Features

• Workstation-grade Platform : Intel® Core™ i9/i7/i5/i3 Processor (14th gen, codename : RPL-S Refresh/RPL-S/ADL-S) running with Intel® R680E PCH supports max 65W TDP CPU
• Supports outstanding AI computing productivity by NVIDIA®/AMD graphics card, up to 200W power budget
• USB 3.2 Gen 2x2 offers a max speed of 20Gbps data transfer
• DC 12V to 50V wide range power input, Software Ignition Control
• 4 Independent 2.5GigE IEEE 802.3at PoE+ LAN, 2 GigE LAN
• Expansion : 2 PCIe x16 slot with x8 signal, 1 PCIe x8 slot with x4 signal , 1 M.2 Key B, 1 M.2 Key E, 1 Mini PCIe
• Supports multiple 5G/WiFi/BT/4G/LTE/GPRS/UMTS
• Supports Intel® vPro, TCC, Time-Sensitive Networking (TSN), and TPM 2.0
• Storage : 4 front-access 2.5" SSD Tray, M.2 Key M
   

 

產品亮點

High-speed AI Inference Platform

• Workstation-grade Platform : Intel® Core™ i9/i7/i5/i3 Processor (14th gen, codename : RPL-S Refresh/RPL-S/ADL-S), up to 65W TDP CPU supported
• Dual DDR5 5200MHz SO-DIMM, up to 96GB ECC/non-ECC Memory
• Supports performance hybrid architecture and advanced AI capabilities for smarter and faster applications   
• Optional delivers outstanding AI computing productivity supporting max 6656 CUDA Cores, 52RT Cores, 208 Tensor Cores, delivers 15.6 TFLOPS RT Core performance, and 63.9 TFLOPS Tensor Core performance
• High-speed data transfer : PCIe 4.0 (16GT/s), USB 3.2 Gen 2x2 Type-C (20G), USB 3.2 Gen 2 (10G), SATA III (6G), 2.5GigE LAN (2.5G), GigE LAN (1G)
• Optional VHub One-Stop AIoT Solution Service supports OpenVINO based AI accelerator and advanced Edge AI applications 



Compact Integration

• GigE Vision : 2 GigE LAN, 4 2.5GigE IEEE 802.3at PoE+ (25.5W/48V), optional with X-coded M12 Connector   
• USB Vision : 6 10G USB 3.2 Gen 2, 1 20G USB 3.2 Gen 2x2 Type-C   
• 4 COM RS-232/422/485, 3 SIM Socket, 32 Isolated DIO   
• Max 4 external 2.5” SSD/HDD trays, Software RAID supported 
• Compact M.2 Configurations : M.2 Key B, M.2 Key E, M.2 Key M 
• Intel Deep Learning Boost (Intel DL Boost) and Intel Distribution of OpenVINO toolkit for AI Computing
• Optional supports advanced NVIDIA Ampere Architecture and NVIDIA Turing Architecture by low-profile graphics cards

 

Flexible Configuration

• Levels of Independent Low-profile PEG : NVIDIA RTX A2000, NVIDIA Tesla A2, NVIDIA Tesla T4, NVIDIA T1000 
• Versatile product configuration for project requirements   
• SIM sockets for multiple 5G/WiFi/BT/4G/LTE/GPRS/UMTS wireless connections   
• PCIe expansions for multiple 10GigE M12 PoE+ LAN, 10GigE PoE+ LAN, 10GigE LAN, 10G SFP+, 2.5GigE PoE+ LAN, 2.5GigE LAN, M12 PoE+ LAN, PoE+ LAN, GigE LAN, LAN Switch, USB, or Independent Graphics   
• Compact Expansions : Mini PCIe, M.2 Key B, and M.2 Key


Optimized Operation

• Max 200W power budget for independent graphics cards   
• Multiple 5G/WiFi/BT/4G/LTE/GPRS/UMTS supports non-stop wireless data transfer
• Lockable SSD/HDD trays, RAID 0, 1, 5, 10 Data Protection     
• DC 12V to 50V wide range power input
• Software Ignition Power Control for in-vehicle applications
• Anti-shock, Anti-vibration
• EN50155, EN50121-3-2   
• Real-time capable with Intel Time Coordinated Computing (Intel TCC) and support for Time-Sensitive Networking (TSN)
• Smart System Management : Intel vPro, TPM 2.0, PXE, Wake on LAN, Watchdog Timer