NF5468M7 is the latest flagship product in the flexible NF5468 AI server series, IEIT’s bestselling server. It supports the industry's most comprehensive Al accelerator card and domestic AI chips, and can be widely used with Internet AI public clouds, enterprise-level AI cloud platforms, smart security, video codecs, etc. NF5468M7 provides large local storage and is the industry's only single machine that can support one-click switching of Balance/Common/Cascade topology software to flexibly respond to the performance tuning requirements of different AI application scenarios.
Technical Specifications
| Item |
Specification |
| Product Model |
NF5468-M7-A0-R0-00 |
| Form Factor |
4U |
| CPU |
Supports 2 x 4th/5th Gen Intel® Xeon® Scalable Processors, TDP up to 350W |
| GPU Cards |
Supports 8 x full-height, full-length, dual-width PCIe GPU cards; |
| Supports ≥4 PCIe 5.0 x16 slots |
| Chipset |
Intel® Emmitsburg PCH C740 |
| Memory |
Supports 32 x DDR5 4800MHz RDIMMs |
| Internal PCIe Cards |
Supports up to 2 PCIe 5.0 expansion cards |
| Front I/O |
2 x USB ports, 1 x VGA port |
| Rear I/O |
1 x Serial port, 1 x USB 3.0 port, 1 x RJ45 management port, |
| 1 x OCP3.0 network adapter (supports NCSI) |
| Local Storage |
Front: 24 x 2.5" or 12 x 3.5" SAS/SATA HDDs; |
| Max supports 16 x NVMe or E3.S drives; |
| Internal: 2 x M.2 NVMe/SATA SSD |
| RAID Support |
Optional RAID 0/1/10/5/50/6/60; |
| Supports cache supercapacitor protection, RAID migration and configuration preservation |
| Operating Systems |
Microsoft Windows Server, Red Hat Enterprise Linux, Ubuntu Linux, CentOS and other mainstream OS |
| System Cooling |
N+1 redundant cooling fans |
| Power Supply |
4 x 1600W/2000W/2200W/3000W 80Plus Platinum/Titanium PSUs, supports N+N redundancy |
| Chassis |
W447mm × H174.5mm × D850mm |
| Operating Temperature |
5 – 35°C / 41°F – 95°F |
| Fully Configured Weight |
≤ 87kg |
Key Features
Excellent Performance
2x Intel® 3rd Gen Xeon® Scalable processors, 3 UPI interconnects, TDP 270W
The PCIe Gen4 platform with doubled bandwidth can support up to 8x 350W GPU cards
4-card pass-through design, no need for PCIe Switch communication between CPU and GPU, greatly reducing P2P communication delays and helping algorithms achieve maximum performance
Flexible Configuration
Three GPU configurations to flexibly match different application scenarios
Supports one-click switching of Balance / Common / Cascade topology software to flexibly respond to the performance tuning requirements of different AI application scenarios.
Meets the needs of customers with high redundancy and high security scenarios, using rich I / O extension interfaces
Rich Ecology
Abundant global AI accelerator card partners, NVIDIA, AMD, Intel, etc.
Leading deep learning framework, TensorFlow, PyTorch, PaddlePaddle, MXNet, etc.
Powerful IEIT AIStation resource platform, algorithm platforms, and application optimization services
Mature meta-brain ecosystem for all-industry and all-scenario support
High Reliability
N+N power supply redundancy can carry 8x 350W GPUs running at full speed
N+1 fans redundancy ensures stable operation at high ambient temperatures
Smart BMC remote management provides zero-contact O&M, and improves O&M efficiency
Your message must be between 20-3,000 characters!
English
