MIC-720AI-00A1 AI Inference System based on NVIDIA® Jetson™ Tegra X2 256 CUDA Cores

MIC-720AI-00A1 AI Inference System based on NVIDIA® Jetson™ Tegra X2 256 CUDA Cores

MIC-720AI is the ARM based system which integrated NVIDIA® Jetson™ Tegra X2 System-on-Module processor, providing 256 CUDA® cores on the NVIDIA® Pascal™architecture. It is designed for the edge applications support rich I/O with low power consumption. The system has 4GB LPDDR4 memory, 4K video decode/encode, 2 USB3.0,POE and mSATA expansion. With wall mount brackets and fanless design which can help you easily install in rugged environment. This small embedded system is an ideal system for AI inference edge and deep learning applications.

• Fanless and compact design
• Supports NVIDIA Jetson® Tegra X2
• 2 USB 3.0 ports and 1 internal USB 2.0
• 1 IEEE 802.3af POE port
• Linux OS with BSP (board support package)
• Supports deep learning trained models
• Wide operating temperature

Processor System

• CPU
• Memory
• Flash

• nVIDIA TX2 Dual Core nVidia Denver2.0 + Quad Core ARM Cortex A57

• 8GB LPDDR4

• 32G of eMMC

Ethernet

• Interface
• Controller
• Speed

• RJ-45

• Intel® i210IT

• 10/100/1000 Mbps 1G

I/O

• Digital I/O
• USB
• Button

• 8 bit

• 2 x USB 3.0 & 1x USB 2.0(internal)

• 1 x Power on/off switch

Expansion

• Storage

• 1 x mSATA slot

Camera Interface PoE

• Controller
• Compliant
• Power Output

• MICROSEMI_PD69101ILQ-TR

• IEEE 802.3af

• 15.4W

Power

• Supply Voltage
• Power Type
• Power Consumption

• 19-24V DC-in

• DC-Jack

• Typical: 6.3W (Max-Q Mode)&Maximum: 48.3W (Max-N Mode)

Environment

• Operating Temperature
• Operating Humidity
• Vibration

• -20 ~ 60 °C with 0.7 m/s air flow (Max-Q mode)

• 95% @ 40 °C, non-condensing

• With SSD: 2 Grms @ 5 ~ 500 Hz, random, 1 hr/axis, With 2.5" HDD: 0.5 Grms @ 5 ~ 500 Hz, random, 1 hr/axis

Mechanical

• Dimension
• Weight
• Install

• 147 x 118 x 52 mm (5.7" x 4.6" x 2.4")

• 1.2 kg

• Desktop/Wall-mounting

OS

• Linux

• Ubuntu 16.04

Certification

• Safety Certifications

• CE/FCC