Mustang-V100-MX8

计算加速卡具有 8 x Movidius Myriad X MA2485 VPU, PCIe gen2 x4 接口, RoHS

  • Overview

    ● Half-Height, Half-Length, Single-slot compact size
    ● Low power consumption ,approximate 25W
    ● Supported OpenVINO™ toolkit, AI edge computing ready device
    ● Eight Intel® Movidius™ Myriad™ X VPU can execute multiple topologies simultaneously.
    Warning: DO NOT install the Mustang-V100-MX8 into the TANK AIoT Dev. Kit before shipment. It is recommended to ship them with their original boxes to prevent the Mustang-V100-MX8 from being damaged.
    In order to increase the compatibility with server grade systems, IEI recommend users to update Mustang-V100-MX8 firmware. If your product's serial number is before 2020 June (ex. Q205XXXXXX stands for 2020 May), please contact with IEI (IEI Technical Support Form), our team will help you to update it.
  • Features

    Mustang-V100-MX8

    加速走向未来

    An Intel® 视觉加速器设计产品
    Intel® 视觉加速器设计采用 Intel® Movidius™ VPU

    Mustang-V100-MX8 vision Accelerator product

    人工智能深度学习推理工作负载的理想选择

    mustang-v100-features

    开放式视觉推理和神经网络优化 (OpenVINO™) 套件

    - 半高, 半长, 单槽紧凑型尺寸.
    - 低功耗 ,每个 Intel® Movidius™ Myriad™ X VPU 约 2.5W
    - 支持开放式视觉推理和神经网络优化 (OpenVINO™) 套件, AI 边缘计算就绪设备.
    - 8个 Intel® Movidius™ Myriad™ X VPU 可同时执行多个拓扑.

    OpenVINO™ 套件

    OpenVINO™ 套件基于卷积神经网络 (CNN), 该套件在 Intel® 硬件上扩展了工作负载, 并最大限度地提高了性能.
    它可以将预先训练好的深度学习模型, 例如 Caffe, MXNET, Tensorflow, 优化为IR二进制文件, 然后跨 Intel® 硬件, 例如 CPU, GPU, Intel® Movidius™ Neural Compute Stick 和 FPGA, 执行推理机.

    openvino-toolkits-process

    应用

    machine vision

    机器视觉

    Smart Retail

    智能零售

    Surveillance

    监视

    Medical Diagnostics

    医学诊断

    IEI 深度学习推理加速卡 - Mustang-V100

    特征

    mustang-v100-accelerator-computer-dimension

    FAQ

    Issue Solution Files
    1. Fail "HDDL_NOT_INITIALIZED"
    Download patch to fix Mustsang can not be reseted in some motherboards issue
    For Linux OS, click here.
     
    For Windows OS, click here.
    2. If install  Mustang  accelerators  VPU quntity not equal to "8"

    Mustang-MPCIE-MX2 (2 VPU),
    Mustang-M2BM-MX2(2 VPU),
    Mustang-V100-MX4(4 VPU),
    two of Mustang-V100-MX8(16 VPU),
    Downloago to change autoboot.config device numbersd patch to fix Mustsang can not be recognized in some motherboards issue
    sudo gedit /opt/intel/openvino_/inference_engine/external/hddl/config/hddl_autoboot.config
    change "total_device_num'" from default "8" to the current quntity!
    ex.
    Mustang-MPCIE-MX2 (2 VPU), "total_device_num"      2
    Mustang-M2BM-MX2(2 VPU), "total_device_num"      2
    Mustang-V100-MX4(4 VPU),     "total_device_num"      4
    two of Mustang-V100-MX8(16 VPU),  "total_device_num"      16
    3. Would like to monitor the performance of each VPU Initial hddldaemon to observe utilization and temperature Document
  • Specifications

    • Hardware Spec.

      Spec Item Description
      System
      ChipsetEight Intel® Movidius™ Myriad™ X MA2485 VPU
      Cooling method / System FanActive fan
      Power
      Power ConsumptionApproximate 25W
      Environment
      Operating Temperature-20°C ~ 60°C
      Humidity5% ~ 90%
      Card Slot Interface
      PCIex4
    • Ordering Info.

      料号 描述
      Mustang-V100-MX8-R11 Computing Accelerator Card with 8 x Movidius Myriad X MA2485 VPU, PCIe gen2 x4 interface, RoHS
    • Package Contents

      • 1 X 全高支架

      • 1 x 外部电源线

      • 1 x QIG

Related Products