 9th floor
Contact Person
Jeffrey