Nvidia shrinks Jetson fro AI at the edge
Claimed to be the smallest, most powerful artificial intelligence (AI) supercomputer, the Jetson Xavier NX has been introduced by Nvidia. Althought smaller than a credit card, the Jetson Xavier NX is designed for robotic and embedded computing devices at the edge.
The Jetson Xavier NX module is built around a new low-power version of the Xavier SoC and delivers up to 21TOPS at 15W and 14TOPS at 10W. These parameters make it suitable for AI workloads suitable for small commercial robots, drones and IoT systems in industry, for example high-resolution sensors, optical inspection as well as network video recorders, portable medical devices and other IoT systems.
The Jetson Xavier NX is based on the Nvidia Volta graphics processor unit (GPU) with 384 Nvidia CUDA cores and 48 Tensor cores and two NVDLA. The CPU is a six-core Carmel Arm 64-bit CPU with 6Mbute L2 + 4Mbyte L3. Memory capacity is 8Gbyte 128-bit LPDDR4x, operating at 51.2Gbyte per second. There are up to sixe CSI cameras for a totla of 36 virtual channels and 12 lanes of MIPI CSI-2.
Its performance parameters enables running multiple neural networks in parallel and processing data from multiple high-resolution sensors simultaneously in a Nano form factor (70 x 45mm). The Jetson Xavier NX supercomputer runs on the same CUDA-X AI software architecture as all Jetson offerings and runs on Ubuntu-based Linux.
It is also supported by Nvidia JetPack software development kit, an AI software stack that can run modern and complex AI networks, accelerated libraries for deep learning as well as computer vision, computer graphics and multimedia.
The Jetson Xavier NX module is built around a new low-power version of the Xavier SoC used in these benchmarks.
Jetson Xavier NX is also pin-compatible with Jetson Nano, allowing shared hardware designs and those with Jetson Nano carrier boards and systems to upgrade to Jetson Xavier NX. It also supports all major AI frameworks, including TensorFlow, PyTorch, MxNet, Caffe and others, advises Nvidia.