The site uses Cookies. When you continue to browse the site, you are agreeing our use of Cookies. Read our privacy policy>

Global/English
Home Products meta brain® Artificial Intelligence Servers AI Servers NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6
NF5468M6

NF5468M6

NF5468M6 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. NF5468M6 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.

Key Features
  • Excellent Performance
    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
    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
    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
    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

Technical Specifications

Model

NF5468M6-P

NF5468M6-T

NF5468M6-V

Height

4U

CPU

2* 3rd Gen Intel® Xeon® IceLake Scalable processors, TDP 270W

GPU Accelerator Card

Supports 8* FHFL DW PCIE GPU

Rear supports up to 4 PCIe4.0 x16 slots 

Supports 4* FHFL DW PCIE GPU

Rear supports 2 PCIe 3.0 x8 and 2 PCIe4.0 x8 slots

Supports 16* FHFL SW GPU

Rear supports up to PCIe4.0 x16 slots

 

Chipset

Intel® C621A series chipset ((LBG-R)

Memory

32* DDR4 3200MHz LRDIMM/RDIMM

Internal PCIe

Support up to 2* internal standard Raid card

Front I/O

2* USB 3.0, 1* VGA, 1* RJ45 serial port

Rear I/O

1* serial port, 2* USB 3.0, 1* RJ45 management port, 1个OCP3.0(support NCSI)

Storage

24* 2.5” or 12* 3.5” SAS/SATA drives(up to 8* NVME SSD), 2* M.2 SATA SSD

16* 2.5” or 12* 3.5” SAS/SATA drives(up to 2* NVME SSD), 2* M.2 SATA SSD

24* 2.5” or 12* 3.5” SAS/SATA drives(8* NVME SSD), 2* M.2 SATA SSD

RAID

Optional supports RAID 0, 1, 10, 5, 50, 6, 60, etc., supports Cache super capacitor protection, provides RAID state migration, RAID configuration memory

OS

Microsoft Windows Sever、Red Hat Enterprise Linux、Ubuntu Linux、CentOS, etc. mainstream OS

Cooling

N+1 Redundant system cooling fan

PSU

4* 1600W/2000W/2200W/3000W 80Plus platinum PSU, supports 2+2 redundancy

Dimensions (W*H*D)

483mm * 175.5mm * 830mm

Temperature

5 - 35°C / 41°F - 95°F

Full Load Weight

≤85kg


Relevant Information
  • Meta Brain Server NF5468M6 White Paper

Please enter the SN or MN code of your product to preview or download it.

Verification code:

By clicking Submit,you acknowledge that you agree to comply with all applicable laws and regulations.

Click here to learn how to inquire a SN code.