The CSB1-N10R3576 is a high-performance 1U server designed for large-model privatization deployment and AI computing tasks. With 60T computing power, it supports demanding workloads like deep learning, big data analysis, and high-performance computing (HPC). Equipped with BMC remote management and powered by Ruixin Micro processors, it ensures efficient, scalable, and secure operations for enterprises and data centers. Ideal for AI training, cloud computing, and large-scale model deployment.
- 60T computing power
- 1U rackmount design (space-efficient)
- BMC remote management (enhanced control)
- Ruixin Micro processors (high efficiency)
- Ideal for AI & large-model deployment
Perfect for data centers, cloud providers, and AI research needing robust, scalable computing.
CSB1-N10R3576 can be configured with up to 10 computing nodes. Each computing node is equipped with Rockchip's high-computing chip RK3576, with a maximum main frequency of 2.2 GHz and a total computing power of up to 60 TOPS, providing a strong computing power guarantee for artificial intelligence and deep learning.
Category | Specification |
---|---|
Server form | 1U rack-mounted computing power server |
Framework | ARM architecture |
Number of nodes | 10 distributed computing nodes (up to 80 ARM cores) + 1 control node |
Compute nodes | Octa-core 64-bit processor RK3576, main frequency up to 2.2GHz |
Control nodes | Octa-core 64-bit processor RK3588, main frequency up to 2.4GHz, highest computing power 6TOPS |
AI computing power | 60TOPS (INT8) |
RAM | 8GB LPDDR4 × 10 (Number of compute nodes) (Optional: 4GB/8GB/16GB) |
Storage | 64GB eMMC × 10 (Number of compute nodes) (Optional: 16GB/32GB/64GB/128GB/256GB) |
Storage Expansion | 3.5-inch/2.5-inch SATA3.0/SSD hard drive slot × 1 (BMC can directly operate the hard drive, and computing child nodes can indirectly access the hard drive through network sharing provided by BMC) |
Power | 550W AC power supply (Input: 90V AC264V AC, 47 Hz63 Hz, 8A) (Hot swappable not supported) |
Fan module | 6 high-speed cooling fans |
Size | 420.0mm(L) × 421.3mm(W) × 44.4mm(H) |
Installation req. | IEC 297 Universal Cabinet: 19 inches wide, 800 mm deep and above; retractable slideway installation; cabinet front/rear hole distance 543.5–848.5mm |
Full weight | Server net weight: 8.1kg, total with packaging: 10.3kg |
Environment | Operating temperature: 0°C 45°C; Storage: -40°C 60°C; Operating humidity: 5%~90% RH (non-condensing) |
BMC | Integrated with web-based management (supports Redfish, VNC, NTP, virtual media); can be redeveloped |
Large model | Supports privatization deployment of ultra-large-scale parametric models under Transformer architecture, including Gemma-2B, ChatGLM3-6B, Qwen-1.8B, Phi-3-3.8B |
Deep learning | Supports CNN, RNN, LSTM architectures, import/export of RKNN models; supports TensorFlow, TensorFlow Lite, PyTorch, Caffe, ONNX, Darknet; supports custom operator development |
Internet | 2 × 10G Ethernet (SFP+), 2 × Gigabit Ethernet (RJ45), 1 × Gigabit Ethernet (RJ45, MGNT for BMC management network) |
Console | 1 × Console (RJ45, BMC debug serial port, baud rate 115200) |
Display | 1 × VGA (max resolution 1080p, BMC management display) |
USB | 2 × USB 3.0 (lower USB is USB3.0 OTG, BMC can be upgraded OTG via USB flash drive) |
Button | 1 × Reset, 1 × UID, 1 × Power button |
Other interfaces | 1 × RS232 (DB9, baud rate 115200), 1 × RS485 (DB9, baud rate 115200) |