NVIDIA Embedded Platform
Nvidia already successfully built the gold standard for AI at the edge with release of its Jetson AGX Xavier/TX2i/TX2/TX1 modules. The newest Jetson AGX Xavier is the most leading Artificial Intelligence (AI) computing platform with high-performance, low-power consumption for deep learning, computer vision and compute-intensive applications in embedded fields.
AGX Xavier keeps the advantage of small form factor. In the same time, it increases more than 20X the performance and 10X the energy efficiency of its predecessor, Jetson TX2. Wide temperature -25°C to 80°C and ECC memory support is also the standard function. The Jetson AGX Xavier's unprecedented processing power, incredible energy efficiency, which is under 30W, and easily embeddable form factor(105mm x 105mm) are revolutionizing the AIoT application.
Aetina offers platforms that combine Jetson with our well-fatured reference carrier boards, or with carriers purpose-built for customers' individual needs. They are all supported by full SDK, including latest APIs, tool kits and libraries. We know different embedded applications have different OS requirements. So, Aetina provides software and OS integration service, the custom operating system built based on Linux kernel to sustain the individuality, reliability and security of the applications.
We also provide Extension Hardware Peripherals Integration Service, by preloading the relative API and driver of multiple Video-Image-Detection(VID) module, from our AIoT eco-system partners, and embedded peripheral I/O modules, from industrial-leading brand, Innodisk®, assisting developers flexibly to expand the systems, meanwhile help to start their deployment easier and faster.
Nvidia Jetson Module Comparision
| Jetson Nano | Jetson TX2 Series | Jetson AGX Xavier Series | ||||
| TX2 4GB | TX2 | TX2i | Jetson AGX Xavier 8GB | Jetson AGX Xavier | ||
| GPU | Nvidia Maxwell with 128 CUDA cores |
Nvidia Pascal with 256 CUDA cores |
Nvidia Volta with 384 CUDA cores & 48 Tensor cores |
Nvidia Volta with 512 CUDA cores & 64 Tensor cores |
||
| 0.5 TFLOPs (FP16) | 1.3 TFLOPS (FP16) | 5.5 TFLOPS (FP16) 11.1 TOPS (INT8) |
11 TFLOPS (FP16) 22 TOPS (INT8) |
|||
| CPU | Quad-core ARM Cortex A57 MPCore processor |
Dual-core Denver 2 64-bit CPU quad-core ARM A57 complex |
6-core ARM v8.2 64-bit CPU 8MB L2 + 4MB L3 |
8-core ARM v8.2 64-bit CPU 8MB L2 + 4MB L3 |
||
| DLA | - | - | 4.1 TFLOPS (FP16) 8.2 TOPS (INT8) |
5 TFLOPS (FP16) 10 TOPS (INT8) |
||
| Memory | 4GB 64-bit LPDDR4 | 4GB 128-bit LPDDR4 | 8GB 128-bit LPDDR4 | 8GB 256-bit LPDDR4x | 16GB 256-bit LPDDR4x | |
| Storage | 16GB eMMC 5.1 | 16GB eMMC 5.1 | 32GB eMMC 5.1 | 32GB eMMC 5.1 | ||
| Camera | 12 lanes MIPI CSI-II DPHY 1.1 (1.5 Gbps) |
12 lanes MIPI CSI-II D-PHY 1.2 (30 Gbps) |
16 lanes MIPI CSI-II D-PHY (40 Gbps) |
16 lanes MIPI CSI-II D-PHY (40 Gbps) |
||
| Power | 5W | 10W | 7.5W | 15W | 10W | 20W | 10W | 20W | 10W | 15W | 30W | |
| Connectivity | Wi-Fi (external chip) |
Wi-Fi (external chip) |
Wi-Fi (onboard) |
Wi-Fi (external chip) |
Wi-Fi (external chip) |
|
| 10/100/1000 BASE-T Ethernet |
10/100/1000 BASE-T Ethernet |
10/100/1000 RGMII Gigabit Ethernet |
||||
| Mechanical | 69.6 mm x 45 mm 260-pin edge connector |
87 mm x 50 mm 400-pin connector |
100 mm x 87 mm 699-pin connector |
|||
| Carrier Board |
Coming Soon | Click Here | Click Here | Click Here | Coming Soon | Click Here |
Product Features
-
Edge AI Stater Package, EASP
Aetina and AIoT ecosystem partners collaborate to meet the intelligent transformation demands of each vertical AI solution. Based on those solution’s scenarios, we launch different Edge AI Starter Package (EASP), including full-function Nvidia embedded Jetson platform and related peripheral hardware modules. In the same time, we integrate algorithms, software, and operating systems into the platform, allowing developers to develop and inspect with the Edge AI Starter Package.
-
Software and OS Integration
AI projects have a wide variety of applications, as the result, the software, algorithms and drivers are different. Therefore, Aetina provides software and hardware integration services to directly import BSP, software and drivers into the Nvidia embedded Jetson platform, saving time of installing the software and driver for customers, they can immediately start to develop after acquiring Aetina's products. At the same time, Aetina provides the patch files for customers to download updates, keeping the software and drivers up to date.
-
Hardware Peripheral Integration
Aetina's Nvidia embedded Jetson platforms provide high connectivity and can be built with various extension peripheral I/O module and video-image-detection (VID) modules to enhance the platform's applicability. In addition, we assist customers in selecting the appropriate hardware peripherals and importing the required software and drivers into the computing module to allow plug-and-play around the extension module. The Nvidia embedded Jetson platform support variety of I/O expansion modules (ex. HDMI, USB, CANBus and RS-232…etc) and up to 6 FHD or 3 4K camera modules with the transfer modules that been inspected and certified. With those benefits, developers can choose the right hardware peripherals for the demands of AI project development.
