Edge AI / Data Analytics

Edge AI accelerator equipped with NVIDIA® Jetson

There has been an increase in expectations for AI technology due to the significant leaps and bounds that have been made in recent years in deep learning technology, and research and development toward the goal of practical application in a variety of fields has been proceeding. In particular, in the industrial sector, the introduction of IoT and 5G communication technologies has allowed for real-time collection of detailed field data, with AI allowing rapid response to changes in the situation, innovating brand new business models and truly ushering in an age of digital transformation (DX).

Application development using NVIDIA’s Jetson developers’ kit has been extremely popular for AI inference execution environments being used out in the field, and by implementing applications on industrial computers that emphasize practicality, it will become possible to meet a wide range of market needs.

AI and deep learning performance differences

Using deep learning to create and execute inference models allows for parallel execution of product-sum operations, meaning the performance is closely related to the product-sum operation capability of the GPU and the number of GPU cores.

Operational precision and product-sum operating performance(INT8 / FP16 / FP32)

The product-sum operation precision in a core includes INT8 for 8-bit integer operations, and FP16 and FP32 for 16-bit and 32-bit half-precision floating-point operations, respectively. If INT8 core precision is indicated as “1 TOPS,” 1 trillion 8-bit integer operations can be performed every second.

Because computational accuracy is important for creating inference models (learning processes), the half-precision floating-point operation (FP16/FP32) performance is emphasized. On the other hand, speed is more important than operation precision for implementing inference models (execution processes), so the integer operation (INT8) performance is emphasized.

However, the results will vary depending on the deep learning structure, the data, and the system components besides the GPU. As such, this value alone is insufficient for determining deep learning performance.

Edge AI accelerator equipped with NVIDIA® Jetson

Ideal for fields requiring on-site AI inference processing, including image inspection and predictive maintenance

DX-U1100 

Industrial Edge AI Computer DX-U1000 - Fanless Embedded PC / with NVIDIA® Jetson Nano™ / ARM Cortex-A57 / DC Power Supply

  • Edge AI accelerator equipped with NVIDIA® Jetson Nano™

  • Long-term supply and maintenance

  • Highly reliable and long-life components

  • Two Gigabit LAN ports for alternating connections between different network layers

  • Flexible installation

  • Supports a wide range of temperatures with operations guaranteed from -20 to 60°C

DXU-1100

DX-U1200

Industrial Edge AI Computer DX-U1000 - Fanless Embedded PC / with NVIDIA® Jetson Xavier NX™ / Carmel ARM v8.2 / DC Power SupplyEdge AI accelerator equipped with NVIDIA® Jetson Nano™

  • Edge AI accelerator equipped with NVIDIA® Jetson Xavier NX™ 

  • Long-term supply and maintenance

  • Highly reliable and long-life components

  • Two Gigabit LAN ports for alternating connections between different network layers

  • Flexible installation

  • Stable operations of this product are possible over a wide range of temperatures from -20 to 60°C .

DXU-1200

FA Computers VPC Series

These industrial computers have a focus on maintainability and reliability that allows us to recommend them not only for factory automation solutions but also for applications that require high availability.

VPC-7000-G Industrial Computer

VPC-7000-G

Industrial Computer VPC-7000 - Desktop / Rackmount PC / Intel 3rd Gen. Scalable Xeon (Ice Lake-SP + C621A) / 3x PCIe 4.0(x16), 2x PCIe 4.0(x8), 1xPCIe 3.0(x4) / RAID

This product is a high-end FA computer with availability and maintainability that assumes 24-hour continuous operation. Equipped with a many-core CPU, large-capacity ECC memory, and a hot-swappable RAID disk array, it is applicable to fields that require stable multiplex processing such as transportation, social infrastructure, medical equipment, security, and amusement parks.

  • High-end FA computer compatible with Intel ® Xeon ® Scalable Processor

  • Supports GPU computing

  • Uses highly reliable and long-life parts on the premise of continuous operation for 24 hours

  • Achieves long-term supply and long-term maintenance

VPC-7000-G

VPC-5000-G Industrial Computer

VPC-5000-G

Industrial Computer VPC-5000-G - Desktop PC / Rackmount PC / Intel 9th Gen. Xeon (Coffee Lake R + C246) / 1x PCIe(x16), 3x PCIe(x4), 3x PCI / RAID

This product is a BTO industrial computer that is equipped with Intel C246 chipset for server workstations. It can be equipped with an Intel Xeon E 2200-series or 9th generation Intel Core processor, up to 64 GB of ECC memory, up to 4 hot-swap hardware RAID configurations and more.

  • High-performance FA computer with Xeon processor compatible

  • Highly reliable and long-life components based on 24-hour operation

  • Support Windows 10 IoT Enterprise LTSC and Windows Server IoT

  • Long-term product availability and maintenance

VPC-5000-G

Product Enquiries.

Contact us today to find out more about how Contec solutions can support your business transformation.